710: That New Mouse Pad Feeling

What are you talking to us through right now? Oh, yeah. So if you've noticed that the tone of my voice is colder and not as warm as it used to be, that's the sound of an ARM CPU. Intel just had that extra...

That extra warmth? Extra soul that's just missing from these modern ARM CPUs. All right. Well, I'm excited for you. I'm on my new computer. If this recording goes wrong, you know, the last one, I just realized that the big Mac Pro that's still sitting next to me, I had for more than half of the life of the show for like seven years.

Is that right? Wow. I think so, right? Wow. I mean, that sounds right. Yeah. Yeah. Anyway, and it has served me well during those seven years, but my microphone is now plugged into a new computer. Oh. We're going to talk about this more later. This is the saddest sound of you getting a new computer I 've ever heard.

I know, right? It's just a tide me over computer. I know, but like it's also a huge upgrade. Like, are you able to like feel happy about this? Do you enjoy it? Does it feel fast? Does it feel good? Oh, yeah. No, I mean, I'll talk about it when we get to all our new products, but the whole household is happy because now I no longer need to hog my wife's computer to do dev work.

And so everybody, there is much happiness in the house about this. She bought it for you here, please. Yes. Well, I am very excited for you. When you had told us in Slack, I don't know, a day or two back that you were starting your migration, which we will talk about later in the show, Marco and I, our immediate reaction was it's the end of an era.

And again, as much as I poke fun at you, John, I know how much you love that computer. And it pains me to be serious and to be kind with you about this because I love making fun of you for it so much. But I know how much you loved the Tower Mac Pro. And I know how sad you are that you do not have a spiritual replacement or maybe at best, you do have a spiritual replacement, but certainly not a more literal replacement.

And so it did serve you well. It was very expensive, but you got a lot of use out of it. And I am sad. I am genuinely sad for your loss, although I'm excited for you to join. What is it? Five years ago, six years ago, whatever it was. So I'm happy for you, even if you sound slightly morose.

It is still September. As we record this, there's a bit of September left. And actually this month, believe it or not, or this year, September actually goes into October. And we're going to talk about that in a second. But September is Childhood Cancer Awareness Month.

And that means it is time for ATP to work in concert with Relay and with St. Jude Children's Research Hospital to try to raise money for St. Jude. St. Jude, I hope by now you're familiar. But like I said, they're a children's research hospital and they do incredible, incredible work to try to meet their mission of no child dying in the dawn of their life.

And generally speaking, that relates to childhood cancer, but it's not exclusively cancer. They do all sorts of absolutely banana stuff. And one of the many things that they offered for us to talk about, although we could say whatever we want, but one of the things they offered was that St. Jude has produced 18 of the world's most highly cited scientists in November 2025.

The 2025 list of highly cited researchers recognized 18 St. Jude scientists as among the world's most impactful, the most in St. Jude history. These researchers authored multiple scientific papers ranging in the top, excuse me, ranking in the top 1% by cit ations for their fields for roughly 10 years.

This is a small example of how I know we have a very international audience and I'm very glad for that. But it would be a reasonable conclusion to say, oh, well, St. Jude is in America and their health care system is barely a system, which yes and yes. But why does that help me, says the person not in America?

Well, it's for things like this. It's because their research, the research done at St. Jude really does have a worldwide impact. And we've talked in the past this month about how they have more concrete and explicit worldwide impacts with things like shipping medicine overseas and stuff like that. But really, St. Jude is a tremendous, tremendous place.

And to remind you, the reason we do this or one of the many reasons we do this is because our dear friend Stephen Hack ett, his eldest child was stricken with childhood cancer. It's something like six months old. That child is now 16, 17 years old or something like that. And that's all because of St. Jude Children's Research Hospital.

And I will tell you, before Marco can explain the Marco offset, I will tell you that in just a couple of weeks, actually a week and a half as I record this, on Friday, October 2nd at noon, one true time zone, I will be joining several other Relay hosts in doing our 12-hour telethon, which we call the podcast-a-thon.

It will be broadcast on Relay's YouTube channel. You can see me, if past years say anything about this, you can see me make a fool of myself for 12 straight hours. And who doesn't love that other than me? So check it out on Relay's YouTube channel. And if you're interested in donating, I will tell you this URL a few times.

Let me start now. Stjude.org slash ATP. S-T-J-U-D-E dot org slash ATP. Marco, we're going to talk in just one second about all the absolute consumerism that the three of us have participated in this year. I feel guilty about that. What can I do? Well, do I have the solution for you?

If you're wondering, how much should I give to St. Jude? Or how can I make myself feel better about all these purchases that I just made or I'm about to make? I have a solution for both of those simultaneously. So what you do is you calculate the Marco offset. This is your suggested minimum donation to St. Jude.

Whatever the base price is of the family of products that you just bought or are going to buy. So, for instance, an iPhone Pro. What's the base price? $10.99? $1.99? Whatever it is? Something like that. Yeah. So whatever your total after everything is for that transaction. Call it sales tax, AppleCare, storage upgrades, case, any kind of accessories, whatever it is.

Suppose your total after everything is $1,500. And the base price of that family device was $1,100. Your Marco offset is $400. That is the suggested minimum donation to St. Jude if you can swing it. The theory is you kind of saw that as like an add-on to this purchase that you maybe didn't necessarily need or maybe could have waited a little bit more.

That was just an add-on to that. So maybe you can spare another add-on to St. Jude. And if you do, you are hereby absolved by the powers vested in me by the Flying Spaghetti Monster. You are hereby absolved. I actually do have that. For all of these purchases.

No more guilt for you. You can go on freely knowing that you've done the right thing even though you bought something ridiculous. Now also, if you want to hit level two of the Marco offset, you take whatever money you spent on paid AI plans or API costs for this month.

And you add that on to the existing Marco offset to get level two. Level three is instead of doing that for this month, you do it for the whole year of your paid AI plan and API costs. So you can make yourself feel better about buying consumer devices that you might necessarily not need right this moment.

And for using AI for your benefit, all of that guilt can be absolved by simply donating to St. Jude in these amounts as your suggested minimum donations. Thank you very much. Feels like we're a little bit behind on fundraising this year. Maybe it's because the podcast-a-thon has been pushed out.

I know we get a lot of fundraising during the podcast-a- thon because people just demand to see all the various Relay hosts making fools of themselves. And that's a great time to donate. But, yeah, don't wait. If it goes into October because of the podcast-a-thon, that 's fine. But we really want to try to, like, beat the total from last year.

So please give if you can. Don't put it off. Yes. Excellent. Thank you, John. All right. Again, one more time. Stjude.org slash ATP. St-J-U-D-E dot org slash ATP. All right. We all have gotten some amount of treats. I think this year, John, you are the treat winner for the year.

I'm very impressed. You have gotten a trio of treats. Can you tell us about all of them, please? Sure. I'll start with my iPhone 18 Pro. I'll start by saying goodbye to my iPhone 16 Pro because I 'm on the every two-year phone plan along with my wife. We alternate years.

And so my 16 Pro was my first experiment in babying my battery because the battery life was so good on the 16 Pro. I'm like, you know what? I'm going to set an 80% limit and see how long I can last on that. And it turns out I lasted maybe, like, 10 and a half months before I said, you know, at 80% battery towards the end of the first year.

Some days I feel like it's a little low. So I cranked it back up to 90. And then eventually I just went to 100% because I'm like, oh, this thing's on its way out the door. I'm going to be replacing it soon. Let me just put it on 100. Anyway, just an update on that. The battery health at the end of two years, I guess, more than 10 months.

I said one year in 10 months. Sorry. At the end of two years, my iPhone 16 Pro's battery health is at 95% and that is after 451 cycles, which is not particularly impressive. I bet people have batteries, phones that they charge 100% on MagSafe all night that have the same stats. But I'm happy with how it went.

I liked only charging to 80% during the whole year. My iPad is permanently on 80% limit because it never leaves my bedside, so it doesn't need to be. And when I went to 100%, it kind of felt like getting a whole new battery. You know, like, wow, this phone still has amazing battery life because I'm using the last 20% for the last two months .

So good job, iPhone 16 Pro. iPhone 18 Pro. Battery-wise, I did set it to 80% because I think it'll be fine with this phone as well. But first, the important stuff. The case situation shockingly resolved itself amazingly well this year.

I guess over the past couple of years, it's just been, you know, a problem with, like, the camera control. Or I guess maybe it was just the last time, two years ago, with the camera control and getting cases with the big opening there. And then the camera control button passed through but not having the open bottom, blah, blah, blah. Anyway, for new listeners, my preference is to have a leather case for grip, essentially, with an open bottom.

Because when I swipe up from the bottom to unlock the phone , I don't like going over the lip of a case. I think the last time I got Apple's clear case, which has had an open bottom for many years now. I got the Apple Beats case, which also has an open bottom and is also from Apple and was fine. I've tried silicone cases without the open bottom and recon firmed that I really wanted to have the open bottom.

But this year, Bullstrap, as soon as the iPhone 18 Pro was announced, Bullstrap had for order immediately the case. This is before you could even order the phone. So the first thing I ordered this year was my Bullstrap leather open bottom case with the camera control button. And it arrived before the phone.

And I put it on my phone. Problem solved. I mean, it's a very expensive case, but it is exactly what I want. And it's on my phone now. So boy, this turned out to be a non-issue this year. I just bought the case they wanted. Oh, and for the bra effect on the back, the Bullstrap case has a kind of curved lip, curved leather color lip around the camera plateau.

But nothing is covering the camera plateau. So it protects the edges of the camera plateau, but it doesn't cover any portion of the camera plateau. And I think this is great. Like, I really don't like the ones like Ryan London only has one with like a metal wall around the camera plateau. And I find that unpleasing on my fingers. So, yeah, got it in one.

Bullstrap case. We'll put a link to it in the show notes. Again, it is expensive, but it is exactly what I want. So thumbs up. The other thing to note about my iPhone 18 Pro is the setup process. I'm going to give myself a thumbs up this year for not having to call Verizon for the second phone in a row.

Congratulations. Yeah. I mean, eSIM, maybe I'm going to chalk that up to eSIM making things better or whatever. But yeah, the number actually transferred successfully, even in my house on iPhone day. That's usually the problem is I'm doing it on the day that everyone gets their iPhones. Phone number transferred over.

Fine. Then I had to do battle with Apple's setup process. I once again chose the same thing I chose for the past several years, which is phone-to-phone transfer. on the idea that it would potentially be the fastest and also have the most fidelity.

Even though iCloud, restoring from iCloud backup, the fidelity of that has gotten much better over the years. It may actually have parity with phone-to-phone transfer, but I always want to try phone-to-phone transfer first. This year, I also decided to try connecting them with a cable because I figured the iPhone 16 Pro has sufficient enough speed from its USB port.

Maybe I didn't even bother looking at the specs. Maybe I'm wrong. But anyway, it's a thing I tried. Wait, now hold on. Which cable though? Because obviously there's like a million different choices here. I didn't know what to do with the cable. I ended up using a Thunderbolt 3 cable. Maybe that was the wrong choice. But I connected them butt to butt.

And I could put that they're both MagSafe, so I could put them on MagSafe chargers so they're also charging. You know what I mean? Yeah, yeah, yeah. So anyway, I'm doing phone-to-phone transfer. I'm doing the whole procedure. He makes the little fuzzy code. Make the camera. See the picture of the little fuzzy circle.

And they go through and it says finish on your other phone. And I got them on MagSafe chargers. And, you know, they're doing their thing. And they're – I don't know if it's using the wire. They're connected with the wire. But I have no idea if it's using it because it's not telling me. I don't know. Whatever. I just leave. I leave them there. I go off to do something else.

Come back like an hour and a half later to an error on the screen that says, oh, the transfer failed. Sorry. But you were definitely using the cable then because I can tell you I have – I did not buy any phones this year, not yet anyway. And so I didn't have this process in my house.

But based on my prior experiences and based on the handful of texts I got from random friends, some nerdy, some not, the iPhone transfer and then later setup process is such a dumpster fire. And it is my belief based on no facts. It's just an assertion.

But my assertion and belief is that the only properly v etted way to transfer between iPhones is via iCloud. I don't think Apple ever tests anything else. They claim to support it. And I don't want to do an iCloud backup for the same reasons you cited, John.

But I think that's the only one that Apple ever properly tests. And everything else, it's just a happy miracle if it works. And iCloud will get you back, quote, unquote, up and running sooner. Because if you're doing the phone-to-phone transfer, you have to leave both phones, like, alone and not touch them for a long time. Whereas the iCloud transfer, it will let you at least start using your new phone immediately, even though half your apps won't be there and stuff or whatever.

So that was attempt number one, which I was like, you know, all things considered, that wasn't that bad. Like, I don't know when it failed. Maybe it could have failed within the first 10 minutes, but I only looked at it an hour and a half later because I assumed it would succeed. Then what followed that was a comedy of errors of me trying to – it's kind of like a video game where your first try on the level, you get pretty far, and then you try to level a bunch more times and you can't even get close to where you got the first try.

That's what it felt like. And the basic problem is that when you're doing a phone-to- phone transfer, first of all, the two phones have to be near each other. I got rid of the cable at this point. I ditched the cable. I forget about the cable. We're just going to do phone-to-phone transfer wireless, okay? The phones have to be near each other.

If they're not near each other, they get angry and the transfer won't work. Also, you have to do stuff on both phones. The phone will say, do this, do that, and you can go to this phone and do that and type your code in on this phone and type the unlock code for that phone on this phone. You're going back and forth between the two phones. You've got to be juggling that as well.

And the final thing is when you're essentially done with the old phone, when you no longer really have to interact with it anymore, you hope, it puts up a sheet or whatever they call it, a little card, like a little white square that takes up the bottom part of the screen, and it says, finish on other phone or something like that.

You've seen this, right? Yep. Yep, yep. And it has a little X in the top of that sheet, and it's like a spinner and it says, finish on other phone. I'm like, fine, I'm perfectly happy to finish on other phone. Now, you may be tempted to say, finish on other phone. All right, well, I guess I can just dismiss this white box here and just finish on the other phone.

No, if you hit that X, the thing ends. And when I say the thing ends, what I mean is not only are you no longer transferring between the phones, but you have to start all the way from the beginning, as in it makes you erase.

Yeah, the destination phone has to be restored. The new phone. Erase it. So no matter how far you think you were through the process , you have to erase it. And how far might you be? Well, there's a bunch of stuff you have to do. It's like, you know, set up Siri, set up all your Apple Pay cards, sign into your Apple ID. Like, you can skip a lot of these, and I highly recommend skipping them because they take a long time.

When you get up to that point where it's like, you know, finish another phone or whatever, you may have been there for five, ten minutes already going back and forth through card issuers, entering three-digit codes from the back of your cards to set up Apple Pay, setting up the hey-dingus stuff. Like, there's a bunch of stuff that you're doing before, you know, you accidentally dismiss the thing or whatever.

So the second time I believe what happened is that that little white card on the old phone disappeared, but the new phone didn't know it disappeared. And I was like, is this going to work? No, eventually it said, I don't know what's up with the other phone. I don't know what happened that time. It just didn't work.

The third time really got me because the third time I'm like, all right, third time's a charm. I'm going to do it this time. I've been through the process twice already. I know exactly what all the steps are. I know which ones I can skip. So I'm doing it like a pro. I'm like, I can skip this. I can skip this. By the way, I'm skipping them because I just want to get across the finish line.

I can always set up the stuff later. I can set up hey-dingus later. And by the way, last year, the hey-dingus one is the one that really got me because it was like, hey, set up Siri here. Say, how's the weather? Say, set a timer for three minutes. Say all these things or whatever. But it makes you say, say, hey-dingus, set a timer for three minutes.

And if you do that, last year, or not last year, last phone , because the other phone was nearby, it would answer me. And when it answered, set a process over, erase the new phone. Oh, gosh. I swear to you, they don't test this. I swear to it. Now, this year, I think they improved it.

But I did do the thing where I took the new phone, what I felt like was a safe distance away, not out of the room, but away. And then I whispered to it. I whispered to it. So the new thing- Wait, but if you whisper to it, then you're not doing the good- You're not doing a good job of giving it sample material.

I think you can skip that step anyway. So anyway, that's that. So the third try, I'm like, I know all the tricks. I'm getting through on the third try. Everything went. I did all the stuff, skipped all the things I could skip. I was careful not to activate the elephant or whatever. And I'm like, now I just need to sit there.

They're doing their transfer. I just need to connect them both to power. Because that's the other thing. I would suggest not trying to have them on chargers when you do this, for reasons I'll get to in a second. But they need to be on chargers. Because at this point, on my third try, their batteries are drained a little bit from, you know, the hour long.

Anyway, so I go over. They were plugged in during the hour long thing. But anyway, just going through it, the batteries are a little bit drained. So I carry them both over to our little household charging station, where I have tons of cables to plug in. And I plug the new phone into a cable and put it down. And I take the old phone and try to plug it into a cable.

And, you know, these phones have buttons all over the outside of them. And I guess I squeezed a little bit too hard when I was plugging in the cable. And I accidentally pressed in the camera control on my old phone. Oh, God. Transfer over. Erase new phone. Oh, my God. This is so bad, John.

Erase. And I'm like, no. Like, it's just at the last. All I needed to do was plug it into power and then let it sit there. But because I pressed the camera control accidentally, that obviously means erase, you know, end entirely. No confirmation. Nothing.

Erase and restore the phone. And every time you erase it, it's got anyway. Fourth time, I was super duper careful. And I actually got it transferred. And it went fine. But boy, that was just ridiculous. Like, you would think that they would, like, make it more resilient.

Like, they would make it kind of like a, on the Macs, which I'll get to in a bit. Like, make it like a locked mode where the old phone, like, none of the buttons do anything anymore. And it's just locked. But instead, it's the most fragile state. If you do anything on the old phone, it's like, well, obviously, whatever you're doing takes priority and screw that transfer.

Even if that transfer had been running for an hour or whatever, it doesn't matter. So, yeah, they get set up. I guess the only thing I've done with my phone that's interesting is I took a picture, took one of my usual pictures of my dog at the foot of my bed with most of the lights off in my room. And it's better at low light photography with the big aperture.

All right. Because I've taken that same photo of my dog a hundred times in various poses. So, I have lots of history to compare to. And all I do is pick up my phone, point it at the dog, and press the, you know, the camera control, the shutter button. And so, I can say, yeah, better, noticeably better low light performance than my iPhone 16 Pro, presumably due to the larger aperture.

So, thumbs up on the camera so far. Well, that's exciting. That's, I mean, I'm sorry for the absolute disaster that was the transfer process, but at least the aperture or the variable aperture is showing some signs of improvement. It could have gone worse.

But, you know, the main thing I'm happy about is I like the color. I like the case. Everything's all settled. My new phone was like, other than the silly setup process, was the best thing. The best new phone setup process I've had in, you know, a couple of phones. I'm sorry, what color and what storage did you get? Oh, I have 512 because I'm a little bit over 256.

I'm like a 380 in my storage. So, I've got a 512 and it's black. Okay. Thank you. AirPods 5 with wireless charging case, meaning the ones that have the volume control thingy on them. So, a couple of things to note about these.

I forget if we mentioned this before, but apparently these are IP57 as compared to IP54. This is a waterproofness and dust ingress rating. So, the old AirPods 4 and AirPods Pro 2 were IP54, but the AirPods Pro 3 are IP57.

And IP57, the 5 stands for how much like dust can come into it. So, 5 is ingress of dust is not entirely prevented, but it must not enter in sufficient quantity to interfere with the safe operation of the equipment. So, dust can get in, but it can't screw it up. And then the 7 is the important part. The 7 means immersion in water up to 1 meter for 30 minutes .

The 4 means it can take like splashes of water, but the 7 means you can put this in up to a meter of water for half an hour and it should, in theory, still work. So, that's impressive and I'm excited to get even more waterproof AirPods, although I don't plan to test that. The other things to note about the AirPods 5, some odd behaviors.

And unfortunately, two variables changed at once here. I got new AirPods and also I got a new phone. So, I'm not sure if I should blame this on my new phone or on the new H3 chip in the AirPods or whatever. But here's the odd behavior. First, when I put these things in my ear and start playing music, it pretty much all the time does a thing with the sound that makes me think, did I accidentally enable head tracking?

You know that? Like, you ever accidentally enable head tracking? And it's like, oh, no, I don't want this. Where is that? Turn it off. It always does like a thing where I hear sound like in one ear and then it transfers to the other ear before they even out. It's not head tracking. It's disabled, right? I have disabled head tracking everywhere.

But it always does this weird thing where the sound travels usually from my left to my right for a second before they settle in and they even out. Really weird. The volume control, I had to watch a YouTube video to figure out how to do it. I've seen it a million times. I said it in the Apple keynotes when they added it to the AirPods Pro years ago.

Oh, it's a little area and you swipe it with your finger to change the volume. So I finally get a pair that has this and I try to swipe with my finger to change the volume and I cannot do it. Weird. I'm like, what is the secret? Tell me the secret. And so I found a YouTube video where someone was like, listen, here's what you have to do. Grab it like this.

Put your finger above the flat area. Swipe downward, starting above the flat area and then ending after the flat area and do that periodically and you'll hear a little boop and it will go and do the reverse up and down. So the video helped. Thank you to that nameless video that I did not save that taught me how to use the volume control.

I am glad to have it. The other weird thing. So one of the reasons I don't use noise cancellation that much in my AirPods 4 and probably won't use it much in my AirPods 5 is noise cancellation means there's a little microphone that's listening to the outside world so it knows how to cancel those sound signals.

And if you're wearing a hood or a hat or something that r ubs against the AirPods, the rubbing against them can make like noise on the microphones. I don't know if you've ever experienced this with the AirPods Pro, but definitely with the regular AirPods. If you rub something on the microphones, you're like, shh, like I don't want to do it by a microphone.

But you know, the sound of something rubbing up against a microphone, extremely unpleasant, especially in the winter when I have like hats on and stuff that are like rubbing on the AirPods, covering my ears. So I'm like, oh, I don't want that. So I turn off noise cancellation. The AirPods 5 with wireless charging case do this weird thing where when I'm taking them out again, I have noise cancellation off.

Take out the right AirPod. I always take out my right one first, put it in the case. Then when I go to take out my left AirPod, I get static iness when my finger touches the little microphone thing. And the microphone thing is in the exact same place as it was on the AirPods 4. So it's not like I'm like, you know, hitting a thing that wasn't there before.

But now when I take off the left AirPod, it's as if it has turned on noise cancellation briefly because I hear that loud, you're rubbing your finger against a microphone and it's playing that rubbing into your ear thing. Very weird. Minor issues. I'll see if they settle down, if there's a firmware update or whatever, but some oddities.

And as for the actual noise cancellation, it is better than the 4s. My usual test is, can I wear them with noise cancellation enabled while vacuuming? And the answer with the 4s is no, not unless I really crank the volume. Like the test, the real test is listen at a comfortable volume, then take out the vacuum, turn it on, start vacuuming.

Do not change the volume on your AirPods. Because if you crank them up enough, it'll work in any condition. But then when you turn off the vacuum cleaner, you're like, oh, this is way too loud. I've been blowing my ears out, right? The 5s can have better noise cancellation and I can just about listen to podcasts while vacuuming.

So thumbs up on, I don't think I'll do this often and it's not what I prefer and it probably would be better if I raised the volume a little bit, but I can listen to podcasts while vacuuming. So believe the hype about the slightly better noise cancellation in these. All right. So now generally speaking, you don't use noise cancellation .

Like if you were on a plane, which I know is your favorite place to be, would you use noise cancellation? I use the big over ear Sony, like mega expensive noise cancellation headphones for that. I figure what they are with it. Whatever the latest MX, the highest MX number is. That's what I have. But day to day, you don't use it generally.

No, I don't use it at all really. For that reason, because like in the wintertime, the hats rub on it in the house when I'm listening to like watching a TV show on my iPad on my bed or whatever . Like I can hear the noise cancellation if that makes sense, especially since they're, you know, they're not, they're open ear things. Like I can hear the effect of it and there's not noise that I really want to cancel anyway.

So I don't use it. And it used to be for battery life too, but so far the battery life on these has been great. We are sponsored this episode by Daft Music, a native Apple Music playing app for Mac and iPhone, custom designed for each and built for the liquid glass generation.

Daft Music is a wonderful third party Apple Music app. Now it isn't just an Apple Music client. It uses Apple Music as a streaming backend, but it does things its own way. So there's all sorts of wonderful features in Daft Music that are opinionated and designed for Apple nerds like us. So they have a library for your favorite songs, albums, artists, playlists, and you can pin everything within the library.

They have a directory of custom selected mixtapes. And this is picked by humans, not AI. In fact, they even have a new feature that you can check any track for AI that will flag recordings that have AI fingerprints left behind by the music generators. So you can tell like, you know, if you want to listen to AI or not, you can easily tell what's AI inside Daft Music.

They have a radio feature that Shazams every song from internet radio tracks as they play. You can import and subscribe directly from Spotify and T idal. And all of this is very lightweight on your machine because it's written with native code, Swift UI and AppKit.

And they have features always powered by the latest Apple techs like just Core AI. They're constantly issuing new updates to Daft Music. There's AirPlay support that came recently. A wonderful feature called instrument isolation that you can actually mute the vocals or drums or bass or everything but those things live during playback , all done on device.

There's a 10-band equalizer, 3D cover art support, and so much more all in Daft Music. So go to daftmusic.app, that's D-A-F-T music.app, daftmusic .app, or just search for Daft Music in the iOS or Mac App Store.

So daftmusic.app, thank you so much to Daft Music for Mac and iPhone for sponsoring our show. This is the real deal. All right, tell me about the Mac Studio. Can you first remind us what are the basic specs of this, please?

This is an M5 Max Max Studio with 64 gigs of RAM and a 4TB SSD. It's the Max with all the cores working. I think those are the only options. And then I have an external 4TB SSD. We linked all these things in a past episode. But that's the setup. What color did you get?

Yeah. So the setup process, like I had prepared the way pretty well. The setup process was surprisingly easy. It does fit on the corner of my desk where I wanted it to go. Although the hole that's underneath it means that the foot doesn't have as much contact friction with everything else.

So I'm not sure what I'll do there. Like it's a little bit easy to move. Although people shouldn't be reaching back there to touch it anyway. I basically have almost every port filled in in the back of it, which doesn't feel great. But when you're coming from a Mac Pro with a million ports, like this is just the way it is. I'm glad I didn't need an additional port.

I don't have anything in the HDMI port. But I think that's it. Every other port of the Mac this machine has something plugged into it. But yeah, all the wires, you know, reached and everything. I didn't, I haven't yet done the thing where I'm going to connect all of my external SSD cases to the back of my desk .

They're currently sitting on my desk. I was deferring that until I got done with the setup process. I only got the computer two days ago, so it hasn't happened yet. But I will probably do that in time for next week. But maybe not in time for the next episode, which we're recording early. Anyway, the transfer process, the migration assistant.

I haven't done this in seven years. I'm usually a big fan of migration assistants. I don't start fresh. I use migration assistant. But here's the thing with migration assistant. I feel like this app, for all its amazing capabilities, has not kept up with the advancement in storage size.

When this thing was made, how much data was migration assistant moving? How many files was it moving? The numbers must have gone up at least by an order of magnitude since then. And so a lot of things in the app don't give you any feedback that would let you distinguish, is this stuck?

Or is this just taking a long time to do a thing and eventually it will be done? Now for this one, I had, I mean, it's kind of overkill since my 2019 Mac Pro just has Thunderbolt 3, I think, right? But the M5 Mac Studio has Thunderbolt 5.

So I had a Thunderbolt 5 cable. Certified real Thunderbolt 5 cable. And I'm going to connect my Mac Pro to my Mac Studio with a Thunderbolt 5 cable while I do migration assistant. And one of the things migration assistant has done for many years is it'll, like, test your connection and say, this connection has been tested to support these speeds.

So I know it found my cable. I know it tested it. And it tested it at really high speeds. So I'm like, great. This will help because I'm transferring four terabytes. It's my Mac Pro at a four terabyte SSD. Mac Studio has a four terabyte SSD. I'm going through my migration assistant, go through the stuff, type these numbers in this computer, blah, blah, blah.

I was basically just screen sharing to my Mac Pro from another Mac so I could see its screen and, you know, doing all the stuff. And it says, what do you want to transfer? It defaults to everything being checked off. I'm like, yep, transfer everything. Go to the next thing. And then you get an indeterminate progress bar, like the progress bar where the filling goes back and forth, left, right, left, right.

You know, that thing. And I think it says, like, starting up or something like that. And it stayed on starting up for a really long time. So long that I had time to start Googling. Migration assistant stuck on starting up. You know, these type of questions.

And if you Google that, boy, will you find years and years of people who said, hey, I tried to use migration system between two Macs. And it showed a progress bar going back and forth. And it says starting up. And it's not making any progress. What's the deal? Years of it. Like, I'm not going to say decades because I think the longest one, oldest one I saw was maybe six or seven years old or whatever.

But this is not a new thing. It's not a good feeling when you do a query like that and you get that answer. Then I start going through some of them. I'm going through, like, Reddit posts from multiple years ago. And it's like, yeah, I've let it run for 20 hours. And it still says starting up. Like, oh, no, I have a podcast tomorrow. 20 hours.

And people are saying, should I just let it go? People are saying they called AppleCare. And here's what AppleCare told me to do or whatever. Then you start getting this sort of received wisdom, like, you know, folklore stuff where a couple of people said, here's what you have to do.

Start over. Okay. And then when you get to the screen where it tells you, like, what do you want to transfer? It's got little checkboxes. Next, like, do you want to transfer the apps? Do you want to transfer this user account? Do you want to transfer system settings? Like, the screen where it comes, where everything is checked off by default said, look, when you get to that screen, do not leave that screen until it calculates the sizes of all those things.

Because they all just say calculating, like, for the sizes, right? You have to stay there until it calculates all the sizes. So this is slightly complicated by the fact that next to each item it says calculating. But then eventually the word calculating disappears and it 's replaced by a number.

Like, let's say, 10 gigabytes. But that doesn't mean it's done calculating. Because if you keep watching, that 10 will change to 12. Then to 15. Then to 30. And you're like, how do I know when it's done calculating? I guess I have to wait until none of the numbers are changing anymore.

Anyway, the calculating process takes a long time. A long time. Because I guess it's counting up every file and every size. I don't know what it's doing. It's a four terabyte drive with literally millions of files on it. But the thing I came up with is, like, look, the sizes should total about four terabytes.

Because my disk is full. So I'll wait until the, I'll wait until the, first I'll wait until nothing says calculated. It says, you know, nothing says calculating anymore. Because it does just like one or two of them. And then eventually it'll start on the third or whatever. It doesn't do them in order. I don't know. But some of them just say calculating for a long time.

I'll wait until all the numbers add up to roughly four ter abytes. And they're not changing anymore. And then I'll hit the continue button. So I did that. Because many things on the internet said you should do this . Of course, in that same thread, there's a bunch of angry people years later coming back to the same Reddit thread saying, After doing that, how long do you have to wait?

How long did it take to finish? Please tell me. But they're long gone. Because this is years ago, right? You know, this is the sad people coming back. Wait, I found your thread. How long did it take after that? So I waited for all the calculating to be done. I hit continue. And then it says starting up. And it shows an indeterminate progress bar going back and forth.

And it stays there for a long time. Can you give me an order of magnitude? Are we talking minutes, hours? I mean, I'm just trying to remain calm. And I just go and do something else for a while. And I look back at it. And it's like, all right, it's been 30 minutes.

I'm doing some more Googling. I'm not finding any more folklore wisdom except for a bunch of people. A bunch of people from back in the day are like learning that you should never do this over Wi-Fi. I'm like way ahead of you. Like, I know. I know. I'm not doing it over Wi-Fi. I'm not even doing it over Ethernet. I'm doing it over Thunderbolt.

And by the way, the Migration Assistant app is not reass uring. And at the bottom, it says like what your current connection is. And it says none the whole time. That's great. But when you click on it, when you click on it or go see like the connection test thing, it will tell you what it tested all the connections for.

So it's like, obviously, this is not none. Because you said this connection has been tested. And it just gave this data rate or whatever. And it tests all of them. So you've got an Ethernet connection. And I tested that. And it gave this data rate. And you've got a Thunderbolt connection. And like, but the thing just says none. All right. So I just let it sit there for a while. I figured, okay, I will give this, let's say, three hours.

To be on the screen that says starting up or preparing. Whatever the hell it says. It is not a progress bar that's going anywhere. It is a back and forth progress bar. And there is no other visibility into this. There's no log window to bring up. There's nothing else that you can see to let you know that this is making progress or not. Maybe before I waited for all to calculate, it would have also completed if I had waited the same amount of time.

Who knows? But I just found something else to do and put it out of my mind. And probably, probably about an hour in, maybe 50 minutes in to the starting up thing. I glanced over from what I was doing on my wife's computer. And I looked back at my computer and I said, it's not going back and forth anymore.

Now there's an actual progress bar. All right. And that's when I sent you guys screenshots with my fingers crossed emoji, where it's just like, it's, it's going and it's giving me a data rate and the data rate kept climbing. I think it was topping out at like over a gigabyte per second, which is great. And it took maybe four hours to finish.

And it did finish. And the computer rebooted and had pretty much all my stuff in it. And I was once again, amazed at the thoroughness of migration assistant. I think it let me inherit all of my time machine backups, including the network ones on the new computer, which is amazing.

Like I didn't take it up on that offer. I wiped the time machine back up because I just wanted to start fresh because I'd know from experience that the network ones get corrupted easily. But I think it let me do all that. All my settings were there. All my stuff from user local was there. Just like everything was in there.

All, you know, it was really amazing. Backblaze was a little bit angry and I had to corral that back into a reasonable state. But yeah, pretty good after getting over that hurdle. But this is this is the this is the flaw in migration system. They need some way like for phases that previously were like, well, this isn't going to be that long.

We can show an indeterminate progress thing here. They can't do that anymore. Like what if you have millions and millions of files like I do? What if you have four terabytes of data spread over millions and millions of files? There's no way to tell whether this is ever going to complete. And most people won't have the patience to wait an hour for it to get to the part part where the progress bar begins.

Right. Before you start seeing it. Once it begins, it gives you a time estimate and you can see it progressing. As long as that progress bar is moving, you think things are happening. But wow, a little bit scary. I'm not sure what I was going to do. Like I could have tried restoring from time machine backup or something else like that.

But I really wanted to do the machine to machine transfer. So I'm glad it eventually worked. And then after that, I've just been on Intel purge. Go to system report, go to like, you know, about this Mac, blah, blah, blah. You eventually get to what used to be called system prof iler. And that will list every app that it knows about on your system, including what architecture the app is using.

Although it lists a bunch of them as either unknown or other or something. But anyway, I went through every Intel app I had and either deleted it and updated it to an ARM version. I cleaned out user local, rebuilt everything for ARM. Didn't rebuild everything. Some things I didn't rebuild eventually will like when I need them. But the stuff I knew I needed, I rebuilt for ARM.

Built new versions of things that I wanted new versions of. Yeah, that's going pretty well. Did a network time machine backup. That's actually still running to my Synology right now. Did a super duper clone. Oh, the enclosure with my two SATA SSDs working okay.

But one time, I think yesterday, both drives, which were both mounted, just disappeared entirely from my computer. And disk utility could not see the enclosure, which is what everybody says. Oh, don't get external enclosures. They just, you know, your drives disconnect.

When that happened, disk utility couldn't even see the enclosure anymore. And my solution was to turn off the enclosure and back on because it's got a power switch. Hasn't happened since then. I have no idea what caused it. Maybe I'll try plugging it into power instead of making it bus powered. But I've got my eye on that one. But anyway, I did do a super duper clone.

Did two super duper clones to it because now I'm cloning my photos drive and my, what do you call it? And my main drive to two separate volumes on there. And I did a time machine backup to it. And that worked. And now the Synology time machine backup is taking forever because it's way slower. But yeah, I'm almost completely up and running.

And during that, while all that stuff was going on, I was doing dev work on my computer instead of my wife's. And she loved that. And I loved it. And it's nice and fast. And so far, I haven't missed the RAM that the Mac Pro had. This is not. Well, so sounds like a little bit of a troublesome journey for both the phone and the computer.

But ultimately, it sounds successful. Yeah, it could have been worse. Oh, an update on photo searching. My photos library did success. This is another thing. Copying your photos library following Apple's instructions and their support note that we put in a past episode. Copying my two terabyte-ish photos library to an external drive following their instructions.

The copy went really fast. Because, hey, I bought a fast Thunderbolt 5 SSD. And I have a, you know, a copy went surprisingly fast for two terabytes worth of library. And then following Apple's instructions for how to open it. And then it says, and then you want to designate it your system photos library.

So go to, you know, settings general. And there's a button that says use the system photo library . And I click that button. And photos beach ball. And that beach ball stayed there for 20, 25 minutes. Oh, jeez. And then eventually wasn't there anymore.

And it was successful. That is another instance of no regular user is going to tolerate that amount of waiting. But I was like, look, what's my alternative? I've learned that sometimes it's just going to block. I mean, the rest of the computer was fine. But photos was 100% beach balling for 20 minutes.

Doing, you know, I have a big photos library. There are a lot of photos. Like, I get it. I don't know what it was doing. But it did eventually succeed. And my photos library did move there. And then I deleted my photos library from the internal SSD, freeing up half of my space. And then once it was all said and done, and once it had done all iCloud syncing and everything, I did search inside photos on my photos library that's on an external drive.

And I searched for pictures of cats and pictures of fences. And both of those things worked. And you did this within photos? Within photos. I don't know or care anything about spotlight finding photos. But within the photos app, when I search for some word that , you know, it works. It works. It worked fast.

I'm satisfied with that experience. I think I'll be okay with my photos library and external drive. And that drive enclosure has not spontaneously disconnected so far. So far. I thought that we were of the understanding that searching within photos was never really up for grabs.

It was searching outside of photos and spotlight. That was the problem. Well, there's a follow-up item in the second where we'll see what Apple's intelligent agent has to say about this. But just experimental evidence. I did not try any kind of spotlight search for photos like as in Command F in the Finder because I don't care.

All I care about is that search inside the photos app works . And it does. Good deal. So all in all, it sounds like success all around. Yeah. So far, so good. I mean, yeah. It's nice to have everything all in one place. I think that time on my wife's computer has really made me appreciate this more.

Made her appreciate it more, obviously, too. But just being back on my own computer with my bigger screen, my own keyboard and mouse. Oh, that's another nice thing I got to treat myself. So I gave myself a new mouse pad. Wow. The old one was fine. But, you know, there's that nice new mouse pad feeling, right?

So I have a new mouse pad and, of course, my keyboard that I bought on eBay turned out not to be a scam. It was, in fact, a new sealed inbox Apple extended keyboard with a full size control key in the lower left corner and no stupid globe thing there and touch ID. So here I am.

New keyboard, new mouse pad. Not a new mouse, although I do have plenty of spares of those up in there as well. So, yeah, it's got that. It's got that new Mac feeling. That's awesome. Congratulations. That's, you really have, you've earned this over a very, very long time.

Hopefully I won't have this computer for seven years, but tune in seven years from now to see how that goes. Good gracious. Does the new computer feel appreciably faster than the Mac Pro did? Oh, yes. Okay. All right. Good deal. Remember, my computer was slower than a phone.

Well, yes. And it showed up like, Xcode builds are so much faster. Yeah. There's no idea. Everything is a lot faster. We are sponsored this episode by Click for Sonos, a premium Sonos companion app.

No lag, no hassle, just music. Click for Sonos is the fastest native Sonos app. It runs on iPhone, iPad, Mac, Apple Watch, even Apple TV and the Apple Vision Pro. And of course, one purchase covers all those devices. Now, as an Apple nerd, you probably expect things like widgets, live activities, shortcut support.

Click has all of those, plus so much more. You know, lock screen support, control center support. On the Mac, they have a menu bar item. And they have all sorts of amazing features for Sonos users . So, for instance, instant control, song changes show up the moment they happen with no lag.

They've unified search across all the different, you know, Apple Music platforms, Spotify, Plex, Tidal, Deezer, Pandora, Sonos Radio, and so much more. All search together at once in one unified list. They have amazing tailored scenes. You can have one tap to change volume, speaker groups, music, and sleep timers, including a timer that ends at the end of the current song.

And it is all backed by these amazing features. Things like, you know, for audiophiles, you have Lossless and Dolby Atmos support, Arc Ultra speech enhancement and dialogue sync. And this is an actively updated app. There's ongoing updates all the time. There's been eight major releases already this year with more on the way.

Click for Sonos is just the best Sonos companion. Again, no lag, no hassle, just music. You can get lifetime access for 50% off using our special link, click.dance.atp. That is, click is spelled C-L-I-C.

So it's C-L-I-C.dance.atp. That gives you lifetime access for 50% off. Click.dance.atp. Thank you so much to Click for Sonos, the fastest native Sonos client, for sponsoring our show.

All right, Marco, you also got some treats. What's going on there? Yeah, so I went to the Apple store for my pickup of my new Apple Watch Series 12. And I got to see all the other products.

Very briefly, the iPhone Pro colors, I think, are fine. I actually, I went today. I needed to go to the mall anyway. And so I just popped in briefly to look at the two new colors. Actually, I didn't even look at black, come to think of it, I should have.

But I saw Glacier and Burgundy. And they clearly are not as cool as orange was. But I actually really like both of them quite a bit. I find, at least, you know, from seeing it in the store for a few minutes, I find that the Glacier, which is that light blue color, the Glacier is lighter than I expected it to be .

And the Burgundy is darker than I expected it to be. I think that's fair. Yeah, seeing YouTubers with the Burgundy, like Apple's videos made it seem so much lighter than the spy shots. But now that I see other YouTubers doing it in different lighting conditions, it looks more like a spy shot. So I think the Burgundy is pleasingly darker than it appeared to be in Apple's videos.

Yeah, it's very dark. Like in moderate to low light, it could be confused for black. It is very dark. And I think that is fine. And I mean, that is kind of like, you know, the style this year is a lot of like reds and purples and stuff. So, yeah, that makes sense. I think, honestly, they went even darker than maybe the style would suggest.

But I'm not a style expert. So I'm actually very happy sticking with my 17 Pro for one more year for lots of reasons. As the various reviews have come out of the 18 Pro, it does seem like it's not really that necessary to upgrade every year, even for idiots like us.

And so I'm I for the moment, I'm going to stick with my 17 Pro. Now, that being said, part of the reason is that I'm waiting for the duo. So I know I have to get one of those, at least for development and everything. So that's going to throw a wrench in the plans. But for the moment, happy with my 17 Pro. 18 Pro does not really do it for me.

And in the same way, I also didn't ever get the Apple Watch Series 11. I'm coming from a Series 10. And I actually, for various developing and testing reasons, I ended up having the Series 10 in both of its sizes.

One of them came refurbished later on. And I just got a Series 12. I decided to go with the smaller size Series 12 again. I spent a good amount of time wearing both sizes on the Series 10. Like I wear one for a few months and switch the other one for a few months.

What really got me with the Series 10 is a couple, like about a month ago, I was out with some friends and they took a picture. And I saw the picture of myself afterwards. And all I could see was this giant screen on my wrist.

I'm like, hmm, I don't like that. Like, I think I want to start reducing that. And so I went with the smaller size Series 12. I got the white ceramic, which I think is awesome. I'll get to that in a moment. But just a little bit more on sizing.

The ceramic is actually a little bit bigger, especially it 's, I think it's like one millimeter wider than the other materials. And also, there's been a change in the screen proportions with Series 12. We're going to get to a little bit in the follow-up.

But I think, in my opinion, I think the small size looks way better on me than the bigger one in this model of watch . And also, I would even like it to be a little bit smaller. I think it could even go like maybe two millimeters smaller and still look great on my wrist, which I think, I mean, I forget.

I should know as a watch nerd. I think my wrist is like 6.75 inches, something like that. I don't know. Something in that ballpark. All right. Well, hold on. So since you've brought this up, we do have some follow-up we're going to get to later. But let's just do it now. And so Sebastian writes, the Apple Watch Series 12 display is smaller than the Series 10 display.

I got mine yesterday, and I noticed it immediately. Also, the speaker grill is linear again, while in the S10 and S11, or Series 10, Series 11, the speaker grills, they were speaker grills with small holes. Both of those are very disappointing changes, in my opinion , according to Sebastian. The Series 10 was a regression from the 789 with thicker bezels, and they made it even thicker with the 12.

So this has been a common refrain, and there's a picture that Sebastian took that makes it fairly clear. Like, I'm not trying to argue that this is incorrect. This did not even register to me. Maybe I'm just not observant enough, but did not even register. Well, there are some changes. Oh, and by the way, before I follow up, I just took out one of those little flexible tape measure things.

I have a six and a half inch wrist, and I think the small size is definitely the right size for that. Anyway, that being said, there are a few changes to some little tweaks to the design. Oh, my Series 12 just rebooted itself.

That keeps happening. Version 1.0 of the software. Anyway, as it reboots, I'll tell you. The black part, like the crystal of the watch, is a bigger proportion of the watch's height.

So it has gotten thicker, and the watch body below it has gotten thinner. So as you look from the side, you see more of the height is black than it was before. So I think that that might be contributing to part of the way of the difference in the look.

Honestly, I think that makes it look better, but that's just me, maybe. Now, that being said, people have taken side-by-side pictures, and indeed, the screen bezels are a little thicker on the 12, and they were on at least the 10, and I think the 11 as well. To me, I actually prefer this.

Now, part of it is because I don't use a watch face with a solid color background. My watch faces have black backgrounds, so I don't even notice this. But when the Series 10 first came out, we noticed, as we used it, and we talked about it, that the screen, by being pushed closer to the edges, it actually warped a lot of the content in the corners and edges of the screen to slightly be in the curvature of the crystal.

And I actually didn't like that. Like, it looked kind of cool for a second, but it made it a little harder to read, and to, like, because a little bit of the screen content was actually on the slope. And I didn't think that was a very good idea. It was great that they could brag about thin bezels, but it actually worked worse.

That's why, even though the outgoing, I don't know if the current ones do, but the outgoing large size, the 46mm Series 10 and 11, actually has a bigger screen than the Apple Watch Ultra. But because of that curvature around its edges, where the Apple Watch Ultra is flat edge to edge, I actually preferred, for large screen use, the Ultra by far.

Now, I think the Series 12, by having it pull the margins a little bit in and actually have thicker bezels a little bit , I think it's actually better because it helps pull that content away from those curved edges, and more of it is on the flat face of the watch.

That, to me, makes it more useful. It definitely does seem like a conscious change, because Apple was bragging so much when they got the bezels smaller and smaller on the Series 10 and 11. They just loved it so much. But it seems like they also learned that there are usability issues. It also makes me think about the 20th anniversary phone that is rumored that is going to take the screen and wrap it all around exactly the same way.

Have they unlearned that lesson for the phone? Are they going to use that setting that pulls in the screen image? Tune in next September to see how that goes. But for now, it seems like the watch team, at least, has made a conscious decision on the Series series of phones that have curved top-class things to pull in the screen image.

So, yes, it does mean that when you take a head-on picture of them, there is more black border around the new phone than the old phone, but that may provide better usability and also better read ability. Yeah. As Sebastian wrote, the speaker grills on the Series 10 and I think 11 are like a bunch of tiny holes.

Previous and now Series 12 models, it's two slots. This, again, as a Series 10 owner, I prefer the slots because the little holes on my Series 10 have had a little bit of sunscreen in them for a year and a half, and I have not been able to get it out.

So, like, the top speaker grill on my Series 10 just has these little tiny white dots. Like, nothing I have done can get that out of there. So, hopefully, the switch back to the slot might make it easier to clean. Plus, now my watch body is white anyway, so I might not notice as much. Oh, also, my outbound Series 10, I had the gold Series 10.

It significantly changed color throughout its life. Oh. The gold DLC coating tarnishes. And if you do some searches on Reddit, you'll find this. And, like, you know, if I flip it over, I could see, like, the parts that, like, rub against my skin are actually a different color than the rest of the case.

I tried... Never do this. I tried polishing cloths. Never do this. And they tell you, never do this. And AI will tell you, never do this. But I tried it, and it helped. But it still did not restore the original color. Never do this.

That is something that if you want a partial restore jewelry polishing cloths, never do this. Might maybe help you, but don't. Why are you saying never to do it? That it will damage the finish or something? Well, what polishing is, is abrasion. You don't mean, like, the Apple, like, microfiber polishing cloth.

Explain what you mean. No, I mean, like, a jewelry... Like, a, you know, a fine metals polishing cloth, like, from Amazon. You know, like, a jewelry polishing cloth. They have, like, tiny abrasive things in it. Yeah, it's a very, very fine abrasive compound. And if you, you know, rub metal with it, it gets it shinier .

But it does that by removing a layer of metal. Now, a DLC coating, which is what Apple uses to color these watches, is very thin. So, if you do too much of any kind of abrasion on it, you will rub off the coating, and you will just expose the, you know, the metal color underneath it.

Thus, the admonition to never do this. Yeah, that's why everything will tell you never to do this. I did not find that to happen on mine. But that's, hey, all bets are off if you try this yourself. That being said, that is something to be aware of. If you pick, like, the gold, like, anything that's not the natural color of titanium or steel, the coatings they use can tarnish.

And in this case, it absolutely did. And it kind of turned me off the gold now. Anyway, but going back to the white ceramic, so far, love it. I was a little concerned that it would be a little bit too much of a cream color. Because in Apple's pictures and stuff, it's kind of hard to tell.

Like, is it white or is it cream? Well, I can tell you, having the actual, you know, ceramic watch, and I have the, you know, their version of the white sport band, which they, I believe, call sand is the color. I have it right here next to the white sport band from the previous generations.

And it is not that different. It is a little bit more of a warm tone, but it's a very subtle difference. They call the watch pearl white and they call the band sand . I think those are pretty accurate color names. It is not, like, I was afraid it would be more like star light.

Like, they're, you know, like, they're kind of warmer versions of this. It's not. It is closer to white than starlight. And so I like it a lot. And I think it looks very, like, fresh and cool. It will restrict what band choices will look good with it. But, because, you know, because there's no metal to match.

Like, nothing with metal lugs will match it. But so far, I'm very happy with it. And that's it so far. I haven't tested all the, you know, spying on people features because those aren't released yet. For what I have right now, it is wonderful. Did you compare the new Ceramic Series 12 with your old Cer amic Series 5?

Or do you not have that anymore? I should dig around in my kind of archive. I don't think I have it anymore. I think I sold it. Oh, there was one other saga. Oh, gosh. Trying to pair this watch for developer mode. This has changed.

Now, I don't know if this is a Series 12 thing. If it's a watchOS 27 thing. Maybe some combination thereof. If it's an Xcode 27 thing. Trying to get this watch to show up in Xcode and be usable for development was ridiculous. The way developer mode works in Apple devices in the modern way is there's a secret setting called developer mode that you have to enable.

This setting will not even show up until you try to issue a build to the watch from Xcode. Then Xcode will error out and say developer mode isn't enabled on this device. Then you have to, at that point, go to the device. Go to the bottom of the security screen.

Find the developer mode. Turn it on. And it will say, all right, we got to reboot. So it reboots. Then after it reboots, it asks you, you sure you want to turn this on? Then you do the passcode thing. And then finally you have developer mode on. Then you can go back to Xcode and issue a build.

After it downloads all the symbols, it will eventually work for that device. Okay, the watch, previously, you would have the little dance where you'd plug the phone into the computer. And the phone would say, trust this computer, question mark . And you have to, you know, if you had the watch at that point paired to the phone, then the watch would also have its own tap, tap, trust this computer.

And you have to, you know, type in the code and everything. Well, no matter what I did, I could not even get that to happen with the new watch. I couldn't get it to even say, trust this computer. What am I doing here? Tried searching, tried AI-ing, tried finding any kind of documentation. I rebooted everything under the sun.

I installed both of their day one security updates. Like everything. I did everything. All of the usual dances. Could not figure it out. Eventually, I don't know how I found this, but somehow I figured out that there is, I think maybe I eventually found like a KBase article.

In the new Xcode device hub, there's a, like in the device list where normally it would show the devices that were waiting for pairing or that, you know, need to develop mode or whatever. It's not even there. You had to hit the little plus and say, pair nearby device. Now, the screen that shows up when you say that has three pictures, like a kind of multi-device icon, Apple TV and Vision Pro.

Now, if you look at this, you might miss the tiny little Apple watch that's in the multi-device icon. It kind of looks like this is not the right pairing screen for anything that's not an Apple TV or a Vision Pro or maybe, you know, an iPad or something.

Who knows? But it kind of looks like this is not the right screen. It is. Then, only then, with your Mac saying waiting to pair on this weird screen, only then will the developer mode option magically show up on the watch. Will it tell you this?

No. No, it will not. You have to just know that when you're looking at the waiting to pair screen, then you go into the watch and find the developer mode thing and then turn it on. Then, of course, we have to reboot, blah, blah, blah, go to the pairing screen. Okay. You reboot.

It says you sure you want to develop mode. Yes. By the way, the Series 12, I think, reboots faster than previous Apple watches, which is not saying a lot, but it is useful. Okay. You finally reboot. The watch is finally in developer mode. And you plug in the phone and you're waiting for that tap, tap, trust this computer.

And it just doesn't happen. What the hell? You know where it is? Not that it's going to tell you this. When you plug in the phone, then the time that you would normally find the trust this computer prompt on the watch, it's not ever going to tell you.

You just have to know to go back to the security screen, to go to the developer mode screen, to scroll to the bottom of the developer mode screen, where nothing has ever been before, and nothing has told you anything is there. But you just have to know to scroll there.

And then at the bottom of it, it says Marco Pro or your computer name here. And then you can tap that to pull up the trust this computer screen. Crazy. Only then can you finally use your watch for development.

Oh my God. Sounds fun. This took me hours. You mentioned the icon leading you to believe that that button is only if you have an Apple TV or whatever because it didn't look anything like a watch. I had a similar experience when I had to set up Apple Care on my Mac Studio, and I was doing it through system settings because it bugs you to do it there.

I'm like, oh yeah, I have to do that. And I go through it all, and it's like, yeah, do this, blah , blah, blah, blah. And then it says, okay, like on the Apple Pay screen where it wants you to fingerprint, authenticate, or whatever, to Apple Pay for your Apple Care Plus. It says, okay, so here you go. This is Apple Care for an M6 slash M5 Pro Mac Mini, right?

Like, uh, no. And so I back out. I go back through it again. I'm like, yeah, all right. I'm doing this on the Mac Studio through its own system settings. It's leading me to say, so you've got a Mac Studio. It's going to cost you this much for this, blah, blah, blah , blah, blah.

It's like it knows it's a Mac Studio. You go through the Apple Pay screen. Okay, so here's that Apple Care for the M6 slash M5 Pro Mac Mini, right? I just went through with the payment. And when it was done, my Mac Studio has Apple Care. But like, yeah, just a couple of missing string constants here and there.

An icon that doesn't make you think this is for watches because they can only put so many devices on it can lead you astray. We are sponsored this week by Quince. Let me tell you, today in Richmond, Virginia, it has been cold, it has been rainy, and it has been gross. And today was the first time since, I don't know, March that I wore jeans.

I didn't wear shorts because generally speaking, that's all I need around Richmond. And so when I put on my jeans, I thought to myself, oh, maybe it's time to take a look through my closet and kind of take an assessment of my wardrobe. And when the seasons change and it starts to get cool, that 's a great time to do it.

And let me tell you, if you want to refresh your wardrobe, Quince is a great way you can do that. Quince focuses on high-quality wardrobe staples made with premium materials like 100% Mongolian cashmere, organic cotton, and merino wool. The styles are timeless, versatile, and designed to become the dependable pieces you reach for day after day.

Further, everything at Quince is priced 50% to 80% less than similar brands. They work directly with ethical factories and cut out the middle people. So you're paying for high-quality, not brand markup. So I was given a little bit of money to spend at Quince over the last few months.

And one of the things I got was the Comfort Stretch Travel er 5 Pocket Pants. And these are a really great in-between where they're more formal than like jeans. But I wouldn't say they're quite as formal as like slacks or something like that. They're just very nice in-between. And this is a very true story.

Earlier today, I was starting to think about what I was going to pack for the podcast-a-thon next week. And it occurred to me that my Comfort Stretch Traveler 5 Pocket Pants would be the perfect thing to wear during the podcast-a-thon. A little bit more formal than jeans, but not so formal that I feel like I've gone overboard.

So if you want to get yourself some of these or some of the many, many, many other things that Quince offers, including stuff for your house, stuff to travel with, stuff for women and for babies, all sorts of different things, you can do all of that. And you can find your next fall favorites at Quince.

Download the Quince app for app-exclusive offers or go to quince.com. Get free shipping on your order and 365-day returns. Now available in Canada and the UK, too. That's Q-I-N-C-E.com slash ATP. Thank you to Quince for sponsoring the show.

Speaking of M6 Mac Minis. Oh, yeah. I got a new Mac Mini. Got my M6 Mac Mini, and I'm happy to say it is fast. You don't say.

I haven't done, like, formal benchmarks. I just, like, installed my transcriber consumer app and just saw, like, how fast can it go? Like, how many transcription minutes per real minute can it do? Like, what is its X speed of, like, you know, it's trans cribing at 200X, 300X?

So far, I've seen it peak at, like, 380X real-time transcription. I didn't see it sustain that, though. I've seen it sustain in, like, the low to mid 300Xs. Whereas the M4 Mac Minis before it would usually sustain more, like, 240X.

So I've gone from, like, 240-ish to, like, 320 to 330-ish. You know, something like that. But again, these are not, like, formal numbers yet. I haven't done controlled testing yet. I haven't had time because I've only had it set up for a few hours. But so far, it seems like the M6 is a substantial jump over the M4 in the area of transcription.

Which makes sense. Like, if you look at Apple's quoted numbers and some of their own performance claims and what they've focused on and what has changed, that does make sense. The only thing that I would really wish for, if anybody is listening to this at Apple who has any input in this, the 27 OSs have the new advanced dictation model that Apple has touted that runs on the highest-end hardware.

This is, like, that new form of model where it's, like, the sparse mixture of experts or whatever it was. Like, this new advanced architecture of these models. It's supported only on, like, the iPhone 16 Pro and up and the M3, I believe, and up.

Well, I have that. I have a bunch of them. All of my Mac minis are M4s or M6s. So I can run it. But there appears to be no API for that model on Mac or iOS . I have tried. I have run tests. There, you have this amazing new transcription model on this hardware.

And there's apparently, as far as I can tell, no way for apps to use it. And that's unfortunate. So I really hope Apple gives us an API for that. And even if it's as simple as, like, a flag to the speaks transcription API that, on those supported devices, can use the higher-end model, please, I would love that.

Well, you know, since you're running this on your own Mac minis, like, forget about that. I know you run transcription on the phone, too, but forget about the phone for a second and just think about your Mac mini farm where you're doing transcription. Private APIs. Send your agent to find them. They're in there somewhere. That's a good point.

You can use them. That's a really good point. No one can stop you. I never thought about that. Thank you. That's actually a really good idea. I am the devil on your shoulder telling you to use an agent to reverse engineer Apple's OSs and find those damn APIs because it will find them. Oh, that's really good. Oh, I like that a lot.

Thank you. Okay, I will. You like that until you're in private API hell like me where nothing is documented and things don't work. You're like, what the hell? Private APIs. I'm just going to walk away whistling like a cartoon character right now. Yeah, I mean, the good thing is when the disaster happens, it won't be on everybody's phones.

It'll just be on your Mac mini farm. So you'll reap what you sow. And it'll all crash at once. Foreshadowing for the after show, by the way. 50% of the Mac minis use private APIs and 50% don't. And 80% of the time it works every time. All right. So you good with your treats?

Yeah. All right. So for me, Aaron and I upgraded from series 10 aluminum watches to series 12. I had gone for a few years with the small watch. Now I'm back on the big watch, which seems fine to me. I think when it got less tall, it made a big difference to how I perceive it on my wrist, which isn't to say that Marco's wrong about him or me for that matter.

I think our wrists are roughly the same size. Yeah. And to be clear, like all three, like the 42, is it now? The 42, the 46, and the Apple Watch Ultra all look fine on us. Like it's just a matter of like, what's your current style?

Like I like a bigger watch in the summer and a smaller watch in the winter. Why? I don't know. But that's just what, you know, what I like. So all summer I wore the 46. All winter last year I wore the 42. And a lot of the summer I switched to the Ultra because it was even bigger. Like that's just, it's just personal style. In any case, so both of us got series 12 aluminums and it's nice.

That's about all I got. I mean, I'm not trying to be like, you know, snarky. It's just, it's a two-year newer watch. Seems faster. Battery life is way better. Seems good. I really just don't have that much more to say other than the bezel difference.

I think, Marco, you nailed it in saying that because I have a black background like you do, did not even register that the bezels are different. It's fine. It's good. I shouldn't say it's fine. It's good. Yeah. There's no complaints. It's good. That's about all I got. Awesome.

All right. So back to follow-up. I don't know what that was. Back to follow-up now. The iPhone Duo DevTools, there was a delay and there was much consternation, including for Marco and myself. And we're going to cover the actual tools themselves a little bit later in this very episode, or at least that's the plan.

But Oscar Echeverry wrote, I'm genuinely curious what Marco and Casey would have Apple do differently with regards to the dev tools for the iPhone Duo. What are your opinions around how Apple would be able to both protect its product secrecy while simultaneously providing more lead time for third-party developers to have access to dev tools before the Duo's release?

I'm very curious to hear what Marco's thoughts are. We did not confer in advance. For me, 27.1, the Xcode 27.1 beta should have been released that day. Like, I'm not clear. Maybe I'm missing something obvious. But I'm not clear why that is impossible.

Like, obviously, it's not what they did. But what precludes that from happening? I don't really get why that can't happen. The other thing I'll say, and I just recently, in the last, like, 48 hours, I just found out that this is at least slightly a thing. But the other thing I'll say is, they should give us access to hardware.

Now, in a perfect world, that means give us some hardware. Do a D2K type thing. But yes, your reaction is exactly correct. There's no way that's happening. Not a chance. In a perfect world, it'd be great. But it's not happening. No. So what's the next best thing? And Marco and I were talking about this a bit privately. But they are doing, I don't think they're officially called labs.

I forget what the official name is. Marco, do you remember? Like, I don't know. I don't even know. It doesn't matter. But they're basically doing labs. Like, the one that Marco and I did together for the Vision Pro a couple of years ago now. Three years ago, something like that. And what they are doing is in Cupertino, in New York, and in Battersea, and I think one in India.

And I think one in, like, Japan-ish area. I don't have the list in front of me. They're doing these things where you can come in for a day or sometimes two and presumably get time with a physical iPhone duo. And I really, really, really, really support this approach.

Genuinely. I love this approach. I wish there was more, particularly in America, than literally one in New York. And that is the only one the East Coast gets. And I'm sorry if you live anywhere between, I don't know, San Francisco and New York, because you're screwed.

You're going to have to find a way to get to either San Francisco or New York. I don't love that. I get that this is complicated. I get it's hard. But if Apple wants the particular, well, I was going to say any developers, but if Apple wants developers to have this thing, to have their apps ready to go on day one, you got to put in the work, Apple.

And yes, I think a lot of us will recompile and it'll be passable on day one. For all I know, that may be where I end up because I'm gone next week doing other things. But maybe it'll be passable for a lot of these apps on day one, but it won't be great.

And I really wish they had more offerings, even just in New York, because selfishly, the one and only time they offered , I can't make that day or two days happen. And so I really wish they took that more seriously. But ultimately, I still don't really understand why we couldn't have gotten Xcode 27.1 on day one.

But Marco, convince me I'm wrong. Nope, I'm pretty much with you. You nailed it. I'm not asking for Apple to give developers access to this kind of stuff at scale before they've announced it. And they don't.

Like Apple, they follow a very predictable pattern where when they're launching new hardware like this, they bring a very small number of select developers like Netflix, you know, or maybe a couple of game developers. Like they'll bring them in and show them the hardware before they've even announced it.

And that way they can have something to demo during the keynote. But that's a very that you're talking like, you know, you can count them on one hand usually, like how many developers would have access to that before the keynote. Then the keynote happens. The thing is announced to the public. And at that point, usually like an hour after the keynote, that same day, developers get an SDK that gives us a simulator for the new hardware.

And for all the new abilities, any new APIs that are enabled by the new hardware, we get an SDK that day. That's the part that didn't happen here. Now, I'm glad that they did give it to us. I mean, what was it, a week and a half later? Something like that. Something like that. That's great. But this is already like this is a very compressed timeline for developers because the duo has all these different screen sizes and aspect ratios and positions it can be in and changes so much about the layout of common system

controls and system interaction patterns that it's a lot of work. And we have one month before they're in customers' hands. So losing, I mean, at the time it was, you know, six weeks. Losing one and a half of those six weeks felt like a need less delay for developers.

Now, it wasn't arbitrary. I'm sure the answer was it wasn't ready yet. And in fact, when they did give us the SDK on Friday, there were some things about it that were kind of rough that it's like, oh, yeah, it isn't ready yet. Or at least like you can tell they really like pushed it out hard.

And even then, they couldn't get it out more than, you know , a week or less than a week and a half after the event. So it's not that they were artificially holding it back. But they obviously weren't prioritizing it enough to get it out there on day one.

It wasn't ready yet on day one was the problem. And look, this stuff happens. I understand. It's software. None of my stuff's ever ready when I want it to be ready. Preach. But I'm not the biggest company in the world. You know, like that's not my job to be that and to do that.

And, yes, and you are not relying on a bunch, like thousands, tens of thousands, hundreds of thousands of other people. To leverage your work product, right? Like it'd be one thing if Apple just, you know, wanted to be late for something that only affects Apple.

But Apple wants all of us developers to be ready on day one . And if you want us to be ready on day one, help us be ready on day one, right? Like it's very frustrating and demoralizing is dramatic, but I can't think of a better word.

It's just frustrating that we're begging for table scraps just so we can get a not embarrassing version of our apps ready for day one. Yeah, because here's the thing. On day one, we will have one-star reviews from people who say that our apps aren't ready or don't look good or don't work well or the touch targets aren't reachable or something.

We will get one-star reviews every single day from day one until we fix those problems. Every day counts. Now... But, Marco, there's no way that actually happens. I assure you what happens. And the thing is, every part of this is an Apple-created problem.

Why don't we have time with the hardware? Well, with the exception of some of these labs that fortunately are happening, I mean, we'll see how many people can actually get into them and everything. But, you know, that helps. But, like, developers don't have early access to hardware for the most part.

Okay. We don't have earlier access to software than, you know, four and a half weeks before the release. Okay. Why can people rate apps on a brand new phone that developers haven't had access to, but yet our apps are still available on that phone? Good question.

That's an Apple problem. Why do app ratings matter so much to developers? That's an Apple problem. Why is it a five-star system when no one ever does anything but one or five? That's an Apple problem. Like, every single one of these... Like, these are all Apple-created problems.

So, there are some things that Apple could do differently here. They could give us... They could prioritize the release of the SDK on day one every single time. They could make one-star reviews not possible on a brand new platform for the first month or two or whatever.

Or they could make them matter less. Or they could do a different system that isn't just one star or five stars. There are so many ways they could solve this problem. And we just don't question them because this has just been how it's... This is how it's been for 20 years almost.

Like, this is just how it's been. And so, we just assume this is the way it has to be. But every single part of this can be improved. Every part of this is an Apple-created situation. And it is within their control to improve it. Oscar didn't ask me to answer this question.

But the reason I put it in here is because you two are all angry about the way the whole Duo DevTool situation worked out. And I thought it was worthwhile thinking about, Okay, well, if you're mad about this, what would you rather have happen? What would have made you happy? And my answer to that is something that is difficult for all the reasons that you've mentioned, but is still the actual answer, which is whatever Apple is releasing,

whatever product they're releasing that they've held secret , because we assume they're going to continue to do that. And honestly, I think that's fine. It's fine to have something secret for a long time. But whatever your secret thing that you're revealing that requires developer support, Casey's point, like it's not just a thing that you put out there. You actually want developers to do something in response to it.

However big that changes, is it just like, oh, make the screen a little bit taller? Or is it like it has two screens and it bends? Like, however big that changes, you need to scale the amount of time developers have to update their apps based on how big a change you just made.

Which means the actual solution is not, oh, make sure you have the DevTools on day one, because that just frustrates people who are impatient like Casey. The real answer is... It wasn't just me, Dad. And Marco. The real answer is... I want credit for my impatience. The real answer is people should not be able to buy and have in their hands a duo for another whole month.

And Apple's like, well, we can't do that. The holiday season and this and iPhone time is always here and blah, blah, blah. And that gets back to, well, these are all choices that you made, Apple. I understand why it's difficult, but that's the actual solution. Not the day or week and a half or whatever. It's you have to give a longer time between the time that you announce the product and the time when consumers are allowed to buy it.

To give your developers more time if the change requires more work. If it's a small change, then give them a few weeks or a month. If it's a duo-sized change, I feel like two months would be more appropriate. And you're like, well, they can't give two months. They're going to miss the holiday season. The things are ready to manufacture. But like, I understand.

But that is the actual solution. You have to give developers time enough to make their changes. It doesn't actually matter whether you have the dev tools on day one and stuff like that. Whenever they have them, the only thing that matters is how long from the time developers get the dev tools to when they have to have their app updated for your new device.

And if Apple cares about that, they have to give developers a sufficient amount of time, which runs afoul of all the other things I said. And I will add, by the way, that the secrecy screws them here, too, because if you're not following the minutia of X code releases, which you're probably not if you're a developer, but like Apple walls off the people who are working on secret projects, even within Apple.

That's just compartmentalization. They've always done it. They continue to do it, which means the team that is working on the iPhone Duo dev tools, like the simulator with the bendy phone and all the new APIs or whatever, they are actually walled off from the rest of the Xcode team. A subset of them are disclosed on the bendy phone, right?

And a bunch of other ones aren't disclosed on the bendy phone. And so they're working on the version of Xcode that's going to have the dev tools that know about the iPhone Duo. And the other Xcode people are working on Xcode that doesn 't know about the Duo because they don't know about it. And so on day one, the main Xcode team is like, well, here 's the latest version of Xcode.

And then the iPhone Duo team comes out and says, surprise, we've been working on one with the bendy phone. And they're like, OK, well, we have the real Xcode that we 're shipping to consumers here. Now we have to like merge your changes. And they said, well, don't worry about it. We'll ship 27.1. And that's our version.

You just ship 27.0. We're going to ship 27.1. That's got the bendy phone in it, OK? But then they're like, but we're about to release another version of 27. It's like, OK, you call that one 27.2. And you can release that one, but 27.1 isn't ready yet. And so you end up with the 27.2 beta that comes out before 27.1 because the team that was working on it in secret hasn 't had time to merge their, you know what I mean?

Like this is this is part of the fallout of secrecy. And I'm not saying they can't do that, but they just have to add that to their timeline of like, all right, well, that just delays how long it's going to take before developers have tools to do this. And now how much time do you think it should take the average development team to adapt their app for this bendy phone with three screens?

And the amount of time they're giving, I feel like is maybe a little bit tight. And in the end, that's the root problem. It's not going to be solved by always making sure you have the SDK available on day one. It's solved by pushing out the date the consumers get the product. And that brings its own whole host of other problems.

So I got to move the whole iPhone date up out of September because now I need more time. Like this is the situation they're in. Like in the end, it'll, you know, Apple will again reap what they sow. They'll have fewer apps ready on day one and all the bad things that happen to developers with one star reviews will happen. And we'll all get through it together.

But yeah, the solution I feel like is to actually give adequate time for things to happen, which is one of the hardest problems in computer science. One of the hardest problems in management period is providing those timelines. Apple does a pretty good job of having timelines and keeping things on them. But there is some collateral damage and the duo and app readiness is going to be one of those collateral pieces of collateral damage.

Yep. Now that is annoyingly the rightest answer of the three of us, which I shouldn't be surprised, but you're exactly right. Photos libraries on external drives. Larry Feynman writes, I was so happy to hear you address this issue again on ATP. I was even happier to hear Jason Snell help solve this long standing problem regarding photos library on an external drive.

I spent countless hours on the phone with Apple geniuses trying to solve this to no avail. I really hope I'm wrong and that the new enhanced Siri is wrong too. Here's a summary of what macOS 27 Siri has to say on the matter. Quote, if you move your Apple photos library to an external hard drive, you will lose the ability to search for specific people, pets or objects.

This happens because modern macOS uses a system wide spotlight spotlight indexing for these advanced searches, which only works if the library is stored on your Mac's internal drive. Oh, swing and a miss new Siri. Sorry. I tested it on your latest OS and it works just fine inside the photos app.

If they mean you will lose the ability to search for specific people, pet or objects when you're searching through command F in the finder or through spotlight, then that is correct. But that's not what they said. So I would continue to caution people. They're like, well, surely if I ask Siri, it will have this knowledge base of Apple's product knowledge and it will give me the right answer.

That continues not to be the case. It will very often give you answers that lead you to settings that do not exist or do exist, but at a different path. And no, it's not because it has outdated knowledge because they move things around. Sometimes it just makes up stuff like, oh, go to this and that. And it's like, no, there's nothing.

There's no setting called that in system settings. There never has been and there never will be. In this case, it gives what I think is on its face an incorrect answer about how this works. Maybe the kernel of truth that is feeding this plausible LL M generated text is the fact that spotlight searching won't work.

And maybe it is also true that macOS uses the system-wide spotlight indexing system for these advanced searches. But the combined answer makes this person think that they won't be able to launch photos and search for cats. And you will. It does work. All right. Moving right along.

Preliminary M5 Ultra GPU Geekbench scores have come in. We didn't have these last episode. All right. So looking at the performance deltas, the increases through different generations. M1 to M2 is about 34% better. M2 to M3, 15% better.

M1 to M3 is a 53% change. And then M3 to M5 Ultra, about a 60% change. And the M1 to M3 comparison in there is because we skipped a generation. So if you're just looking at 1 to 2 and 2 to 3, you're like , oh, 34%, 15%.

But then look, M3 to M5 is 60%. Well, you're jumping. You know, there's a delta of 2 there. So the 1 to 3 comparison is the closest one. And that was a 53% gain. And this is 60%. So still, this is the best M whatever Ultra increase, two generation increase chump, I think.

Although I didn't actually do M1 to M3. But anyway, look at the chart. We'll put a link in the show notes. The M5 Ultra has good GPU gains. But a lot of that goodness has to do with the fact that there never was an M4 Ultra. So its comparison is to the M3 Ultra.

And in that comparison, it looks pretty good. All right, then let's talk readiness score. This is a new thing in the Apple Watch. I have slept with my watch on like three or four nights, but not enough to get. Hey, baby.

Hey, girl. But anyways, I haven't slept with it enough to get a readiness score because I think you need to have it on overnight seven times or something like that. But cardiologist Michael Katz pointed us to an article by physician scientist Eric Topol that says, Readiness scores on devices like the Apple Watch are proprietary scores that integrate different metrics for each of the wearables.

No algorithms have been disclosed. They are unvalidated against health outcomes. So there's a whole substack about that. Then back to Michael Katz. It is infuriating to hear about new sensors, the marketing and wellness in the absence of any meaningful evidence. Doctors are mad that this is just made up stuff.

But I mean, I do feel like it's mostly for entertainment value. And I'm sure Apple has a million disclaimers saying this is not real health advice, blah, blah, blah, blah, blah. But it is worth pointing out that a lot of these things these sensors do is just like this might be a fun thing to tell people. Setting aside like, you know, heart arrhythmia and, you know, dangerous conditions where your blood oxygen is too low or whatever.

Things like readiness scores and even the sleep tracking. It's they just take a bunch of numbers and grind them up and they give you some kind of score. And you should mostly just view that as like, well, whatever they're doing, I can move this number around. Whatever they're doing to calculate the number, presumably me moving the number around does something.

But the readiness scores and stuff, cardiologist is saying that's not really a thing. Just enjoy it for entertainment value, but don't put too much stake in it. Yeah. And this is I mean, this is honestly true of every device that offers this. You know, you have whoop bands and aura rings and Fitbits and all these different health wearable things that also do similar things.

And they all do stuff like this. They all have some kind of metrics or stats in their apps that try to tell you like, you know, you're this is your energy for the day. Even like my my Suunto watch and the Garmin watches like they all they all do this.

And they're all using their own algorithms with their own data that they don't tell you. So there's no way to like even judge it like, oh, this one is accurate and this one is not like there's this is not this is not a scientific endeavor. It is merely a consumer facing product with some fun numbers for you. Yeah.

And as always, these are, you know, as John was saying, like it's not really medical advice. It's something that can like maybe direct you to seek medical advice if it detects some kind of problem or detect something that might be a problem. But they're not medical devices.

Now, also. It's it's less about like, oh, the Apple watch gave me a seven and the Woot band gave me a six and more about that device measuring things relative to itself. And it's and your own experience with that device over the last X days or months.

Getting a baseline for each device and having it measure relative measurements to itself on you in the past. That is actually potentially useful for certain things, but you can't compare them or treat them as absolute value. So it's something with like when the Apple watch tells you the amount of calories you burn in a workout.

These are estimates based on stats and heuristics and things. It's not actually measuring how many calories your body is burning. It can't. But what it is measuring is some amount of heart rate or movement or some combination thereof.

And you can compare that today to what you did at your workout two days ago or last week or whatever. And you can say, oh, this was a harder workout than the last one or this was an easier workout like that relative to the same device measuring the same kinds of things a few days earlier.

That could be a helpful thing to give you a relative idea. These just aren't absolutes. And and you can't compare them between devices. But you don't usually need to. And the problem with even those measures is like, well, how are they calculating that number? Are they waiting something in there that I've massively changed a number that turns out to be heavily weighted in their algorithm?

So it's changed my score a lot. But in the end, I haven't actually changed the thing they 're trying to tell me about. Like if you read the this article by Eric Topol, he spent some time railing on heart rate variability and how that like these a lot of these things incorporate heart rate variability into the algorithms to try to give you some score.

And you're saying there is no evidence that heart rate variability should be a factor at all in this algorithm and could lead to misleading results because it's now weighted in to give you whatever your score is. But no one has no one has shown in a scientific study that measuring heart rate variability is a useful thing to do with respect to the thing they're trying to show.

So, yeah, I just you got to take it all with a big grain of salt. And even for things like, you know, the calorie burn one again, because it's got to give some kind of estimate because they don't tell you here are the numbers we're incorporating into the algorithm that derives the calorie burn thing.

You don't know if it's just like, well, you know, if it's some particular sensors like measuring your body heat or something, I don't think this is the case. But like and it happens to be like a hot day or you're running a fever or something and it skews your score. Like if you don't know what it's incorporated into the algorithm, first of all, you can't.

No, scientists can validate. Yes, this algorithm is giving you actionable data. And second, you don't know if even just the relative thing like you may be changing a variable that actually is unrelated to the thing it's trying to measure. But you change that variable a lot and it makes your score yo-yo up and down.

But in reality, you're not actually changing the number of calories burned because of the way they calculated. Anyway, it's it's tricky. Like, I do feel like the main the main purpose, the main value of things like this is simply getting people to take notice of some aspect of their health or activity and in a way that they weren't before.

Like, that's the gamification, like even just like closing the rings. That's just that's basically just a video game. It's like a Tamagotchi. Like, you got to close your rings. What what is closing the rings do? I don't know. I just like them to be closed. In the end, it is making you be slightly more physically active than you were before. And there is a foundation for being slightly more physically active than you were before is good.

And that's the level these things are working at. It's like herd you towards what we hope are outcomes that are scientifically proven, as in doing more physical activity instead of less is better. If you're doing very little activity, adding a little bit more can help your health. All right.

Let's get to topics. And let's start with, as we talked about earlier, the iPhone Duo DevTools are released. John, you took us through a kind of overview as to all the different versions of Xcode that are currently out right now. But yes, 27.1 landed, I believe, Friday as we record this.

And it's interesting. I think Marco had said earlier, or one of you had said, that it's a very stripped down version of Xcode. And the device hub simulator, whatever you want to call it, for the Duo only includes Safari and nothing else.

Or maybe settings, maybe. Yeah, it's definitely, it's coming in hot. But they did add a 3D model of the phone, which is the thing I suggested last episode. Like, we were talking about iPhone mirroring. I'm like, how are they going to do the phone and iPhone mirroring? Oh, they should have to do a 3D model that you can fold. Well, they do have a 3D model. I think this is from Device Hub. Iron Vey posted a video of it.

And apparently, if you hold on the option key, you can control the folding state. And it even does the effect or whatever. Just follow the link to Iron Vey's post to see the pictures of it. But I put some things in the notes here. If you look at the bottom one with the thing closing, notice how the screen is disconnecting from the frame a little bit.

So, yeah, coming in hot, I feel like. For people who aren't looking at the exact image, and this is from a video. Like, they show a 3D model of the phone folding. But it's clear that the screen portion of the 3D model is not connected to the frame of the phone portion.

Because the screen is folding faster than the frame portion , leaving gaps between them. A little bit of jank. But it does look pretty cool. Yeah, I mean, I've been using this a fair bit. What I have discovered, and I don't want to go on too long about this. Because I don't think that's the purpose here.

But what I've discovered is I saved myself. Past Casey is both wonderful and an apple. And the reason I say that is because just a few months ago, I really took a hard look at my iPad design.

And really revamped it and made it no longer actively embarrassing. I wouldn't say it's perfect, but it's efficient. And when you open up the internal screen, I can never remember what the official names for these are. But anyway, if you open up the internal screen, it shows, for Callsheet anyway, it shows the iPad version of it.

And it looks mostly good. It's a little cramped compared to an iPad because, of course, it is. But mostly pretty good. Mostly okay with it. And on the outer screen, I have some fiddling to do with, like, toolbars and stuff. But by and large, it mostly looks like the iPhone version.

Mostly pretty good. However, where Past Casey is a jerk is that I kept kicking the can around some of the... I'll say esoteric, and someone will probably disagree with me here. But some of the esoteric stuff around iPad. For example, I never really use windowing in iPad.

I'm genuinely incredibly glad it exists, but I almost never use it. I almost never use Stage Manager or anything like that. And so having multiple instances of Callsheet running concurrently was never really a thing I did. And thus, I never really made it work. It does work out of the box the way the released version of Callsheet is on the Duo simulator.

However, if you navigate in one of the two side-by-side Callsheets that you have running concurrently, if you navigate on the left-hand side, the right-hand side does it too. And so I need to properly support... Shoot, I forget the... What is it?

Scenes. Yeah. Rather than just lumping everything into, like, one macro window. I would not call that working, by the way, for two side-by- side instances. I would say that is in the realm of not working. Well, sure. I just mean it's not crashing or anything. Yeah. I would also say, like, you might be the first person to ever actually try to run your app this way.

Like, this is not something that a lot of people are doing. And, you know, you might do it in Safari, oftentimes accidentally, because you tear a tab off the window or something when you're trying to... It's just... This is not a style of multitasking that people frequently do intentionally on iPads for most apps.

So I think while John is technically correct that this is not, like, the right behavior and it's not acceptable behavior... It would be like if you had two Safari instances and wherever you wanted one Safari window, the other window went there, too. It's like, well, what's the point of this? Yes, exactly that.

But I would also say, like, fixing that should not take priority over almost any other, you know, need for your app for your customers. I agree, but what it boils down to is at this point, I need ... I feel like I need to make that work, at least for the most part.

And as I sit here tonight, it's mostly there. It's not 100% there, but it's mostly there. I think as long as I have another couple of days to work on it, I'll get it there. But with regard to most of the other stuff, I actually am sitting far prettier than I expected. And I think that's because, by and large, I am using out-of -the-box controls.

Not 100%, but for the most part. And that means that this is not the absolutely ruinous experience that I worried it was going to be. Now, I think my kind of minimum viable product is just, does it work appropriately or reasonably in, you know, the modes that the Duo supports, of which there's like 312 of them?

Does it actively embarrass me or not? And right now, it does actively embarrass me, but I think I 'm pretty close. I'm within striking distance of not active embarrassment. Do you have any plans for tent mode? Not really. I haven't thought too much about that because the lower-st akes stuff is front of mind at the moment.

If you or listeners have suggestions, I'm all ears. Here's my suggestion. Because I keep thinking of Apple's demo of like, yeah, there's tent mode, but then like you put it on like your nightstand and it shows like a nice like clock or whatever, like an alarm thing. If you put call sheet in tent mode as like a little night stand thing, it should have like a little screensaver of floating heads of like the last 10 people you looked up.

You know, that is one way you could do it. Another thing I am considering is if you're actively watching a movie. Now, maybe that's because you've told call sheet you're watching a movie or show or whatever. And call sheet has for a long time supported live activities where you can go into call sheet and say, start a live activity.

And it will put something in the dynamic island so you can always get to it lickety-split. Even better, if you're integrating with Plex or Jellyfin or Channels, I believe has duration as well. If I know the duration, then in tent mode, I could show like the poster art in a status bar.

Not a back and forth status bar, mind you, John, but an actual status bar or a progress meter. Excuse me, I keep saying status bar. Progress meter. Showing how far you are through the movie or show or what have you. Like I could do something like that. I haven't thought too much about that. Like I said, I want MVP for now.

And I think I'm going to get there before launch day. And then inevitably, and I have a feeling Marco is going to say the same. Once we get the hardware or perhaps spend time with the hardware, one or both of us can make it to New York, we'll see that all of our assumptions are wrong and we're going to need to rejigger everything.

But I just want it to be not actively embarrassing. And I'm within striking distance. And while this simulator is not perfect, it has been sufficient enough for me to get my work done. As opposed to like, I think it was Greg Pierce was fussing on the internet, which is not his style, about how rough it 's been for him.

So I don't know what he's hitting that I'm not. But either way, it's been fine for me. And I'm glad it's here. I'm glad it was only a week and a half late and not later. And I am trying furiously to get work done as quickly as I possibly can. Marco, what do you got going on?

Basically that. I had a busy family weekend. So I didn't actually, I wasn't expecting it to drop on a Friday afternoon. So I thought I'd have another week. So I've only just started like over the last couple of days . But so far, it's fine.

I am not doing anything very like deluxe for this adaptation until I have the hardware. I'm doing the like a normal job. Make sure controls look decent. Get stuff away from the edges whenever it's possible without causing a lot more bugs.

That's what I'm doing. Any I'm not doing any kind of redesigns yet, like just a basic adaptation. So my apps don't look broken. And for now, that's it. And then once I actually have the hardware and start to use it, and by the way, start to see how it sells, then I will start evaluating whether to do deeper optimization for it.

But I think it's probably going to sell very well, as I've said before. Even though very well for, you know, a kind of a specialty version of the iPhone, that's only going to maybe be three to 10% of iPhones that are sold.

I'm trying to avoid the Vision Pro scenario, where the Vision Pro was just a massive distraction for all of us for months in various ways. I mean, for Apple, it was a massive distraction for years.

But it ended up that, you know, any work I did trying to make Overcast work on the Vision Pro was a waste of time. I hate to say it so bluntly, but it was a waste of time. It was a huge distraction. And I wish I would have spent that time working on the app for everybody else, all my other customers, instead of the literally like 35 that ended up using the Vision Pro.

And I don't think, I'm not saying the Duo is going to do that. It's not. But I don't know how much of my time the Duo is worth. It's not worth zero. The Vision Pro is worth zero. The Duo, I'm sure, will be worth more than zero. I'm sure it will sell, and people will like it, and it will probably stick around as a form factor indefinitely into the future.

So, it is worth optimizing my app for the Duo. But is it worth an entire month? Maybe, hopefully not. Is it worth multiple months? Almost certainly not. So, try to do something like, you know, make like a tent version of Overcast.

No one's going to do that. Like what I was saying with Casey with your multi-window on your iPad. No one's going to do that with our apps. Or close enough to no one that we shouldn't spend almost any of our time doing stuff like that when there are needs that all of our iPhone customers can benefit from when we work on the regular rectangle version of the app.

You know? Like, whatever we're going to call, like the regular iPhone screen. Like, all the work that we do on the Duo has to be put in context of this is going to benefit less than 10% of our iPhone customers and possibly a lot less than 10% of our iPhone customers for a while and maybe forever.

So, I'm going to accommodate the Duo, but I'm not really going to focus my design efforts on the Duo. Yeah, the left and right safe area insets are going to benefit probably a larger number of people when the edge-to-edge screen iPhone 20th anniversary phone comes out next year.

Because presumably a lot more people will buy that than bought the Duo. And so, just basically having your apps understand non-zero values left and right will go a long way to getting ready for that a bit. Yeah, definitely. All right. Thank you to our sponsors this episode, Daft Music, Click for Sonos, and Quince.

And thank you to our members who support us directly. You can join us at atp.fm slash join. One of the many perks of membership is ATP Overtime, our weekly bonus topic. This week in Overtime, we're going to be talking about a new Ubiquity home Wi-Fi product that does full duplex MLO.

This is some pretty advanced stuff. Covering that in Overtime, you can join and listen at atp.f m slash join. Thanks, everybody. And we'll talk to you next week. Now the show is over. They didn't even mean to begin.

Because it was accidental. Accidental. Oh, it was accidental. Accidental. John didn't do any research. Marco and Casey wouldn't let him. Because it was accidental. Accidental. It was accidental.

Accidental. And you can find the show notes at atp.fm. And if you're into Mastodon, you can follow them at C-A-S-E-Y-L-I-S-S.

So that's Casey Liss. M-A-R-C-O-A-R-M-N-T. Marco Arment. S-I-R-A-C-U-S-A-C-R-A-C-U-S-A-C-R-A-C-U-S. It's accidental. Accidental. They didn't mean to.

Accidental. Accidental. So I run a bunch of Mac minis. You do. When I first launched the Overcast transcription, when I first announced, rather, the Overcast transcription cluster in the data center, my 48 Mac minis, I had mentioned that the way I'd set these up was literally just setting up 48 Mac minis , like manually.

I enabled Remote Desktop to eventually, you know, kind of govern them all. But the setup was literally just clicking through the setup screens 48 times. In some of them, like installing the betas as they came out for the first summer, like over and over again manually, like it was ridiculous.

Yeah, we had some follow up about like MDM and stuff like that. Yeah, exactly. Like people, like, of course, all of the like Mac admins out there who run fleets of Macs, they're like, oh my God, you're not running an MDM? This is crazy. What has also happened, and now one of the reasons that I hadn't been running an MDM service before is that the cheapest ones I could find were like seven or eight bucks per month per device.

So this is, this would be a substantial amount of money to manage my fleet of Mac minis. And when you looked at like the features they offered, it's more, it was more geared at like managing employee devices for like a whole team of employees at a business. Like that's not, they weren't really solving a lot of needs I had, and it was going to cost a lot of money.

So I just did it all without that. In the meantime, this past spring, Apple launched a new Apple business platform that included kind of a basic free MDM service right from Apple. So I signed up for that. And so I wanted to give an update on where the Mac minis stand and, and my experiences using the Apple business MDM.

And this is being my first MDM experience ever. So first of all, a small update on my Ubiquiti high availability setup. I mentioned briefly a few episodes back that I had gotten a second router and was running it in Ubiquiti's high availability shadow mode where you basically have like a hot spare router effectively.

Um, and I was splitting the internet connection between the two of them with the Ubiquiti WAN switch. That doesn't work. Oh, cool. My Ubiquiti WAN switch just doesn't work reliably. Like it works for a while and then doesn't.

And I don't know what the problem is. So I'm currently investigating like alternatives. Like apparently you can just use any switch if you configure it right to like isolate the, you know, the level, whatever, and the subnets and VLANs and stuff. So I have to figure all that stuff out.

Um, but not pleased with the WAN switch so far, but anyway, that's, and by the way, try to find any documentation on the Ubiquiti WAN switch on the internet. I, I challenge you. There is none. And as far as I can tell, I'm the only person who ever bought one, not recommended.

Anyway. All right. A few months back, there was a, uh, remote desktop security vulnerability that was patched, um, in Apple remote desktop. And this vulnerability allowed access like to any, anybody with the public IP address. If the machine was open to the public internet, anybody could perform this exploit and whatever it was and gain access to the machine and install whatever they want and run it.

I had Mac minis at a hosting company that had public IP addresses. Not many. I had four of them there. Oh no. And right as all of this was happening, I'm like, you know, I should probably take those back in house, put them with the rest of the data center of Mac minis.

Um, I should probably just get those out of there. And that way I can centrally manage them all because having these on these public IP addresses was always a bit of a pain because it was like out of my management system. Like if I had to like reboot them or log in like with a KVM or screen share for them, it was always a little bit clunky or would require me to go through their UI or submit like a support ticket or something.

And like, it was a big pain. So I'm like, I'm just gonna bring these in house. So I, I ended things with them. And I started to realize like, you know, when that exploit came out, I'm like, I need to patch all these machines. I need to make sure they're all are running at least 26.2. I believe it was. How do I do that with 48 machines?

Do I actually have to go on 48 machines and go and click, click, click and type in the password and do all like the software update steps 48 times. So I'm like, all right, I'm going to get these all on MDMs. I got Apple business going. This is not an easy system to understand.

I can't say I find that surprising. Yeah. Fortunately, we have AI now. Back then it was not as good and not as available and didn 't know as much. Now it's better. So a lot of my questions I was able to find better answers to. And a lot of them I was able to have AI like generate me config profiles or generate me shell scripts to run that would fix problems or that would set things up a little bit better.

But it is incredibly complicated and incredibly clunky. There's two different ways you can enroll a Mac in MDM management. One is like if you physically have access to it and you're like restoring it from scratch or setting it up from scratch. You can use the iPhone configurator app to like set up a special mode where like your profile kind of gets locked to it.

Or if it's already set up and you don't want to like restore or you don't have physical access like in a data center, you can just like from the Mac OS control panel, you can like add like log in that way and add the configuration that way. And it's like there's some difference between those two modes.

For my purposes, they don't matter that much. But anyway, you can do it. But it is incredibly cumbersome and time consuming because there are so many steps that need to be done individually and can't be automated or skipped. And there's so many manually typed logins and passwords and so many 2FA codes that have to be sent to my phone and then typed in through a physical

keyboard or a screen share or something. I cannot tell you how much I hate that of the like three different Apple accounts that are required to do this kind of setup, they all require 2FA . And there's no as far as I can tell, there's no way to have them not have 2FA.

Just be glad those Apple accounts weren't getting locked every 24 to 12 hours. That's fair. Although I will say people probably write in with this. If you had done this in the other order and gotten the Apple business thing before you bought the Mac minis, isn't there a way that you could buy the Mac minis from Apple so they come out of the box like pre-enrolled or like prim ed to enroll themselves?

Yes. And I can't, I didn't do that when I bought them. You can do that when you... Time travel, but... Yeah. If you buy new ones like through your business reps, they can auto-enroll them so they get so they auto-configure as soon as you plug them into the internet. But the process that I do, like the ones that I've restored from in-person use, I've currently done I think about 12 of them this way.

The ones when you restore it and use configurator on your phone, like during setup to like do that, it ends up with the same result as that. It just takes more steps. But oh my God, there's so many steps. And if you get anything slightly wrong, restore it, start over.

That sounds familiar from earlier this episode. Yep. It's so cumbersome. And I can tell you all these stupid logins and passwords and 2FA codes. And the other thing is like almost every step that or every few steps when you're setting up a new Mac shows a spinner and you just have to wait.

Why can't we batch some of these things so that you can ask me all the questions you need to ask me and then do all the slow stuff as one batch? Like I know that's not always possible, but it seems like it's not even considered at this point. There's also like the onboarding slash setup screens in mac OS.

There are so many times when you click a button and the button does not change state, but it is waiting. So the button does not get disabled or it doesn't become a spinner. The button pops back up and is still enabled.

And there might even be a spinner on the other side of the window, but you can still click the button again. So it's unclear whether are you should I be waiting right now? Like John was saying earlier, like there's so many cases like you need to be waiting, but the interface does not fully support that mental model because some button is enabled again or it's silently spinning somewhere else, but not displaying a spinner to you.

It's so incredibly sloppy. And this is the first experience for new Apple customers. Like when they get a new Mac, it needs a lot more attention from Apple. Honestly, it's very sloppy and it's full of rudimentary like, you know, level 101 interface bugs and shortcomings that Apple is above that in every other way.

And yet here we are like the very first experience you get when setting up Mac OS is just amateur hour. But anyway, so many long spinners, so much waiting for Apple or the system to do something that's unclear. Like why does agreeing to the terms and conditions have a spinner that sometimes lasts 15 or 20 seconds?

I have no idea. Why does creating a user account take so long? I don't know. There's probably a reason, but why? I don't know. It's a long time. Why does activating the Mac take so long? And then after it's activated, why does it take a full minute to reboot? It can't do anything else.

Why not reboot immediately? Like there's so many of these steps and every single one. It's just like, why am I waiting right now? And maybe I'm sure there's a good reason for some of them, but like it's so slow. It's so cumbersome. And then you just have to like wait for a while, stare at it, and then like click another button and then wait some more because they didn't batch those two things.

Also, MDM has the capability to skip certain setup screens, which is great. I love that. That's a very common need. If you're setting up or, you know, re-imaging or whatever, a whole bunch of Macs, a lot of those setup screens you either don't want to face or you don't want your users to face, or you've already set a policy in place that makes that decision for you.

So for instance, on these Macs, I don't use FileVault because it's a lot easier not to in various ways. So I just don't use FileVault. There's no sensitive data on them. There's no customer data on them. There's no personal data on them. The only thing on them is a binary of my transcriber app.

And there's nothing sensitive in it. There's no credentials. There's like, there's nothing in it. Every single time there's a software update. Why do I have to keep saying no thanks to FileVault? It's like, because MDM, at least this version of MDM, can't skip it. Every version of Mac OS, every even point updates, when these machines install them and reboot, I have to say something like, you know, adjust the liquid glass slider or choose my age range.

I'm an adult. There's always something. There's a welcome to Apple intelligence. Welcome to the new image playgrounds. Do I want to use the new image playground training feature? No, I don't want to be presented with. Any of these things. Aren't there defaults for admins?

And the answer to some of them is yes, but not all of them. Because it seems like Apple, not only, I've played about this before, not only does Apple not respect users at all during that setup process by showing us so much crap over and over again, and there's so many steps. You can, like, this is like, you know, when people say, like, the org chart is showing, you can see this with whatever decision makers at Apple are in charge of, like, keeping the prompts to the user during

setup to a minimum, they don't have enough power in the company. Like, they are being overruled or something. Like, it's awful. There's so many prompts with everything. Every single version of something new is more prompts. And Apple is adding them faster than the MDM team can give you ways to skip them.

So anyway, that sucks. So many unskippable screens. They have to click through and click through and click through. Macs are so not made for this. And Apple never lets you forget it. They are not made for automated management.

They are not made for bulk deployment. They're not made for bulk management. And it is so incredibly cumbersome. But once you get through all of that, once you have MDM configured, enrolled, all the things are deployed properly, and all that stuff is done, then you can have automated software updates.

Which, when you're running a bunch of servers, is pretty useful. Or so I thought. Oh, no. So, this morning, I was riding the train from the city. I opened up my laptop for the first time in the morning, and I'm checking my transcript progress from all my other vital stats that are not super urgent.

And I see all of my transcript machines are down. Oh, gosh. Hmm. They haven't reported anything since 2 in the morning. Well, that's suspicious. That sounds like a software update time. Last night, all of my transcription machines, at the same time, upgraded to Mac OS 27 Golden Gate.

Oh, great. That's stupendous. Now, I probably shouldn't have allowed that. Like, I admit that was on me. Okay. I probably should configure these so that, first of all, it doesn't do major updates automatically. I should probably also configure these so that it doesn't do them all at the same time.

Maybe do them in batches. Maybe have, like, some test ones that I do first. Like, I admit, I should probably, you know, this is MDM newbie having newbie problems. I fully admit that. So, why did the Mac OS Golden Gate upgrade take all of them down?

Because every single one of those machines was waiting on the screen that says, software update complete. Your Mac has been updated to Mac OS 27 Golden Gate. Oh, my gosh. Continue. Cool.

These are so not made for this. Like, again, like, when I say that Macs are so not designed for this and actively fight this kind of use, this is the kind of crap I'm talking about. Every single machine was stuck with this. Now, also, whatever state they're in to be stuck on this, my background tail scale demon wasn't running either.

So, I couldn't reach them by any of their tail scale IPs, which means I couldn't reach most of them. This is why you need a KVM in your life. Well, funny that you mentioned that. Oh. A friend of ours at GLINet recently sent me a couple of their remote KVM products.

The Comet Q and the new PoE one. I forget its exact name. And it just so happened that last week, I installed both of them in the data center on two different Mac minis. And so, that is how I was able to actually start fixing this.

Because I was able to on the train over tethering log into the GLINet Comet Q interface. So, what this is, it's basically, it's a KVM device. It plugs into the USB-C slash Thunderbolt port on the back of a Mac mini.

And it fully emulates to the Mac mini a screen, keyboard, a mouse. But it projects it to you over a web browser. and it has built-in Tailscale client support. So, the Comet Q KVM shows up as a Tailscale host name.

So, as long as I'm logged into Tailscale on any of my devices or whatever, I can access its IP address in a web browser, log in directly, and have KVM from anywhere to that machine. So, that worked because that's not remote desktop to the Mac.

It shows up as hardware devices. So, I was able to log into that, proceed through the software update complete screen on that Mac, get through like the set the liquid glass slider, like all this BS that I do not care and I wish I could skip.

Another welcome! You know, the little animation of the little tubular glass, like, oh my god. But, so I get through that and then I enable and once I'm through that software update complete screen, then Tailscale runs again so then I can use regular remote desktop after that, which is a little bit faster.

So then I close that, I go to remote desktop to that machine, then I remote desktop from that machine. Yep, I was gonna say, I see where this is going. Yep. If I had seen the movie Inception, I would probably make an Inception joke here. But like, so through remote desktop to the machine that I had unlocked, I then made a sub remote desktop from that one to the other machines on its local network.

Can you go from that machine back to the original machine again? Probably, yeah. I didn't try that. And I'm doing all this on a moving train over tethering. I saw that in that AT&T commercial in the 80s. It felt, it felt glorious because the truth is it was fine.

Like, yeah, you know, a couple of cursor jumps here and there, like, you know, a couple of minor moments of lag, but I was on a moving train over cellular doing remote desktop to remote desktop. Like, come on. Like, that was amazing. Like, it actually worked. So, I had to log in to each one of all of the Mac minis and proceed it past the software update complete screen so then I could start administering it again and it could start running its software again.

You know, speaking of your AI buddies, one thing you might want to consider doing in this scenario is use computer control on a spare home Mac mini and tell it to do 48 machines worth of clicking. Oh, yeah, that would be wise. I'm not sure how well it does with screen reading that because like, it obviously won't be able to use like its accessibility views of the stuff.

But I don't know if it will work. I'm just saying like, when I think of you manually doing anything on these 48 machines that are enrolled in MDM it makes me think, well, if they're already not working and there's nothing and they're all disposable, they're all, they're all, you know, cattle's not, not pets. Yeah, right.

Throw, throw a computer control Mac mini at them and say, yeah, go finish all these software updates and if it hoses all the machines, how much worse off are you? I don't know. Worth considering. Yeah, I mean, at this point, so I'll tell you, I actually have never used any of those features of the AI apps yet. I'm trying to keep that kind of thing a little bit contained and so, same, on my to-do list is to designate one of these Mac minis as my agent machine and give it like, you know, only the credentials it needs

and only the access to the things it needs and have it be able to do that kind of stuff. Like, so this is like an ideal scenario. Like I said, there's no data, there's no personal anything on any of these machines and to the degree that you can isolate them, you can like, oh, you know, go nuts. Like the worst you can do is hose the Mac mini I've given you access to, right? Exactly.

And, you know, and it's funny, like everyone else is like, man, I wish I could have a Mac mini to run AI stuff and I'm like, I have some. but what you needed to do is to, you need the AI to manage the other Mac minis you have. That's what you need to do. Like, this is what MDM is supposed to do.

This makes me think like, um, maybe in an upcoming episode this will come up in my long queue of stuff to add in, but there have been some recent rumors about Apple potentially getting back into the server game. and if you're wondering like, well, who cares about Apple servers? Why would anyone bother with that? It's like, well, if Apple's hardware turns out to be good for various tasks for various reasons, like transcribing audio or bang for the buck for getting M4 Mac minis or whatever, what the server of hardware would bring is presumably that hardware would actually be tailored to,

uh, the things that these Mac minis are not, which is, hey, can I run a software update on X serve and not have to click through a, a welcome screen? I'm assuming the answer would be yes. Well, the thing is like, if Apple does get into the, the server game like that, are they really selling to us or are they selling to like, you know, Google and AWS?

Like I, it doesn't, we'll see like, if that rumor ends up being a thing. It won't sell to anyone who will buy them if the X server is any less than. Right, but like, basically, are they running Mac OS or are they going to run like the private cloud compute OS? Mac OS 10 server. Yeah, right. I know, what was the name of the OS?

It was like they had, it was some name for their server OS, but it was like some iOS based thing. I forget what it was, but someone told us at some point, but yeah, like this is all just rumors, but just all I'm pointing out is that the, it's like you said, the Mac minis, Mac minis are not made for this. You can use them for it, but they're not made for it.

and I think you're totally right about the team that makes the OS is always going to be outrunning the MDM team's ability to skip those things because yeah, like just think of every new version of Mac OS has some new BS in there. They remove old BS sometimes, but there's always something new and do you think the MDM team gets a heads up so they have time to update all the MDM stuff so it knows how to skip all the new screens?

Hell no. They probably don't even see those screens until two days before the keynote because they weren't disclosed on whatever features they're advertising. if they even see it before the keynote. It's very obvious that all of the enterprise and business and IT admin support stuff within Apple is not a high priority.

They don't get the resources they need. They don't get told about stuff ahead of time enough. That's very clear and I think that's been true for a while. I hope that as Apple figures out its role in the AI age, I think what we're already seeing is there is a role for people to have computers that they can carefully designate as AI running computers and give it only stuff it needs and everything like that.

Obviously, this is going to stay mostly a nerds and businesses thing. It's not going to really be a thing. Consumers will just install the meta thing and give it access to everything. Consumers will have the home hub screen device or whatever where it will be the same type of thing, a tiny home device running AI answering your questions but not anything like a server.

When it comes to running the chat GPT or meta or clawed apps to automate your computer doing things, most people are just not going to care about security and access and are just going to give it access to everything. But nerds and businesses care.

there is a market. As we see right now, that's why Mac meetings were so hard to get last year because everybody wanted to run OpenClaw and do exactly this. Give them an always-on connection so you don't have to worry.

It's not like running on your laptop and you close it and you can't do stuff. But also be able to isolate it to only some of your data, only some of your credentials, things like that. Apple can lean into this. Apple's role in AI so far is they make really good chips that run AI really well and they have a really capable OS in macOS that all of the AI apps launch great features like day one on macOS in their macOS apps.

And when it comes to like I want an AI to automate tasks that involve different apps on my system, possibly through MCP, possibly through accessibility APIs or like scraping or whatever else. And I want to be able to have a web browser that's also being directed by the AI and all this stuff.

The fact is Apple's other platforms that are iOS based are too locked down for any AI to have access to that except Apple. See also the EU BS. But anyway, not for today. The Mac is the one platform that they can all use and that people actually want to use.

Like obviously Windows and Linux can do this too, but nobody wants to do that people are actually wanting to run.

And Apple could embrace this. And one way they could embrace this is by making some kind of like mode for the OS. Maybe it requires MDM. I hope not. Maybe it's just like a mode. It could be like developer mode. You got to like, you know, wave some wands near your Mac and maybe know where the setting is then.

But some kind of mode that macOS can be in that you don't have all of these stupid prompts. You don't have all these, you know, like after it reboots, after a software update, you have to click this button before any of your software runs. Like have a mode to run like agent mode for macOS or something, automated mode, whatever they want to call it.

Call it agent mode. Take advantage of the AI hype. Agent mode for macOS. That would be a great feature for macOS 28. Agent mode would skip all this crap. It would, like so much of the stuff that my setup script does for me now, like removing all of the icons from the dock besides the apps I actually use.

Having like, you know, a launch D configuration that makes sure this one app is always running. And if it ever quits, restart it. Like there's so much stuff like that that you can do now manually, but would be so much better if Apple just gave it some thought and some time.

And what a great OS 28 feature that would be. So maybe think about it, Apple. Make all of these stupid hacks I'm doing unnecessary. Make it so that other people can do this. And all the IT admins, save them so much time. Because if they don't do this, running Macs as servers just sucks.

I'm doing it because the hardware and software, like the applications I'm running and the hardware I'm running them on are great. But Mac OS is just fighting me at every turn trying to run this not like somebody setting up a new iPhone for the first time.

Like please let me use my computer like a computer and optimize it for that a little bit please.