Is Apple Silicon Ready? (thenextweb.com) 132
Programmer Abdullah Diaa has put together a website to help determine if your favorite apps work on Apple Silicon yet. An anonymous reader shares a report from The Next Web: ... [P]lease say hello to Is Apple silicon ready? The idea behind the site is simple: it shows you if specific apps will work on laptops and desktops with Apple's in-house chip. Easy to get your head around, right? It shows you a list of software and, if they have native M1 support, they're given a green tick.
Here's an image that shows you what's going on far clearer than lots of words could. As you can see, the site also shows you if the app you're after has Rosetta 2 support. Effectively, Rosetta 2 is an emulator, allowing a large number of apps designed for Intel machines to run on Apple Silicon. If this is supported, you will still be able to use that software on an M1-toting machine. Further reading: Linus Torvalds Would Like To Use An M1 Mac For Linux, But...
Here's an image that shows you what's going on far clearer than lots of words could. As you can see, the site also shows you if the app you're after has Rosetta 2 support. Effectively, Rosetta 2 is an emulator, allowing a large number of apps designed for Intel machines to run on Apple Silicon. If this is supported, you will still be able to use that software on an M1-toting machine. Further reading: Linus Torvalds Would Like To Use An M1 Mac For Linux, But...
pornhub (Score:2)
Re: (Score:1)
Rosetta does the natives...
I don't see the point (Score:2)
Other than potentially aiming to do "app shaming" because they're "not ready", but then, that's why Apple implemented Rosetta-2, for those developers that might not have the same level of 1000+ engineering resources Apple has to adapt their code for compatibility, and need more time. And then they'll announce when they're ready.
Besides, there's several apps I've seen publicly state they're ready for Apple Silicon, but aren't listed here. Are developers supposed to "magically" know this list even exist, and
Re: (Score:2)
1. Rosetta2 is a translator, not an emulator, but it still doesn't provide full native performance. It is better for apps to be natively compiled from source.
2. Not everything works with Rosetta2.
Re: (Score:2)
Stop nitpicking about "translator" versus "emulator" - you are not qualified to talk about it. For all "computer science purpose or talk: it is an emulator. An emulator that uses translation. Oops, that is so easy.
Secondly, the "emulated intel binary" runs faster than the original binary on older Mac hardware, so much to "not full native performance".
It is better for apps to be natively compiled from source.
That is obviously a no brainer. However the gains might be minuscule.
Re: (Score:2)
Secondly, the "emulated intel binary" runs faster than the original binary on older Mac hardware, so much to "not full native performance".
Everything runs faster on new hardware than old hardware.
When I said "native performance", I meant the performance on the computer you just bought, not the computer you are replacing. I apologize for assuming that was so obvious that it didn't need to be explained.
Re:I don't see the point (Score:5, Informative)
When I said "native performance", I meant the performance on the computer you just bought, not the computer you are replacing. I apologize for assuming that was so obvious that it didn't need to be explained.
Apple replaced three low-end machines. In every case, Intel code compiled with Rosetta 2 runs twice as fast as the same code ran on the machine that was replaced.
Re: (Score:2)
What will happen when the power users try to upgrade their higher end macs? Will the startup sound be crickets?
Apple announced there will be a transition phase until the end of 2022. I bet they are working hard right now and the next chip with 8+4 cores, 16 CPU cores, twice the RAM, twice the ports, to replace their mid-end chips. Until that is ready, "power" users (that's an expression that always makes me laugh) will be told that they have to wait.
Re: (Score:2)
Hmmm I guess that proves Rosetta isn't an emulator, right? Condescension fixes everything.
Re: (Score:2)
An emulator treats the source instruction set like bytecodes and executes them as it steps through the code.
A translator reads the entire program, converts it to native code, and then runs it.
A translator will run faster but with a lag at startup.
The difference is similar to the difference between an interpreter and a compiler. They serve the same purpose but are not the same thing.
Re: (Score:2)
The point is that emulation as well as interpretation can - and often does - translate the original binary code into another code.
That does not change the fact that e.g. the Java Virtual Machine is an interpreter ... and ... Tata! Rosetta 2 is an emulator.
The fact that both use cross-translation does not change that. I suggest to go to a CS class and stop this silly nitpicking about it.
Re: (Score:2)
Rosetta is an emulator, it emulates a different computing platform. It performs codes translation as part of this emulation. translation and emulation are not mutually exclusive.
One can argue, even, that translator and emulator are synonyms in this context. It's ridiculous how people contort themselves to argue against "it's an emulator" merely because Apple marketers said this in a cheerleading session. Rosetta is an emulator.
Re: I don't see the point (Score:2)
Not true. Rosetta 2 is actually both. It uses translation when it can and uses emulation when translation isn't possible. (For ex. With every Electron app thatâ(TM)s using JIT to compile JavaScript at runtime.)
Re: (Score:2)
Re: I don't see the point (Score:4, Informative)
I don't think it's about "shaming" at all. In fact it looks quite useful and time saving. If you actually look at it, you will also see that some apps aren't working with Rosetta 2. It looks like a useful place to quickly assess the impact of switching to Apple Silicon, although there are still some missing for me like Cisco AnyConnect, but I think that will improve.
Re: (Score:2)
What if they don't know - will they end up getting "shamed" for it?
No, because most Apple users don't give a shit about what some obscure blogger has to say about which apps are ready or not, they'll simply install it from the App Store and they'll have their answer immediately.
Isn't it (Score:1)
Re: (Score:3)
Re: (Score:1)
Re: (Score:2)
modern silicon fab is a lot more complicated than just building chips based on design from customer.
But, the design is what determines what the product is and does. I'm not saying the fab isn't involved in the process. I'm just saying that ultimately the fab creates what it's paid to create. It's a factor in how the chip is designed, but it's still ultimately designed by Apple in this case.
Re: Isn't it (Score:2)
. they probably buy most of the sub components from IP core vendors such as Mentor, etc.
Right.
If that were true, weâ(TM)d see Qualcomm equaling or exceeding Appleâ(TM)s SoC performance long ago.
Re: (Score:2)
Apple Silicon is the brand name, just like Intel Core or AMD Ryzen. Apple Silicon is an SOC that implements the arm64 ISA, Intel Core and AMD Ryzen are CPUs that implement x86_64.
Yes (Score:5, Informative)
Typing this comment on an Apple Silicon machine. Yes, it's ready. There are almost no downsides whatsoever to the experience. x86 compatibility through Rosetta2? Check. Runs x86 code faster than my previous Intel i7-powered MacBook Pro, in fact. About twice as fast for single-threaded tasks and four times as fast for multi-threaded tasks that leverage the four high-performance cores. Apps that are optimized for Apple Silicon (ARM) absolutely fly... they're just icing on the cake.
In short, loving Apple Silicon and never looking back. Ever.
Re: Yes (Score:1)
No expectations, no downsides. Doubleplusgood!
Re: Yes (Score:5, Informative)
Your previous Mac was perma-throttling due to inadequate cooling because looks were more important, by the way. So de-facto not an i7 anymore, unless you count boost clock times.
Re: (Score:3, Informative)
Was going to post something similar. The M1 is about 75% the speed of a mid range Ryzen Mobile CPU and the GPU is even less competitive with mobile Radeon GPUs. And that's for native ARM applications, when you include x86 translation it falls even further behind.
All the Apple x86 laptops in the last few years have been thermally limited. They hit 99C within seconds and slash clock speeds to keep over overheating. Intel's crappy 14nm+++ process just can't compete now AMD is back.
Re: (Score:3)
Was going to post something similar. The M1 is about 75% the speed of a mid range Ryzen Mobile CPU and the GPU is even less competitive with mobile Radeon GPUs. And that's for native ARM applications, when you include x86 translation it falls even further behind.
If it makes you happy... (And the numbers I saw are different from yours, the 75% comes from a site comparing the iPad A12 chip with a Ryzen chip). But you can't get a Mac with a Ryzen processor, and you never will. That said, you are comparing a low end M1 with a mid range AMD chip. Mid-range M1 expected in March will have 8 performance cores and leave any mid-range Ryzen chip in the dust.
Re: (Score:2)
Midrange in March? Source?
In an argument over M1 vs. Intel why should it matter whether the machine is a Mac or not? There are Intel machines faster than the M1 Macs, just few Intel Macs. Plus, if the point is how superior Apple Silicon is, why shouldn't it be compared to AMD, and why shouldn't GPU performance be considered?
And what's with this "if it makes you happy" bullshit? All this M1 rah rah is to make Apple cheerleaders happy.
Re: Yes (Score:2)
There are Intel machines faster than the M1 Macs, just few Intel Macs.
Apple Baaaad! You are tilting at windmills.
Nobody is claiming that the M1 in an ultraportable is faster than every Intel processor ever made. You can unpucker your butthole safely now.
Re: Yes (Score:4, Insightful)
Apple have published the performance delta between the M1 MacBook Air 13-inch and the previous generation, top-spec Intel model configured with 1.2GHz quad-core Intel Core i7, 16GB RAM and 2TB SSD.
The MacBook Air with an i7 is likely using the Core i7-10510Y - Quad-core 1.2GHz with turbo boost to 4.5GHz:
https://ark.intel.com/content/... [intel.com]
Picking a task with the smallest delta, exporting images from Adobe Lightroom was 2.3x faster on the M1. Keep in mind that this is the M1 running emulated code in Rosetta 2, as Lightroom is not native yet. If the M1 is 75% of a mid-range Ryzen Mobile, then this means that this mid-range Ryzen is 3x faster than a Core i7 mobile CPU. Is that correct?
For the various hardware configurations that Apple have tested, refer to the footnotes on this page:
https://www.apple.com/macbook-... [apple.com]
Re: (Score:2, Interesting)
Unless you absolutely have to have a Mac because you are locked into the Apple ecosystem then comparing with a last gen Macbook is just gaming the numbers. Apple know their last gen Macbooks are crap, they thermal throttle under load.
Re: (Score:2)
Be that as it may, does it mean that a mid-range Ryzen Mobile is 3x faster than a Core i7-10510Y?
Re: (Score:2)
Yes it does, at least an i7 in a Macbook where it is thermally limited.
Re: (Score:3)
Geekbench says a Ryzen 9 4900HS scores 1093 single, 7090 multi core.
Intel Core i7-10510U scores 1152 single, 3765 multi core.
Apple's M1-based Mac Mini scores 1743 single, 7710 multi core.
Yeah, I know this may be a case of bringing facts to a values fight. I'm sure someone so emotionally invested in AMD Ryzen (which is twice as fast for multithreaded workloads and similar for single thread) will have plenty of reasons to still claim mobile Ryzen is 3X faster.
Re: (Score:2)
What's Cinebench say?
How about SPEC CPU 2017?
Geekbench is a toy.
Re: (Score:3)
SPEC 2017 has the M1 doing pretty good. Not the exact Ryzen model you mentioned but basically neck to neck with high end Intel and AMD desktop chips.
https://www.anandtech.com/show... [anandtech.com]
Re: (Score:2)
Geekbench is a toy.
Geekbench is a "toy" because it shows good numbers for Apple, so that makes it a toy.
Cinebench is the real thing, especially since the first published and widely quoted benchmark didn't test the M1, but the A12 from an iPad Pro.
Re: (Score:2)
That's very good single-threaded performance. Now, what happens if you run it on all cores?
Re: (Score:2)
Fortunately the very next page has that.
https://www.anandtech.com/show... [anandtech.com]
Re: (Score:2)
Obviously I didn't bother to click that far. That's very good CPU performance overall, as far as synthetic benchmarking.
It will be interesting when Cinebench runs on it. If that's good as well then it will be necessary to concede that Apple has really accomplished something :)
Seriously though, it's already quite impressive.
Re: (Score:2)
Geekbench is a "toy" because it shows good numbers for Apple, so that makes it a toy.
Geekbench is a "toy" because no one takes it seriously.
Even on Mobile, Antutu is considered to be the benchmark of benchmarks. On desktop, it's a hodgepodge of non-synthetic benchmarks.
The SPEC results are very good, though, and I take those seriously even though that is a synthetic benchmark.
Re: (Score:2)
Believe it or not.... same article has Cinebench.
https://www.anandtech.com/show... [anandtech.com]
I hope these chips can scale to higher core counts and that they wake the beast like last time AMD beat intel when 64-but was new. Hopefully Intel recovers their glory days, competition is good for everyone.
Re: (Score:3)
> The M1 is about 75% the speed of a mid range Ryzen Mobile CPU
Looking at both SPEC and GeekBench, the AMD Ryzen 9 4900HS, which I believe is still their highest-end mobile product and has been called "AMD's most important product ever", is about 70% as fast as the M1 in single-core and 95% in multi-core in spite of having twice as many cores.
> They hit 99C within seconds and slash clock speeds to keep over overheating.
If we're going to talk about TDPs the contrast is even starker. The M1 is a SOC wit
Re: Yes (Score:5, Informative)
> You know why Apple fans love SPEC and Geekbench?
Because they are used universally on practically every CPU and platform available.
> It's because Apple products do well on them
Non-M1 Apple products do quite poorly on them, which is why Wintel, and especially AMD nerds, would often quote them in Apple threads. They also remain extremely popular in AMD-vs-Intel flame wars.
Bu now, when the M1 is winning these handily, suddenly those very same people say these tests suck. Hmmm.
>Look at Cinebench or the time needed to export a video from Final Cut.
If you insist. Let's start with Cinebench. From:
https://www.cpu-monkey.com/en/cpu-amd_ryzen_9_4900hs-1285
https://www.cpu-monkey.com/en/cpu-apple_m1-1804
In single thread, the Ryzen 9 4900HS, AMD's flagship laptop chips (according to their web site anyway), gets 1246. The M1, Apple's lowest-end laptop chip, gets 1514. What was that? Oh right, "falling massively".
On that second page, the M1 easily outperforms all mobile chips and is only *slightly* behind the 5800X, what AMD describes as an "Elite Gaming Desktop Processor", with a TDP of 105W.
I'm not sure what you hoped this comparison would illustrate, but whatever it was, this seems to be doing precisely the opposite.
But perhaps you were referring to multi-core? In that case, the 4900HS scores 11006 vs. 7760 for the M1.
That does sound awesome for the 4900U until you consider that the M1 has only four full-performance cores while the 4900HS has eight, and has a TDP of 15 vs. 35 for the 4900U.
And FCP?
https://techcrunch.com/wp-content/uploads/2020/11/Final-Cut-Pro-5-minute-8K-Timeline-Export.png
The M1 MBP absolutely *crushes* every other machine *with integrated graphics*. In fact, so does the MBA and MM, both of which are the lowest-end products in Apple's lineup. Yes, indeed, the bit about "integrated graphics" is important, and we should not forget that. And we won't, because...
It is clear to anyone with two brain cells to rub together that whatever Apple chooses to call their mid-performance CPU, perhaps "N2Y", will replace the integrated graphics with more performance cores and an external GPU, because that's precisely what all of their mid-line machines already do.
Looking over the design layout, it was clearly intended to make this easy (note the way the memory fabric is positioned relative to the various cores). At a first glance, it appears the GPU portion of the die would hold at least four more high-performance cores, perhaps six if one uses the room for the other four low-performance cores, and another 12 MB of cache.
It is clear that this design, which requires only re-layout and no changes to the fab, will appear in the new year.
But sure, and Trump won too.
Re: (Score:2)
If you like, how about compile performance? https://techcrunch.com/2020/11... [techcrunch.com]
Not only does the M1 compile Webkit much, much faster than any other Mac laptop (with Intel processors, admittedly), it only uses 9% of the total battery to do it, while the old processors can take around 3/4 of the battery for the same task. And it's an apples-to-apples comparison because those old laptops and the new ones have basically the same battery life.
Like, this chip is just BETTER. It is. ARM is better at low power applic
Re: (Score:2)
Are they really saying it's faster at compilation, a highly parallel task, than a 28 core Xeon or whatever the Mac Pro has? Well they don't say what model of Mac Pro, so who knows...
Is the Mac Pro compiling from HDD or something? Or is the compiler single threaded? In fact what compiler is it, because they didn't bother to say?
This is what happens when idiots do benchmarks. I think we will have to wait for the Gamer's Nexus results, those guys are one of the few who do proper controlled, repeatable tests us
Re: (Score:3)
It's right in the article. But you have to read it.
* 2019 Mac Pro 12-Core 3.3GHz 48GB w/AMD Radeon Pro Vega II 32GB
I'm assuming they're compiling in XCode, so Clang is the compiler. In any case, they'll be using what's default on the system, not mixing and matching compilers.
But yes, the M1 is faster at compiling than the 12 core Mac Pro.
The new Mac Mini M1 also performs pretty well when compared with a 6-core i7 and external GPU (Radeon XT5700) doing rendering/exporting/audio tasks while only drawing about
Re: Yes (Score:2)
It was perma-throttling because intel can't get it's shit together and make a chip that doesnâ(TM)t run hot and suck an asstonne of power, and demands that you blast fans at it non-stop. The lesson of the M1 is that you actually can make chips that don't destroy your battery but can still do actual work. That M1 is in an identical case to the old chip and runs circles around it, so clearly the case and thermal design isn't the problem. In fact, I suspect no amount of air cooling would make that i7 perf
Re: (Score:2)
Re: (Score:1)
The 7448 Could do 2+ GHz, but by the time it was far too late by that point. Apple Intel hardware was already in development.
The G4 powerbooks were power cookers. There was a big thing about people getting burns from the thermal dissipation issue. You conveniently seem to have forgotten about that.
Also IBM weren't interested in apples direction for the G4. (also was news worthy) which is why they got dumped and apple went to intel. The decision to dump was already made by the time the 7448 came around (Ap
Re: (Score:2)
Rumor has it? Rumor has it? People say? I heard? Sounds legit.
The PowerPC never significantly advanced beyond the original 601/603/604 designs because Moto/Freescale was incompetent, not withstanding the IBM G5 point product. That's what led to the switch, there was no shady conspiracy to dump a superior architecture, just an attempt to become competitive.
Re: (Score:3)
The PowerPC never significantly advanced beyond the original 601/603/604 designs because Moto/Freescale was incompetent
Not because they were incompetent, but because they were not interested. Apple had the choice between PowerPC from Motorola which was perfect for running tiny little routers needing very little CPU power, or POWER from IBM which beat the shit of any Intel chip (including Xeon and Itanium, which was still relevant at the time), at a cost and power consumption just out of this world. Nobody produced a PowerPC chip that was suitable for a desktop or laptop PC.
Re: (Score:1)
Re: (Score:2)
All is fine if what you do doesn't require more than 8/16GB memory. If that happens, anything coupled with more memory will wipe the floor with your iSilicon..
Obviously that is the case, Apple pretty clearly say that it maxes out at 16 GB. This is the entry-level M1. There will be a more powerful version released next year, named something like M1X or M1Z, that will be used in the higher-end 13-inch MacBook Pro, as well as the 16-inch MacBook Pro and the iMac. This SOC will have more cores and support more than 16 GB RAM.
Re: (Score:2)
Typing this comment on an Apple Silicon machine. Yes, it's ready. There are almost no downsides whatsoever to the experience. x86 compatibility through Rosetta2? Check. Runs x86 code faster than my previous Intel i7-powered MacBook Pro, in fact. About twice as fast for single-threaded tasks and four times as fast for multi-threaded tasks that leverage the four high-performance cores. Apps that are optimized for Apple Silicon (ARM) absolutely fly... they're just icing on the cake.
In short, loving Apple Silicon and never looking back. Ever.
Yes, but does it come in Rose Gold? If not, it's a deal breaker for me. If it doesn't match my keyboard, my code won't compile right. I just know it.
And it better come with wheels too...
Re: (Score:2)
Wait, it can run 32-bit apps again like old Office 2011?
Re: (Score:1)
Doesn't run java. Keeps hanging. Multiple VMs. Can't use eclipse. That's pretty big for a lot of people.
Multiple odd things also happening in other development apps. Plus the usual 'frozen' and 'hanging' from x86 stuff.
Yes. Apple silicon still needs work at the OS level. Typed on an M1.
Re: (Score:2)
Several video editing reviews on YouTube have specifically mentioned most 3rd party plugins for Final Cut Pro do not yet work. Apparently Rosetta 2 can't run them. Maybe it's because Final Cut Pro is running native ARM code?
The no-plugins Final Cut Pro performance looks absolutely astounding, so I'm pretty sure those 3rd parties will soon port their code. But if you edit video in Final Cut and use any plugins, the reality today isn't quite "almost no downsides whatsoever".
Likewise for virtual machines or
Re: (Score:2)
You can't mix and match x86 and ARM code in the same process. So if the parent process is ARM, the plugins must be ARM as well. This probably applies to just about any program with a plugin architecture. Note: you may be able to launch Final Cut in x86 mode and have the plugins work? (With a performance penalty, of course.) Not sure, I don't have either an M1 Mac or Final Cut.
Ready? (Score:3, Interesting)
It would be like trying to run Android on a new iPhone. It's so far out of the realm of possible, that it isn't worth discussing.
Re: (Score:2)
Yeah, from a technical point of view I think M1 is a great step in the right direction. Intel has clearly left a lot on the table in terms of power efficiency (we'll see about performance soon I expect). But I do think there is a good chance that once M1 goes mainstream for Apple they will make everything go through the Apple Store. They could just tell folks who want to install their own software to buy the Intel macs for the next couple of years and then ditch them after that.
As you said, most users won't
Re: (Score:1)
Re: (Score:2)
That fact that you're saying that means the commercial worked brilliantly.
Re: (Score:2)
An Android emulator would run just fine on an iPhone.
Same as Linux VMs run just fine on M1.
Or like android VMs running on Mac OS, oops, you did not know that?
I don't recommend "Andy" thought, it installs mallware on your Mac.
Re: (Score:2)
It's ready for Apple. These new processors are going to be like a dream for Apple as it will allow them to completely lock down the environment that their users are able to access.
Initially I was also very afraid of that, but at least for now they are specifically not doing that. They even took time to explicitly mention this in the WWDC Apple Silicon overview [apple.com] video, mentioning that various hardware features that lock down the OS can be disabled in case you're developing kernel extensions, are a security researcher, or simply want to explore the OS.
The only one that's not mentioned, and which is quite important, is whether you can boot unsigned OSes. I would assume it's possible sinc
Re: (Score:3)
The only one that's not mentioned, and which is quite important, is whether you can boot unsigned OSes. I would assume it's possible since they did mention that you can disable code signature verification to install and boot older macOS releases that are no longer signed by Apple, but I'm not sure.
It was mentioned on WWDC. You can run unsigned OSes from an external hard drive or SSD drive. One class is older MacOS versions which won't be signed at some point (not relevant to M1 chips right now, but in the future). And of course any other OS, like Linux.
Correlary (Score:2)
These new processors are going to be like a dream for Apple as it will allow them to completely lock down the environment that their users are able to access.
The corollary to this is it will allow them to implement a chipset that doesn't have mandatory Intel Management Engine backdoors installed.
Like it or not, the technology that keeps users out of their own system is the same technology that keeps everyone else out, too.
Re: Ready? (Score:2)
These new processors are going to be like a reoccurring wet-dream for Apple haters as it will allow them to completely lock down the environment that their users are able to access.
Wake up, clean yourself off, it didn't happen.
I'm looking out for your safety here so you don't beat your meat purple trying to make the dream real. Apple 110% fully controls the firmware on all of their devices right? Linux ran on PowerPC Macs? Linux ran on Intel/EFI Macs? Linux ran on Intel/UEFI Macs? Any of those could have been locked down tighter than a game console. If Linux doesn't run on the new architecture it's because it doesn't have support yet. If that didn't wake you up, I don't know wh
apps that don't work (Score:5, Informative)
LFO Tool plugin
Native instruments
Nectar plugin
Neutron plugin
Pro tools
Screenflow
Google Drive File Stream
Virtualbox
Revit
Parallels
VMWare Fusion
Haskell
I understand why VM software wouldn't work, but google drive file stream?
Re: (Score:2)
Given what google drive file stream does, it probably needs to integrate with the filesystem layer rather than run as a standalone application. It's the integration that breaks, since you can't integrate code from different architectures into a single running program or driver. That's why a lot of the other failed applications are plugins for something else.
Standalone applications tend to run fine.
Re: apps that don't work (Score:2)
So the missing information is: does it run on Big Sur at all?
Re: (Score:1)
It doesn't matter why, or how, or whatever other cope or shilling you might try to state, this situation, no matter what you or any other person says, is *still* unacceptable.
Reject, reject, reject. Do not buy. Send the message. AMD or nothing.
Re: (Score:3)
Google File File Stream is mounting a drive, and probably is using an obscure kernel call for that, instead of doing it "normally".
More surprising is that Haskell is not running.
Re: (Score:3)
From Apple's developer page (https://developer.apple.com/documentation/apple_silicon/about_the_rosetta_translation_environment)
What Can't Be Translated?
Rosetta can translate most Intel-based apps, including apps that contain just-in-time (JIT) compilers. However, Rosetta doesn’t translate the following executables:
Kernel extensions
Virtual Machine apps that virtualize x86_64 computer platforms
Rosetta translates all x86_64 instructions, but it doesn’t support the execution of some newer instruction sets and processor features, such as AVX, AVX2, and AVX512 vector instructions. If you include these newer instructions in your code, execute them only after verifying that they are available. For example, to determine if AVX512 vector instructions are available, use the sysctlbyname function to check the hw.optional.avx512f attribute.
From what I understand, there is some software that doesn't correctly test for the presence of AVX, etc, and these programs might fail. Perhaps Haskell has a bug in that it doesn't test for AVX when it should?
Re: (Score:2)
VMWare and Parallels are in the works. Both companies have committed to M1 native releases.
As has Adobe for its CS offering.
I bought a new 16in Intel MBP three months ago (to replace a 5 yr old 13in MacBook). I knew full well that the M1 hardware was coming but I needed to get one in a hurry (grandson spilt his friut juice all over the keyboard)
If this one lasts as long as the previous one then by the time I need to upgrade, we will probably wondering what the fuss is all about for the Intel/ASI transition.
Re: (Score:2)
It's a shame they don't list if the app runs well too. It's hard to measure relative performance since x86 Macbooks are thermally limited, but a comparison with a similarly priced Ryzen laptop would be very informative.
Re: (Score:2)
Sure but say you have a particular app that you use a lot, you might want to know if it runs well on an M1. For example video export, which can be either CPU or GPU bound depending on the encoder.
Clearly, M1 is not ready for business. (Score:3)
Adobe InDesign, Lightroom, After Effects, Illustrator and Premiere Pro. Acrobat isn't even on the list.
There's a Photoshop beta that works, but that's it. Granted, it's the big one, but the only one?
People who use Macs professionally, at least in my experience, tend to be working with Adobe products.
Oh, and not only is Parallels not marked as working, JAMF isn't even on the list. Kaseya hasn'
Words ... hard to get you head around. (Score:2)
"Why are you tryin' to read that word? ... You a fag??" has become reality.
Alright, sirs, I'm commencin' to receive my tattoo now.
Sponsored by Carl's Junior.
Re: Words ... hard to get you head around. (Score:2)
I can't buy Apple silicon (Score:2)
I can't buy their chips and put it in a project. So why are we talking about it? Whatever Apple does might as well be a black box.
I remember when people complained an Intel Mac had to emulate a PowerPC, and before that when people complained that a Power Mac had to emulate 68K. But after a few years nobody cared, because usually software developers are forced along the path of updating their development tools to support the latest OS and hardware. It's just me and like 5 other people in the world that still
Re: (Score:2)
...It's just me and like 5 other people in the world that still have Macs with OS X 10.4 (Tiger), people running actual businesses don't care about old crap that Apple refuses to support.
People running actual businesses using old crap that vendors refuse to support, will find themselves suddenly caring about the actual reality of Shit Happens.
Meh, perhaps I'm overreacting...it's not like hardware ever dies and software becomes incompatible...
Re: (Score:2)
People who are serious about running a business will ensure they're never in a position that they're stuck with something no longer supportable.
Vendors publish support timelines that you can see in advance, you have to plan your migration strategy to occur before support ends. You have to also plan for contingencies should the supplier change their plans - what is your exit strategy for replacing this vendor's products with something else?
Re: (Score:3)
People who are serious about running a business will ensure they're never in a position that they're stuck with something no longer supportable. Vendors publish support timelines that you can see in advance, you have to plan your migration strategy to occur before support ends. You have to also plan for contingencies should the supplier change their plans - what is your exit strategy for replacing this vendor's products with something else?
You speak so much common sense that it's comedic when compared to reality.
If the business world was so "serious" about maintaining products properly, we probably wouldn't have been worried about the entire computing world melting down due to Y2K, because most critical systems were still running on ancient COBOL, operating for decades under the if-it-works-don't-fix-it mantra.
Also feel free to consult most who are now stuck running Oracle or SAP. That's the other problem with keeping software up to date. I
Re: (Score:2)
I remember when people complained an Intel Mac had to emulate a PowerPC, and before that when people complained that a Power Mac had to emulate 68K.
I think the first 150 MHz PowerPC was the fastest 68k processor ever. (Don't know if they made faster 68k processors since). So fast that some guys managed to make Atari OS run on it, and it was the fastest Atari computer ever built.
Re: (Score:2)
Old man yells at cloud.
Corporate Arrogance. (Score:2)
Hey you. App developer.
Yeah. You. Come here.
WE have new hardware to introduce to the computing world, and YOU need to make sure YOUR shit works on it. Oh, it doesn't? Wow. Sucks to be you. Now go fix YOUR shit.
(I've spoken about Corporate Arrogance for quite a while now, but if you were looking for a steaming pile of an example...)
Re: (Score:2)
i = 0;
While (i != 3000)
Yo, App developer_i.
Yeah You_i. Come here.
We'd like to build new silicon, tell us which special sauce you_i require.
i = i + 1
end While;
Okay, you hardware guys, we have 3000 specs for you to fulfill. Get cranking, there'll be another 3000 next week.
Not ready for music production yet (Score:3)
I've no doubt it will be in fairly short order so this isn't a criticism, more...just noting that for some areas it's best to wait right now.
Or visual stuff (Score:2)
Re:shilling (Score:5, Funny)
Apple is desperate for this M1 mac to succeed.
In fairness, have you ever seen a company that's only mildly interested in a new product launch to succeed?
Re:shilling (Score:4, Informative)
Apple TV was famously described as "a hobby" for the company by Steve Jobs. So, yes.
Re: (Score:3)
Stop paying attention to PR. Jobs was 100% about PR. You couldn't trust anything he said. He spun their lukewarm success as a lukewarm attempt.
Re: (Score:1)
They have every reason to LIE about performace of the first generation machines because unlike the last platform switch, they forgot to deliberately nerf the last gen of Intel macs harder than usual.
Just like how they nerfed the last powerbooks to promote intel macs and killed powerpc support for snow leopard midway through the beta testing.
An
Re: shilling (Score:5, Insightful)
They lie about the performance. All the reviewers lie about the performance. All the news people lie about performance. It is a conspiracy! There will be a press conference about it at the Four Seasons ... Landscaping.
Re: (Score:1)
They have every reason to LIE about performace
But the reviewers don't. You can get a lot more eyeballs on you YouTube video if you can debunk a claim by a major manufacturer than if you confirm it.
Re: (Score:2)
we've never in human history had companies worth $2Tn dollars
You might need to learn a bit of history. While strictly true if you just look at the number of dollars (without adjusting for inflation, etc), there have been companies that were worth this much in the past. In particular, the Dutch East Indies company was estimated to be worth nearly $8Tn in today's dollars. Of course, comparisons to times long-ago can be really hard to do.
https://dutchreview.com/culture/history/how-rich-was-the-dutch-east-india-company/
Re: (Score:2)
> They have every reason to LIE about performace
LIARS!
> And if you don't believe that
SHEEPLE!
> All it costs Apple is a bunch of free shit.
CONSPIRACY!
> All the reviewers lie about the performance
CONSPIRACY!
> Reviewers are usually bought off with free stuff
CONSPIRACY!
I remember this in the Atari ST/Amiga days, it brings a tear to my eye that the more things change...
Re: (Score:2)
I really wish I had mod points for you, I laughed way too hard at this.
Re: (Score:2)
MS with the Zune. And their phones.