The No.1 Website for Pro Audio
Apple to announce ARM chips for all Macs
Old 3rd August 2020
  #1081
Gear Maniac
It would be nice, say in logic, that one could run the system at low buffers until the system is near max. Currently my iMac is hardly taxed and the glitches start, one cpu meter throws the whole thing, needing buffers adjusted and the fans blow!
I can’t say my iPad has been pushed at all given I only browse, so it’s not clear yet the benefits. But hopefully that’s the focus, use the full abilities of the cpu.
Old 3rd August 2020
  #1082
Special Guest
 
TheHanes's Avatar
Quote:
Originally Posted by bgrotto View Post
You'd hafta be crazy to run a business around your computer and invest in 1st gen processor tech. Anyone who's recently bought a mac pro will be in great shape for upwards of a decade.
I agree. We'll be buying the current models of Mac Pro soon to make sure we've got the established processors. These computer should last us until approximately retirement.

Our current trashcan Mac Pro Late 2013 model has served well but we need to get on board with the upgrade before everything needs a rewrite for the new processors. TBH the current setup works so well I would use it another couple of years, but this processor change has pushed that decision forward.
Old 3rd August 2020
  #1083
Quote:
Originally Posted by TheHanes View Post
I agree. We'll be buying the current models of Mac Pro soon to make sure we've got the established processors. These computer should last us until approximately retirement.

Our current trashcan Mac Pro Late 2013 model has served well but we need to get on board with the upgrade before everything needs a rewrite for the new processors. TBH the current setup works so well I would use it another couple of years, but this processor change has pushed that decision forward.
We’re eyeballing a similar game plan for my shop. We ran our cheese grater G5 for a solid decade, and have been using a Mac mini (!!!) in the interim for the past four years, which outperformed the old G5. We bought it as a placeholder while we awaited the then-new Mac Pros to be vetted, but it worked so well we kept it going. But lately it’s showing signs of its age (and build quality/longevity) so we gotta pull the trigger on something more long term. I expect we’ll be in fine shape for at least ten years.
Old 3rd August 2020
  #1084
Lives for gear
 
uOpt's Avatar
Quote:
Originally Posted by bace View Post
I think it is ok to have port limitations on machines that does not have the power for it. But Im afraid that Apple remove ports for aesthetic reasons.
A USB-C plug is actually pretty big and invasive on the machine side. You need to create some space for it, which on the 12" macbook is at a premium.

And then you need to wire it up. On a modern machine that means data and it means 40W or so of power. Those are big power lines by the standards of small electronics.
Old 3rd August 2020
  #1085
Gear Addict
 

Quote:
Originally Posted by TheHanes View Post
These computer should last us until approximately retirement.
I agree 100%, but if the new chips are smoking the old intel ones as far as performance and DAW workflow, people will want to upgrade whether or not the old ones work fine, just depends how much better the new Apple chips will be and peoples budgets.

But yeah those new Mac Pros will be great for a very long time. I should be able to get another 7 years out of my iMac Pro, which costs the same today as 2 years and 8 moths ago when it was released.
Old 3rd August 2020
  #1086
Lives for gear
Quote:
Originally Posted by tbirdparis View Post

In the computer industry in general, it has _never_ made sense for any manufacturer to launch a new machine that is slower than the one it replaces. Apple could not be stupid enough to be about to attempt to do this, especially not when they are hailing in their own technology that they want to make a big deal about.

Yet, that is what Apple already have done once. Many years ago before my work place parted with their PPC-G5, that was getting replaced with an Intel tower Mac with more cores, we ran a heavy real-project render test in FCP on both. The several years old G5 ran circles around the new Mac with the then current FCP.

Of course, not long after there were new compiles that took advantage of the new architecture. I feel we have reason to believe that it will not be as bad this time. But that reality is yet to be confirmed.
Old 3rd August 2020
  #1087
Special Guest
 
TheHanes's Avatar
Quote:
Originally Posted by Artcutech View Post
I agree 100%, but if the new chips are smoking the old intel ones as far as performance and DAW workflow, people will want to upgrade whether or not the old ones work fine, just depends how much better the new Apple chips will be and peoples budgets.

But yeah those new Mac Pros will be great for a very long time. I should be able to get another 7 years out of my iMac Pro, which costs the same today as 2 years and 8 moths ago when it was released.
The problems are going to be EVERY piece of software needing deep rewrites.

Big Sur OS will be the first problem. Big Sur will get rid of the .kext portion of software programming. Next, any computer shipped after Big Sur release will not be able to go back to previous OS generations.

Expect a lot of software to never be re-written, or pay upgrade fees on everything.
Old 3rd August 2020
  #1088
Gear Nut
 

Quote:
Originally Posted by Mikael B View Post
Yet, that is what Apple already have done once. Many years ago before my work place parted with their PPC-G5, that was getting replaced with an Intel tower Mac with more cores, we ran a heavy real-project render test in FCP on both. The several years old G5 ran circles around the new Mac with the then current FCP.

Of course, not long after there were new compiles that took advantage of the new architecture. I feel we have reason to believe that it will not be as bad this time. But that reality is yet to be confirmed.
Good point, but as you pointed out, there were a few mitigating factors that made it not just about the pure technical specs of the hardware itself. At the time I recall there being talk that FCP was not optimised, I can't recall the details but it might have been to do with a lot of the work that should have gone to the GPU not doing so because driver support wasn't implemented properly on Intel yet. So even if you had better on-paper specs, you didn't get results that correlated with this, at least not right away.

In any case, they certainly didn't launch the Intel Macs with objectively lower hardware specs than the outgoing models they replaced. I don't recall there being a single Intel version of an existing machine that had lower theoretical performance than the outgoing model. Nor was there one when they went to PPC from 68K either. If there were any actual real-world caveats to this, it came down to issues to do with the transition itself - code emulation and unoptimised software limiting things with overheads and bottlenecks. Basically exactly what happened with your FCP example, something temporary and due to transition issues not actual pure hardware specs.

But you do make a good point, because this is a real concern. There's always going to be a period of cutting through the reeds during any kind of transition, especially early on when early adopters are running a hodgepodge of native and translated apps. This is of course the reason why typically in our industry (and many others), no one goes in early.

But this time round it should smoother because they already have several other transitions under their belt... and they also have a whoooole lot more money to invest in this than they've ever had as a company. You don't even have to believe they've been benevolently investing in this just to delight the Mac faithful. It's all a culmination of what has clearly been a very long term (and ongoing) technology investment which has not just been for unseen prototypes - it has driven the iOS products to where they are now with plenty of return on investment already.

Combine all this with the fact that there's no longer a 3rd party involved in as fundamental and crucial a component as the CPU architecture itself (meaning that they have more control over what they're transitioning _to_ than ever before), and the outlook should be better for a relatively smooth transition.

So I don't have any doubt that just like every other time Apple has launched a new generation of any given machine, someone will get up on a stage and show charts that demonstrate a large visible margin of improved hardware stats across the board. For every model, like usual. And to much applause, like usual. All computer manufacturers do the same, just with less fanfare. They all refresh machines as newer CPUs come in, or anoint a new model when it's more than just a refresh. But never do they do it while also saying, well the hardware is less capable than the previous one but hey, buy it because it's new!

However, I also don't have any doubt this will all need to be taken with a pinch of salt, depending on the transition status of the software people need to use, the availability of reliable drivers and so on. Regardless of demonstrably faster hardware, it doesn't always mean better results till things are native and optimised - of course.

Personally I can see myself picking up a general purpose Apple Silicon laptop far sooner than I'd attempt to switch my work machines over. I could use a better personal laptop than what I have now, and there's basically nothing at stake when I don't need it to do anything more than use the software it already comes with. I'll probably end up being one of the last who moves to the new machines for my workstations though, even if the transition turns out to be smoother than expected...
Old 3rd August 2020
  #1089
Lives for gear
 
juiseman's Avatar
Quote:
Originally Posted by TheHanes View Post
The problems are going to be EVERY piece of software needing deep rewrites.

Big Sur OS will be the first problem. Big Sur will get rid of the .kext portion of software programming. Next, any computer shipped after Big Sur release will not be able to go back to previous OS generations.

Expect a lot of software to never be re-written, or pay upgrade fees on everything.
Not to mention A lot Audio interfaces that are not USB class Compliant...RIP...well you can guess what will happen there. I'm glad to hear someone with the same concerns as myself. I keep bringing up how protools will handle this....well..lol they might just "not" handle it. I can see backwards
compatibility being an issue. They may have to make a translation app to open old projects.

This transition will take some time. My guess is the continued support for Intel based mac's will have to be longer than the PPC-Intel transition.
So that's somewhat a relief for the guys who got those 2019 Mac Pro's

There has been lot of talk about how great these Apple Soc's will be...and that is great!! (I'm holding my excitement back) my view is the software is going to be the biggest hurdle. I think folks forget how many months i takes 3rd party software guys to catch up with the normal mac os updates; now, then let's toss an architecture change in there too, plus I guess were still in the middle of a pandemic?, so there is that.. we can't expect the smaller software developers to magically rewrite, test, verify, debug, troubleshoot...ect their plugs while eating the cost at a snap of our fingers....
I could be wrong.. it happens a lot; and I'm not afraid to admit it.
Old 3rd August 2020
  #1090
Gear Nut
The reality will be somewhat in between. A lot of stuff that has been properly written for MacOS will just work. Most of the time, a vast majority of it, developers don't need to be so close to the metal, and the difference is just in compiling for another architecture. If they're doing their jobs correctly most of the code will cross-compile correctly. There's a testing burden for sure, but again if all goes according to plan, that'll be that.

The rest is just fear-mongering.

Apple has no motivation to release a product that performs worse or is less stable than their last product. I'm sure there will be incidents where that turns out to be not true, but for most people, most of the time, there won't be an appreciable difference.

Non-class-compliant drivers are a different story. I've got some fears there myself.
Old 3rd August 2020
  #1091
Gear Nut
 

Quote:
Originally Posted by juiseman View Post
I keep bringing up how protools will handle this....well..lol they might just "not" handle it. I can see backwards
compatibility being an issue. They may have to make a translation app to open old projects.
It's actually hilarious to imagine ProTools being fast with any kind of update, let alone a major transition with OS updates along the way just for kicks. I don't directly use PT myself very often but it's a major part of the picture in my corner of the industry (film scoring), so I'll definitely be hearing about it all from colleagues.

I think you're right that the continued support of x86 down the track might have to go longer than previous transitions if Avid are going to stay onboard. I don't know if Avid have made any official announcement on what their plans are yet (have they?), but it would definitely be a seismic shock in the industry if they decide to no longer support Mac OS. Will definitely be keeping an eye on that.
Old 3rd August 2020
  #1092
Lives for gear
 
uOpt's Avatar
If NVidia indeed buys ARM then Apple is probably switching to RISC-V in 2025 at the latest :D
Old 3rd August 2020
  #1093
Lives for gear
 
TAFKAT's Avatar
 

Quote:
Originally Posted by TheHanes View Post
I agree. We'll be buying the current models of Mac Pro soon to make sure we've got the established processors. These computer should last us until approximately retirement.
Thats interesting, as I was wondering what potential clients for the current Mac Pro were thinking moving forward with this latest announcement.

I have heard of some being spooked and weighing up an exodus.

And you are right, plenty will be able to buy and lock down the configurations at a critical point, put their heads down and just work for years , then stick their heads up after all the storm has blown over and re-evaluate.

It will only those wanting/needing to be on the cutting edge that will be concerned with the latest/greatest at any given point.
Old 3rd August 2020
  #1094
Lives for gear
 
ponzi's Avatar
New product announcements killing current product sales has always been true in computing. Apple hopes to suppress pc sales with these ‘leaks’, but they get blowback on their own revenues. Always a balancing act for product managers. Not in the least unique to Apple. Probably true outside of the computer industry--read/watch on 'product lifecycle'. I wish kids were taught this stuff in high school so they could be more sophisticated customers.

Last edited by ponzi; 4th August 2020 at 02:33 AM..
Old 4th August 2020
  #1095
Lives for gear
 
Lady Gaia's Avatar
Quote:
Originally Posted by TAFKAT View Post
Thats interesting, as I was wondering what potential clients for the current Mac Pro were thinking moving forward with this latest announcement.
The high end gear is usually purchased by those who make enough money using it that it's just a cost of doing business. They're risk averse, so they wouldn't be early adopters of a new architecture anyway, and chances are the Mac Pro is the last model to get replaced.

If they need the hardware they'll buy it.

It's the crazed enthusiast who has to have the very latest that is likely to be spooked and anxiously wonder what the right path will be, but that's not likely to be the majority of the Mac Pro market.

The real risk for Apple is in not being decisive about making a strategic bet. History is littered with the Digital Equipment Corporations of the world who tried to cling to a successful model way past its sell-by date.
Old 4th August 2020
  #1096
Quote:
Originally Posted by Lady Gaia View Post
...History is littered with the Digital Equipment Corporations of the world who tried to cling to a successful model way past its sell-by date.
Thinking about DEC makes me sad. I grew up in Massachusetts. In Jr High (US grades 6-8), they brought their computer bus (a kitted-out tour bus) to our school, parked in the playground, and brought in interested kids during math classes to tour the bus and try out the computers. They had a bunch of VT terminals and a mini that I don't recall. It was 1983/1984.

They also provided a couple CP/M DEC Robins to the school. That was where I learned BASIC and Multiplan, before moving on to the C64. Then, in college, were were all on VAX/VMS or Ultrix.

No record of that computer bus anywhere that I can find. Too bad, really, as it was pretty amazing. Better than the bookmobile

Anyway, I wish DEC had survived. A little bit of Boston-area culture long-gone.

I'll step away from my rocking chair now :D

Pete
Old 4th August 2020
  #1097
Lives for gear
 
Dave_Ionic's Avatar
Quote:
Originally Posted by Psychlist1972 View Post
Thinking about DEC makes me sad. I grew up in Massachusetts. In Jr High (US grades 6-8), they brought their computer bus (a kitted-out tour bus) to our school, parked in the playground, and brought in interested kids during math classes to tour the bus and try out the computers. They had a bunch of VT terminals and a mini that I don't recall. It was 1983/1984.

They also provided a couple CP/M DEC Robins to the school. That was where I learned BASIC and Multiplan, before moving on to the C64. Then, in college, were were all on VAX/VMS or Ultrix.

No record of that computer bus anywhere that I can find. Too bad, really, as it was pretty amazing. Better than the bookmobile

Anyway, I wish DEC had survived. A little bit of Boston-area culture long-gone.

I'll step away from my rocking chair now :D

Pete
heh My dads chips ran DEC’s. Performance PR4000 RISC processor fastest chip in the valley in the late 80’s early 90’s. and yes I can equivocally say that.
Oh and before he went Solo he was at Fairchild and before that a bit of Mass for you he was Vice President of Transitron between his days at GTE Sylvania and Fairchild.
Old 4th August 2020
  #1098
Lives for gear
Quote:
Originally Posted by TNM View Post
Steinberg in fact, introduced asio guard simply because there was no other way to get decent OSX performance and too many people were complaining of the performance disparity.

it doesn't matter the DAW, OSX starts evening up with windows performance at around 512 buffer and above, hence, logic dual buffering, asio guard, pro tools dual buffer and so on.

It was not steinberg's fault whatsoever and I have no idea where you are getting this from.
OSX is broken at low latencies, it was 10 years ago, and it still is now.
Cubase prior version 6 ran at the real buffer setting throughout which is why it collapsed on OSX.
Even some so called "pros" still don't understand how it works to this day.

For example, there are current videos of pro tools expert saying how amazing the mac pro performance is with PT at 64 buffer, but no tracks were record armed which means every single one of them was using a 1024 playback buffer.. Same with the popular channel that did the test with mac pro and logic

In 2005, i wrote a 46 track song, half VI's half audio vocal tracks, lots of effects, on a single core athlon 2600 with ESI Julia interface, Cubase SX3 and 64 buffer. Yeah, by the end of it, the cpu was beyond hammered, there was just no asio headroom left, so I did end up putting it at 256 so it could be smooth playback, but the point is that windows is just way better at low latencies..at REAL low latencies I mean.

Reason performed over 3x as well on windows till they introduced the dual buffer. Live performs better on windows now cause it still uses the real buffer.
S1 performed up to 8x better on windows before the dual buffer.. on the very same mac.

IN FACT, and this is a FACT...
on my previous macbook pro sandy bridge where OS 10.9 was the final I installed, quad core 2.2ghz, Studio one on windows at 64 buffer (completely prior to the dual buffer scheme) via bootcamp, outperformed Logic 9 on the same macbook in OSX, and that was using the medium process buffer (1024 samples). I tested with non record armed tracks using identical effects and S1 in windows at 64 buffer got around 30% more FX plugins than Logic.
So even with the trick dual buffer, Logic couldn't compete.

Avid, steinberg, presonus et al introduced dual buffer because it's the ONLY way to get decent performance on OSX, period.

Does that mean the DAW is broken if it doesn't have a dual buffer? No, that means the OS itself is broken.
Asio outperforms core audio.
Core audio has aggregate features and the natural multiclient and integration rocks.. I agree with that..

But let's take an up to date example.. latest pro tools on bootcamp on my imac pro thrashes pro tools on OSX on the same imac pro (whether mojave or catalina). This is due to OSX aggressive speed step. Same PT projects on same PT version.

So we have two problems here with OSX.. it's poor low latency performance and not being able to take full advantage of any cpu in any mac currently available with regards to DAW use due to constant frequency hopping.

I don't see how any of this is Steinberg's fault.. they basically made a patch, like other daw devs did, to work around a ****ty performing os for LL DAW use.
Apologies I had a lot of things to do, and this conversation was going nowhere. Let's agree on a few things, if we can.

I never compared Windows to OS X performance at low latencies, or made any assumptions about what is true or not true there.

I explicitly said Cubase was underperforming at low latencies on OS X compared to Logic and Digital Performer. At the time DP had no variable buffer (Pre Gen), for unarmed tracks. So yes, I put some blame for that on Steinberg. Now that DP has PreGen it performs better than Logic, before it was slightly behind, but not near as bad at low latencies as Cubase.

I also said that all the "buffer tricks" that people have leveled against Logic and apparently OS X, are not necessarily due to OS X, but to DAWs evolving, I'm sticking by that, every DAW does this these days, on every OS, because it works, we all get higher track counts at lower buffer settings on armed tracks.

This line strikes me though.
Quote:
This is due to OSX aggressive speed step.
So, you realize of course that Apple can only do so much about Intels speed stepping, that Windows has had roughly 15 more years than Apple at dealing with Intel, and that this is exactly the type of reason that Apples move to Arm chips should be seen as a possible boon? Plus the fact that Intel has a much more vested interest in Windows in terms of their chips, with probably 9 times as many chips sold to machines with Windows as the main OS.

In around 2009 I had a serious issue with Kontakt and a huge dropout when a song in Live got too busy, and it turned out to be due to a hardware shortcoming of the particular Intel dual 2.4 ghz. chip. It was aggressively speed stepping down, and couldn't ramp up fast enough. The workaround was to use a little utility that fooled the chip into thinking CPU was being used all the time. These types of problems hopefully will be a thing of the past.

I mean, what's the point of being a company that supposedly makes the software and hardware as an integrated package as the supposed benefit of using that software, then integrating a chip that has a vastly more lucrative deal with your chief OS rival? I think beyond the money aspects of this, the whole thing makes sense as far as their "ecosystem" approach.
Old 4th August 2020
  #1099
Lives for gear
Quote:
Originally Posted by juiseman View Post
Not to mention A lot Audio interfaces that are not USB class Compliant...RIP...well you can guess what will happen there. I'm glad to hear someone with the same concerns as myself. I keep bringing up how protools will handle this....well..lol they might just "not" handle it. I can see backwards
compatibility being an issue. They may have to make a translation app to open old projects.

This transition will take some time. My guess is the continued support for Intel based mac's will have to be longer than the PPC-Intel transition.
So that's somewhat a relief for the guys who got those 2019 Mac Pro's

There has been lot of talk about how great these Apple Soc's will be...and that is great!! (I'm holding my excitement back) my view is the software is going to be the biggest hurdle. I think folks forget how many months i takes 3rd party software guys to catch up with the normal mac os updates; now, then let's toss an architecture change in there too, plus I guess were still in the middle of a pandemic?, so there is that.. we can't expect the smaller software developers to magically rewrite, test, verify, debug, troubleshoot...ect their plugs while eating the cost at a snap of our fingers....
I could be wrong.. it happens a lot; and I'm not afraid to admit it.
What about stuff like UAD DSP-card? Now obsolete?
Old 4th August 2020
  #1100
Lives for gear
 
juiseman's Avatar
Most of the speed-step issues can be attributed to the bad thermal performance on all macs (minus the mac pro's).. all macs are within the TDP specs for the CPU's.
realistically, you could overclock the mac pro 5,1 & 7,1..but of course they would never allow that. No way would that ever work with imac ,mac mini, mac book..
I'm certain this was a big driver behind the switch to Apple SOC; Apple makes product's thin and light. Intel CPU's have run hotter and hotter since 2014.
This is the only way they CAN keep their products with the sleek design that people love. Control the heat!! that means drop Intel. On a side note, AMD would have been a great option...just sayin...

Thin and lite computers have never been important in a workstation class computer. If your into building your own computers (which a ton of guys are doing) a quick 2 min guide on case air flow and cpu cooling will tell you that open air (more space, bigger case) and direction of air flow are key to a cool and quiet setup. I think thats why guys loved their hackntosh's so much; the great user experience of OSX in the power house hardware of the PC...best of both worlds.

https://www.theverge.com/2020/6/22/2...hips-wwdc-2020

Not sure if anyone has seen this yet; but its the best sum up of the whole thing I've seen so far. I want to see that A12Z performance in a Arm version of Logic.
Old 4th August 2020
  #1101
TNM
Lives for gear
Quote:
Originally Posted by kludgeaudio View Post
ARM is a good choice because it is a clean architecture that is designed for efficiency. The x86 is an ugly architecture that has been made fast mostly through crazy optimizations and extremely fast physical hardware. If you were to build an ARM on Intel's fastest process, it would be faster than anything Intel is currently selling.

It's important to split the three parts: the instruction set architecture, the implementation architecture, and the process. In this case, Apple is changing all three of them, and the first one is going to be a win immediately. The second one will take some time to improve and the third one may never be a win. But the first one is a big step up.
--scott
well I'll re consider my opinion when there are arm chips that can run 512+ track orchestral projects out there with kontakt on every track, and when at least 99% of plugins are compatible. Simple. It might be a better architecture but that's not what i really meant whatsoever. Why is it a good move for US daw users, in what way will it benefit us, especially those on mac who don't use Logic? I see years of pain ahead.
Old 4th August 2020
  #1102
TNM
Lives for gear
Quote:
Originally Posted by juiseman View Post
Most of the speed-step issues can be attributed to the bad thermal performance on all macs (minus the mac pro's).. all macs are within the TDP specs for the CPU's.[/url]
Not totally true.. OSX is broken at a fundamental level and I have proven this over and over again. Yes, there are some macs out there that are thermally constrained.. But my 8 core imac pro is at 48 or 50 degrees when it suddenly drops from 4ghz to 3.2 and that's what causes all sorts of performance issues with DAWs. The 16" macbook pro was under 60 degrees when it dropped from 3.8 to 2.8, a whole gigahertz, and guess what, DAW overload!

standard imacs with the 8 core are very thermally constrained and I guess minis are too.. mac pros of course as you yourself said would be beasts.

In windows, my imac pro never, EVER drops below the 4ghz all core turbo if high performance profile is enabled. This is what makes such a massive difference between pro tools/reaper windows vs OSX on the very same machine. And I have never seen the temps leave the 60's in DAW use. So sure, let's say thermal issues on some macs, but combined with an OS that has overly aggressive frequency hopping. It's not really even debatable as I have tested it myself to death on 7 different macs of different eras in the last 8 years. I always note the temp when the dropping occurs and I only count the times when the temps are well under throttle point..

PS the current mac pro has 20% belter performance in windows.. and it ain't due to thermals!
Old 4th August 2020
  #1103
Lives for gear
 
juiseman's Avatar
Quote:
Originally Posted by TNM View Post
Not totally true.. OSX is broken at a fundamental level and I have proven this over and over again. Yes, there are some macs out there that are thermally constrained.. But my 8 core imac pro is at 48 or 50 degrees when it suddenly drops from 4ghz to 3.2 and that's what causes all sorts of performance issues with DAWs. The 16" macbook pro was under 60 degrees when it dropped from 3.8 to 2.8, a whole gigahertz, and guess what, DAW overload!

standard imacs with the 8 core are very thermally constrained and I guess minis are too.. mac pros of course as you yourself said would be beasts.

In windows, my imac pro never, EVER drops below the 4ghz all core turbo if high performance profile is enabled. This is what makes such a massive difference between pro tools/reaper windows vs OSX on the very same machine. And I have never seen the temps leave the 60's in DAW use. So sure, let's say thermal issues on some macs, but combined with an OS that has overly aggressive frequency hopping. It's not really even debatable as I have tested it myself to death on 7 different macs of different eras in the last 8 years. I always note the temp when the dropping occurs and I only count the times when the temps are well under throttle point..

PS the current mac pro has 20% belter performance in windows.. and it ain't due to thermals!
I think were saying the same thing; I'm just not very good at explaining correctly, or making my point. I think they further limit frequency to control thermals beyond what Intel states as the power requirement. Intel does this a lot, a 95W CPU actually pulling way over 95W on a non-overclocked system. I'm not talking about just AVX-512 stuff either. the OS/Firmware can obviously
control further. It could be baked into Mac OS like your saying regardless of the system or CPU. Why? well, the thermal performance is bad on Mac's (except MP)

https://www.thebroadcastbridge.com/u...Figure_5-1.png

This is baked into the CPU

Last edited by juiseman; 4th August 2020 at 06:24 PM.. Reason: link added
Old 4th August 2020
  #1104
Lives for gear
 
juiseman's Avatar
Quote:
Originally Posted by TNM View Post
well I'll re consider my opinion when there are arm chips that can run 512+ track orchestral projects out there with kontakt on every track, and when at least 99% of plugins are compatible. Simple. It might be a better architecture but that's not what i really meant whatsoever. Why is it a good move for US daw users, in what way will it benefit us, especially those on mac who don't use Logic? I see years of pain ahead.
I say this as optimistic as I can; it may be awhile sir.
Old 4th August 2020
  #1105
Lives for gear
 

Quote:
Originally Posted by TNM View Post
well I'll re consider my opinion when there are arm chips that can run 512+ track orchestral projects out there with kontakt on every track, and when at least 99% of plugins are compatible. Simple. It might be a better architecture but that's not what i really meant whatsoever. Why is it a good move for US daw users, in what way will it benefit us, especially those on mac who don't use Logic? I see years of pain ahead.
On the software side, yes, there will be years of transition ahead. This is a definite.

But as far as long term benefits, Apple SoCs will be able to match or beat the performance of Intel chips like the 10th Gen 15 or 9th Gen i9, for like, HALF the same heat generated and power consumption. (The A12Z in the transition kit already matches the 10th Gen i5, and the A14 and A14X coming this year are expected to be even more powerful, thanks to TSMC's 5 nm process node).

In practice, this means possibly dead silent Macs across the board. No power-limiting from hitting the thermal ceiling. Way better battery life in Macbook Pros; the ability to actually do work in a DAW remotely, if needed, and not wreck your battery completely. Possibly lower latency as well, since SoCs put every part of the system right on the same chip. The high-powered and low-powered cores in Apple's design will also mean even better multi-tasking on your system while running a DAW.

These are the possibilities just from knowing what advantages Apple Silicon offers inherently.

Fan noise is a big one for audio. We all know Apple has had a life-long hatred of fan noise, going back to the first Macintosh. The new Mac Pro is DEAD SILENT (no thanks to the Intel CPU, but thanks to Apple's design). This is obviously something Apple is still thinking about, and the low power consumption and heat of Apple Silicon, with the same or better performance, offers them a way to make the rest of the Mac lineup dead silent, except in the most intense processing scenarios. I don't see why Apple wouldn't take advantage of that.
Old 4th August 2020
  #1106
Gear Nut
 

Quote:
Originally Posted by TNM View Post
well I'll re consider my opinion when there are arm chips that can run 512+ track orchestral projects out there with kontakt on every track, and when at least 99% of plugins are compatible. Simple. It might be a better architecture but that's not what i really meant whatsoever. Why is it a good move for US daw users, in what way will it benefit us, especially those on mac who don't use Logic? I see years of pain ahead.
From what I can remember from every other transition till now, everyone said the same thing. Then people waited x amount of time until their stuff was ready to go, and stuck with what they had till then. There's not much reason to think this time will be any different at all.
Old 4th August 2020
  #1107
Lives for gear
 
crufty's Avatar
Quote:
Originally Posted by MattyJoe View Post
On Apple SoCs will be able to match or beat the performance of Intel chips like the 10th Gen 15 or 9th Gen i9, for like, HALF the same heat generated and power consumption.
So that would be very sweet if true. However...how are you able to make these claims???

Everything I know of ARM goes thusly:
1) great for low power
2) Great at integer math
3) ok (at best) with float math
4) trash when parallel compute is required

Intel’s chipset is so strong in the areas DAWs require and AMD is no slouch.

What evidence is there that ARM can really perform?
Old 4th August 2020
  #1108
Gear Addict
Quote:
Originally Posted by TNM View Post
and that's just fine...
but don't then complain if something doesn't work as well as on windows, or even work at all (i am not saying you are complaining I am saying others here that have posted), when it's apple's fault and not the DAW developer's.
Heh. I have found nothing so far that works better on Windows than on Mac OS outside of gaming. Guess I'm lucky I've not had any audio software issues as of yet on the Mac.
Old 5th August 2020
  #1109
Lives for gear
 
Lady Gaia's Avatar
Quote:
Originally Posted by crufty View Post
What evidence is there that ARM can really perform?
The SoCs Apple has been using in their devices for years, for which there are plenty of available benchmarks? Comparing this Intel Core i9-9900KS system's results again this phone's results you'll see that the iPhone gets about 75% of the Intel CPU's single-core integer score, and about 78% of the floating point score while running at 68% of the clock speed. So it's doing startlingly well considering it's running off a battery in a passively cooled device where thermal considerations and battery life are major restrictions.

This iPad's results from a year before show what happens when you have a little more battery, still passively cooled, and can afford a larger die. It scales fairly shockingly well with only four performance cores.

Is this directly competitive? No, but it's not what they're likely to ship in a Mac, either. Bump it up to eight performance cores (which is what the rumor mill suggests is the initial target), raise the clock speed with active cooling, and add another year's iteration on the design and it's not inconceivable that it'll be very competitive.

Will it work out for Apple as a high performance competitor? Nobody who knows for sure can say one way or another, but I think it's foolish to assume they didn't already know the answer before they publicly committed to the move. We'll know a whole lot more by the end of the year when they announce, and likely even ship, new hardware based on their own design.
Old 5th August 2020
  #1110
Lives for gear
Quote:
Originally Posted by crufty View Post
So that would be very sweet if true. However...how are you able to make these claims???

Everything I know of ARM goes thusly:
1) great for low power
2) Great at integer math
3) ok (at best) with float math
4) trash when parallel compute is required

Intel’s chipset is so strong in the areas DAWs require and AMD is no slouch.

What evidence is there that ARM can really perform?
None. At least not for people without NDA's. We know what Arm from Arm can do. We know what apple can do with their phones. The rest is marketing bull****, hype or guesses. There are leaked results from developer kits that is based on old socs and the measurement is a unspecified benchmark score. Apple don't what us to know, so we don't.
📝 Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 4589 views: 578011
Avatar for stixstudios
stixstudios 20th May 2020
replies: 4920 views: 282415
Avatar for TJT
TJT 1 minute ago
replies: 96 views: 7212
Avatar for juiseman
juiseman 1 week ago
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
🖨️ Show Printable Version
✉️ Email this Page
🔍 Search thread
🎙️ View mentioned gear
Forum Jump
Forum Jump