Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
There are 2 types of Mac Pro users. The pragmatists and the true believers.

Pragmatists care about performance. That is why AMD processors keep coming up. We didn't come here because of P.T. Barnum and his reality distortion field. We went Mac because it was the best performance option. Those of us that have been circling the airport, waiting on a new Mac Pro have decisions to make. As a pragmatist, that means we need to collect data, so we can make an informed decision.

I am sure that it is irrelevant to the true believers and you don't care. That being said, those of us that are pragmatic are looking at all of our options. Do I stay or do I go (with a tip of the hat to Joe Strummer), do we make do with what we have, is there a 7,1 that actually fits our workflow, or do we move onto greener pastures (with more performance).

Decisions, decisions.

Wrong. Performance freaks care about performance above all else, not pragmatists. For example: the TB3 stability issues that are well-known when you are off the pure Intel platform. A pragmatist needs things to WORK, and not have to worry about compatibility just to eke out a tiny bit more performance or save a little money. Also, there is nothing pragmatic about endlessly speculating about a processor option that doesn't exist; it's just wasting time.

So no, this isn't about pragmatism or a "reality distortion field." It's actually about reality: The reality that the only processor options that exist. Speculation about an AMD processor in a Mac Pro is a total waste of time--that's the true "reality distortion field" here.
 
Last edited:
Wrong. Performance freaks care about performance above all else. Note the stability of TB3 issues that are well-known when you are off the pure Intel platform, for example. A pragmatist needs things to WORK, not worry about compatibility just to eke out a tiny bit more performance. Also, there is nothing pragmatic about endlessly speculating about a processor option that doesn't exist; it's just wasting time.

So no, this isn't about pragmatism or a "reality distortion field." It's actually about reality: The reality that the only option that exists. Speculation about an AMD processor is a total waste of time with regards to the Mac Pro--that's the true "reality distortion field" here.

😂
 
  • Like
Reactions: 09872738
Thank you for posting these benchmarks on the 16-core model, as this is the processor I was leaning towards to replace my 2012 Mac Pro (2 x 2.66 Ghz 6-Core Intel Xeon). I'm leaning towards the same GPU but less installed RAM and smaller system drive, around $11K.

I'm mainly interested in the Cinebench score as Cinema 4D is my main tool for our small motion graphics/3D animation business.

Is this the Cinebench R20 score? My 2012 Mac Pro gets a score of 2641 on that, so it looks like I could be getting a 2.5x C4D render speed boost on a new system. I'd love to get the 24-core processor, but the $4000 jump to get that is a bit too much to justify.

Yes, some PC running AMD Threadrippers would be cheaper and faster in terms of raw processing power, but there's more to my work life than that. In short, I really really dislike working in Windows. The interface is gross. It's poorly organized. It won't let me sort all files and folders by date the way I can on my Mac.

We're currently in possession of a 2 x 12 core Xeon Windows workstation with a Cinebench 20 score of 8809 that a client gave to us for a project in 2016 (and never asked for its return), and I use it only sporadically. In terms of raw CPU power it's 3.3x as fast as my 2012 Mac Pro, but I'd rather do the bulk of my animation and compositing on my slower Mac Pro than have to deal with Windows 10. When it comes time for my final animation renders I send those to Pixel Plow.

The Mac in general has NEVER been the best platform for 3D animation, but that hasn't stopped my wife and me from running a fairly successful small animation business that celebrated its 20th anniversary in 2019. We're not rich, but we're comfortable, and we can justify a new $11K Mac Pro after 7-8 years of no workstation upgrades.
Yes, Cinebench R20 score
 
  • Like
Reactions: WShawn
Wrong. Performance freaks care about performance above all else, not pragmatists. For example: the TB3 stability issues that are well-known when you are off the pure Intel platform. A pragmatist needs things to WORK, and not have to worry about compatibility just to eke out a tiny bit more performance or save a little money. Also, there is nothing pragmatic about endlessly speculating about a processor option that doesn't exist; it's just wasting time.

So no, this isn't about pragmatism or a "reality distortion field." It's actually about reality: The reality that the only processor options that exist. Speculation about an AMD processor in a Mac Pro is a total waste of time--that's the true "reality distortion field" here.
Why are you in this thread? You don't really know very much about computers or processors. I mean at least the ones being discussed; that's apparent. You can't even be bothered to research and understand Threadripper iterations. You're obviously incapable of recognizing that there has been a paradigm shift in computing due to technological advancements. You do like to argue for the sake of arguing. You post bogus benchmarks and then cry when people call you out on it....eventually moving the goal posts to "only performance freaks care about performance." which is just utterly disingenuous when considering you're in a thread about raw performance.
 
Last edited:
Do I stay or do I go (with a tip of the hat to Joe Strummer), do we make do with what we have, is there a 7,1 that actually fits our workflow, or do we move onto greener pastures (with more performance).

Decisions, decisions.
The old saying the grass is always greener springs to mind. One problem solved another three discovered.

I personally think its simpler than you suggest. There are those of us (potential Mac Pro buyers) that want to be on the cutting edge and the cost of the Mac Pro doesn’t sit well with them. Then there are those who just want the fastest Mac they can buy.
 
  • Like
Reactions: chfilm
You post bogus benchmarks and then cry when people call you out on it....eventually moving the goal posts to "only performance freaks care about performance." which is just utterly disingenuous when considering you're in a thread about raw performance.
Moving the goal posts???

I thought this discussion was about Mac Pro performance to help those interested in buying one rather than the rather unhelpful ”The sky is falling in!” discussion its descended in to if we use a new Mac Pro for anything work related over a AMD CPU’d workstation.
 
Wrong. Performance freaks care about performance above all else, not pragmatists. For example: the TB3 stability issues that are well-known when you are off the pure Intel platform. A pragmatist needs things to WORK, and not have to worry about compatibility just to eke out a tiny bit more performance or save a little money. Also, there is nothing pragmatic about endlessly speculating about a processor option that doesn't exist; it's just wasting time.
Do you mean like people here do with ARM? :D
 
Wrong. Performance freaks care about performance above all else. Note the stability of TB3 issues that are well-known when you are off the pure Intel platform, for example. A pragmatist needs things to WORK, not worry about compatibility just to eke out a tiny bit more performance. Also, there is nothing pragmatic about endlessly speculating about a processor option that doesn't exist; it's just wasting time.

So no, this isn't about pragmatism or a "reality distortion field." It's actually about reality: The reality that the only option that exists. Speculation about an AMD processor is a total waste of time with regards to the Mac Pro--that's the true "reality distortion field" here.

Performance freaks care about performance above all else
- well said Captain Obvious. This is a Mac Pro forum, not a mac mini forum.

I own 2 4,1 (flashed to 5,1 - see sig) Tell me why I would care about Thunderbolt.

I agree, a pragmatist needs things to WORK.

Now - go look at most of the active threads in this forum. It is a whole lot of Does this product work in OSX? And it isn't just the folks that are using vintage hardware either. I shouldn't be trying to figure out which RX 580 works with a Mac Pro - I should be able to just buy one off the shelf and stick it in my Mac Pro. Pity it doesn't actually work like that.

In computing - there are many options - there are even more if you are technically literate. I was that long before I moved to OSX, and if I move away from it. If you choose to limit yourself, that is up to you.

But I need performance. A 40 - 50% increase in render speeds matters to me, and that is what I am looking at. That isn't my definition of a "tiny increase".

Your "reality" is built around limiting yourself to the 7,1, OSX, and shoehorning a workflow around that. I start at the other end and build the hardware out from that.
 
Why are you in this thread? You don't really know very much about computers or processors. I mean at least the ones being discussed; that's apparent. You can't even be bothered to research and understand Threadripper iterations. You're obviously incapable of recognizing that there has been a paradigm shift in computing due to technological advancements. You do like to argue for the sake of arguing. You post bogus benchmarks and then cry when people call you out on it....eventually moving the goal posts to "only performance freaks care about performance." which is just utterly disingenuous when considering you're in a thread about raw performance.

Sorry, wrong. The ad hominem attack is what you resort to when you can't argue on the merits. BTW, I have over 25 years in technology, including scientific computational computing, two stints as the CTO of major national and multinational corporations. I've managed massive data center operations and all of that has been with non-Mac systems, including both AMD and Intel processor systems. So I'm pretty comfortable with my qualifications.

My points remain: This is a Mac forum, discussing Mac Pro benchmarks. Needless flogging of an AMD thread ripper option is completely pointless. Moreover, when *actual* benchmarks show minor performance advantages for certain tasks the argument of their vast superiority becomes moot anyway.
 
Now - go look at most of the active threads in this forum. It is a whole lot of Does this product work in OSX?

Everything you said is important. I just wanted to highlite this specific part though because this is actually a huge issue. Since there hasn't been a Mac that could take expansion cards since just before the 2013 Model released most hardware vendors have not provided drivers for macOS for any of their PCIe expansion cards.

This was very well illustrated in the recent Snazzy Labs video where he tried a bunch of cards and found many didn't work. Now we could argue that over time this situation will get better now that a new Mac Pro is here but we could also say that things probably wont change due to how niche the Mac Pro is due to its price point.

If you're making a sound card or something like that it may not be worth the development effort if you're not going to sell many to Mac users due to the small market the Mac Pro will have.

Now if Apple also released a mid-sized tower with more consumer orientated parts and 2-3 expansion slots then I could see that changing.

So while on the PC in general you have some Thunderbolt 3 issues, there is a massive ecosystem of add-in cards available. Personally I think the PCIe cards are more important on a desktop. I have Thunderbolt 3 support on my Windows PC and I don't use it but I do on my laptop where docks and things make perfect sense for that form factor.
 
  • Like
Reactions: ssgbryan
Performance freaks care about performance above all else - well said Captain Obvious. This is a Mac Pro forum, not a mac mini forum.

I own 2 4,1 (flashed to 5,1 - see sig) Tell me why I would care about Thunderbolt.

I agree, a pragmatist needs things to WORK.

Now - go look at most of the active threads in this forum. It is a whole lot of Does this product work in OSX? And it isn't just the folks that are using vintage hardware either. I shouldn't be trying to figure out which RX 580 works with a Mac Pro - I should be able to just buy one off the shelf and stick it in my Mac Pro. Pity it doesn't actually work like that.

In computing - there are many options - there are even more if you are technically literate. I was that long before I moved to OSX, and if I move away from it. If you choose to limit yourself, that is up to you.

But I need performance. A 40 - 50% increase in render speeds matters to me, and that is what I am looking at. That isn't my definition of a "tiny increase".

Your "reality" is built around limiting yourself to the 7,1, OSX, and shoehorning a workflow around that. I start at the other end and build the hardware out from that.

Please show me a benchmark that shows a 40-50% increase in render speed for one of the new W-32xx series Xeons vs. an AMD processor. Also please make sure it's consistent across the board, and not just one application.
 
Please show me a benchmark that shows a 40-50% increase in render speed for one of the new W-32xx series Xeons vs. an AMD processor. Also please make sure it's consistent across the board, and not just one application.

The 28 Core XEON and the 32 Core Threadripper are quite close in video editing rendering. But you can purchase a 32 Core Threadripper for the price of the 8 Core XEON. Thus obtaining 4x higher performance for the same price.

Or you could purchase two Threadripper 32 Core systems for the price of one Mac Pro 28 Core system and do distributed rendering if your software supports that (I don't think many do).

Then next year, in about 2-3 months AMD is releasing a 64 Core Threadripper which is socket compatible with the 32 Core parts. A doubling of performance. A system configured with that chip would cost about $7,500 and would for sure double the 28 Core Intel in rendering.

None of this takes into account the Afterburner card though which I think is real innovation from Apple and I'm glad they did it. I don't think even the 64 Core Threadripper could compete with that card in Final Cut. Of course it is a $2,000 premium but we're talking maximum obtainable performance after all and $2,000 for what that card can do seems like a bargain to me.
 
The 28 Core XEON and the 32 Core Threadripper are quite close in video editing rendering. But you can purchase a 32 Core Threadripper for the price of the 8 Core XEON. Thus obtaining 4x higher performance for the same price.

Or you could purchase two Threadripper 32 Core systems for the price of one Mac Pro 28 Core system and do distributed rendering if your software supports that (I don't think many do).

Then next year, in about 2-3 months AMD is releasing a 64 Core Threadripper which is socket compatible with the 32 Core parts. A doubling of performance. A system configured with that chip would cost about $7,500 and would for sure double the 28 Core Intel in rendering.

None of this takes into account the Afterburner card though which I think is real innovation from Apple and I'm glad they did it. I don't think even the 64 Core Threadripper could compete with that card in Final Cut. Of course it is a $2,000 premium but we're talking maximum obtainable performance after all and $2,000 for what that card can do seems like a bargain to me.

Right, that's all true, but now we're talking price vs. performance, and those are two different things. It's been pointed out repeatedly, none of the major workstation vendors (big corporations, not the small guys) offer AMD workstations either. So everyone is charging the premium price Intel is charging--and clearly there are customers willing to pay that. As to why, there are many reasons, and it's not worth speculating why. But in terms of pure raw performance, the actual difference is pretty negligible even excluding things like the Afterburner card.

I think half of the angst here would be resolved if Apple had a mid-range expandable system but I don't see that ever happening. They haven't done that in ages and their business hasn't really suffered. The reason is that market has been shrinking on for ages. About the only people who buy expandable PCs these days are gamers and corporations and neither one of those markets are ones Apple has ever been a viable option in.
 
I think half of the angst here would be resolved if Apple had a mid-range expandable system but I don't see that ever happening. They haven't done that in ages and their business hasn't really suffered. The reason is that market has been shrinking on for ages. About the only people who buy expandable PCs these days are gamers and corporations and neither one of those markets are ones Apple has ever been a viable option in.
Their business is phones, Macintosh is no longer their core business.
 
Performance freaks care about performance above all else - well said Captain Obvious. This is a Mac Pro forum, not a mac mini forum.

I own 2 4,1 (flashed to 5,1 - see sig) Tell me why I would care about Thunderbolt.

I agree, a pragmatist needs things to WORK.

Now - go look at most of the active threads in this forum. It is a whole lot of Does this product work in OSX? And it isn't just the folks that are using vintage hardware either. I shouldn't be trying to figure out which RX 580 works with a Mac Pro - I should be able to just buy one off the shelf and stick it in my Mac Pro. Pity it doesn't actually work like that.

In computing - there are many options - there are even more if you are technically literate. I was that long before I moved to OSX, and if I move away from it. If you choose to limit yourself, that is up to you.

But I need performance. A 40 - 50% increase in render speeds matters to me, and that is what I am looking at. That isn't my definition of a "tiny increase".

Your "reality" is built around limiting yourself to the 7,1, OSX, and shoehorning a workflow around that. I start at the other end and build the hardware out from that.
40-50% performance increase! 😮

Ok, if this is all true and the AMD CPU numbers do look like 50% improvements how about you start a thread educating all us would be Mac Pro buyers to the real world results you claim.

If your goal really is to educate I’d be very interested in these somewhat revolutionary real world benchmarks against a Mac Pro. It could actually turn into a rather helpful resource.

Please don’t take my response as hostile, I’m truly interested in the findings and may help ebb the flow if ill feelings in what should be a Mac Pro performance discussion.
 
  • Like
Reactions: chfilm
I finally have my Mac Pro updated to where I want with the extra RAM I ordered so this morning I ran the Puget Systems Photoshop benchmark. At the moment, it's the only "use case" benchmark vs. synthetic benchmark that I could find that a) works cross-platform and b) has recent data from other systems to compare to. My results in the graphic attached, Puget's most recent data at the following link:


A few points of comparison: My overall score of 926.2 is less than 5% lower than the highest overall score which went to one of the newest AMD processors, the AMD Ryzen 9 3950X 16-core. Impressive performance for the AMD at it's price point but it's also a consumer chip and not entirely comparable (there are limitations in consumer chips). On the other hand, given that it has a almost 10% higher base clock speed, almost 7% higher turbo clock speed and 92% more cache (73MB vs. 38MB), one could argue that the only 5% total performance increase is a bit underwhelming. But the price! Very inexpensive, so kudos AMD (and a very low TDP, thanks to that 7nm).

On the "General Score" my system actually gets the highest score of any of the Puget Systems: 115.3, vs the top score of 110.3 for the AMD TR 3960X 24-core--and with 33% fewer cores, too boot.

On the filter score, I only got a 76.7, only good enough for 5th place, but the top score is only 81.1, so a little less than 6% different--and the top score in that category goes to an Intel 9900K 8-core anyway, not an AMD chip. The best AMD chip is the Ryzen 9 3950X 16-core, at 79.9, or just about 4% better than my system.

The Photomerge score I am getting is bizarrely low. In fact, without that being so low my system may have been the top overall performer. I suspect something is wrong here as it scores a 79.1, so low it's off the bottom of the chart. On the bright side, I never use Photoshop to merge photos, although I don't know if this is actually a Photoshop issue or something in Puget's benchmark, which is still listed as a "beta."

Finally the GPU score; they use the same Nvidia RTX 2080 Ti card with 11GB of GDDR6 memory vs. my system, which has the Radeon Vega II with 32MB of HBM2 memory. Scores are tight there, not surprising. My 99.8 score is the third highest behind an Intel 9900K and an AMD Ryzen 7 system, but not sure we can read much overall into this score. The differences across the top ten scores are negligible and of course Photoshop doesn't leverage the GPU as heavily as many other applications do.

Some caveats: It seems that they may have run all of their benchmarks on version 20 of Photoshop; I am on version 21.0.2. I could downgrade just to run the tests, but the oldest version I can install is actually 20.0.8, so I'm not sure what difference it would make. It could be why the photo merge score is odd though. Also, Photoshop is not very good at all in terms of using multiple cores/threads. That means the raw speed of the CPU and overall CPU/system architecture have a greater impact than core number. This was obvious as I ran the test as the CPU monitor showed the system barely doing anything....CPU usage rarely spiked to more than 50%, and typically was way down in the 10% range. Same with turbo boost, it mostly sat at the base Ghz and I only saw it spike as high as 4.1Ghz briefly.

So, a real world benchmark of a sort. Performance of the Mac Pro seems to be excellent, on par with the best systems Puget lists, with the exception of that off Photomerge score. I'm going to email Puget about that, as it seems like it has to be bug. Would an AMD processor be cheaper? Yes. Are they "40-50% better" as we've seen claimed here? Certainly not in this benchmark.

Meanwhile, if you intend to use Photoshop heavily with your Mac Pro, these results might be interesting and useful to you.

PsBenchResults_12-31-9-37.jpg
 
I finally have my Mac Pro updated to where I want with the extra RAM I ordered so this morning I ran the Puget Systems Photoshop benchmark. At the moment, it's the only "use case" benchmark vs. synthetic benchmark that I could find that a) works cross-platform and b) has recent data from other systems to compare to. My results in the graphic attached, Puget's most recent data at the following link:


A few points of comparison: My overall score of 926.2 is less than 5% lower than the highest overall score which went to one of the newest AMD processors, the AMD Ryzen 9 3950X 16-core. Impressive performance for the AMD at it's price point but it's also a consumer chip and not entirely comparable (there are limitations in consumer chips). On the other hand, given that it has a almost 10% higher base clock speed, almost 7% higher turbo clock speed and 92% more cache (73MB vs. 38MB), one could argue that the only 5% total performance increase is a bit underwhelming. But the price! Very inexpensive, so kudos AMD (and a very low TDP, thanks to that 7nm).

On the "General Score" my system actually gets the highest score of any of the Puget Systems: 115.3, vs the top score of 110.3 for the AMD TR 3960X 24-core--and with 33% fewer cores, too boot.

On the filter score, I only got a 76.7, only good enough for 5th place, but the top score is only 81.1, so a little less than 6% different--and the top score in that category goes to an Intel 9900K 8-core anyway, not an AMD chip. The best AMD chip is the Ryzen 9 3950X 16-core, at 79.9, or just about 4% better than my system.

The Photomerge score I am getting is bizarrely low. In fact, without that being so low my system may have been the top overall performer. I suspect something is wrong here as it scores a 79.1, so low it's off the bottom of the chart. On the bright side, I never use Photoshop to merge photos, although I don't know if this is actually a Photoshop issue or something in Puget's benchmark, which is still listed as a "beta."

Finally the GPU score; they use the same Nvidia RTX 2080 Ti card with 11GB of GDDR6 memory vs. my system, which has the Radeon Vega II with 32MB of HBM2 memory. Scores are tight there, not surprising. My 99.8 score is the third highest behind an Intel 9900K and an AMD Ryzen 7 system, but not sure we can read much overall into this score. The differences across the top ten scores are negligible and of course Photoshop doesn't leverage the GPU as heavily as many other applications do.

Some caveats: It seems that they may have run all of their benchmarks on version 20 of Photoshop; I am on version 21.0.2. I could downgrade just to run the tests, but the oldest version I can install is actually 20.0.8, so I'm not sure what difference it would make. It could be why the photo merge score is odd though. Also, Photoshop is not very good at all in terms of using multiple cores/threads. That means the raw speed of the CPU and overall CPU/system architecture have a greater impact than core number. This was obvious as I ran the test as the CPU monitor showed the system barely doing anything....CPU usage rarely spiked to more than 50%, and typically was way down in the 10% range. Same with turbo boost, it mostly sat at the base Ghz and I only saw it spike as high as 4.1Ghz briefly.

So, a real world benchmark of a sort. Performance of the Mac Pro seems to be excellent, on par with the best systems Puget lists, with the exception of that off Photomerge score. I'm going to email Puget about that, as it seems like it has to be bug. Would an AMD processor be cheaper? Yes. Are they "40-50% better" as we've seen claimed here? Certainly not in this benchmark.

Meanwhile, if you intend to use Photoshop heavily with your Mac Pro, these results might be interesting and useful to you.

View attachment 885864
Thanks for your effort, finally someone who just answered what this thread was meant for!
 
  • Like
Reactions: sirio76 and gazwas
Just to update...it occurred to me that I hadn't yet launched Photoshop to check all of the settings, and I know that Camera Raw has it's own performance settings and the Photomerge test uses a bunch of RAW files so I checked the settings and sure enough, it was set to "Auto" for the graphics acceleration in Camera Raw so I set that to use full GPU acceleration just like Photoshop is set up, and ran it again. New results are better across the board, but the Photomerge score is oddly unaffected. Something is not right with that benchmark I think.

The new overall score is now 947.2, only 2.4% off the top score. The general score has jumped to 120.4, putting it 9% above every other system Puget has tested to date. The filter score of 77 is still good for 5th place, but the differences between systems on that measure is very small--it's only 5% off the top score (held by Intel i9 9900K). Then the anomalous Photomerge score, which actually went down by 0.3--probably just random variation, but clearly something odd there. Finally the GPU score jumped to 105.7, now good enough for first place in that category as well.

Given that the Photomerge score is fully 20% of the total score, if I am right and there's a bug in that test then this system could see a big boost in it's already high total score. I'm going to contact Puget and see what they say.

PsBenchResults_12-31-12-14 2.jpg

[automerge]1577820290[/automerge]
Oh, and BTW...those Adobe scores on Photomerge, if accurate, show how behind Adobe is on some things. I use Helicon Focus for focus stacking images. On my new Mac Pro I can load 30 61-megapixel DNG files and Helicon Focus will align and stack them in 6 seconds flat. Its really uses the processors to the max as well the GPU compute engine. On the processor side, it leverages the AVX-512 instruction set which has been shown time and again to be extremely helpful.

While running these tests I continued to watch the results, and my Mac Pro was basically snoozing through them. Max CPU spiked to 4.1Ghz, and rarely. Watching the cores, it basically ran everything on the first three cores, with the fourth getting less use and the fifth even less. All of the other 11 cores were basically flatlined all of the time doing nothing, except for brief spikes of 10-15% use. With the Activity Monitor processor window open, none of the virtual cores ever saw *any* use. Max RAM use was very minimal. It never got beyond 40% of the GPU RAM and never more than 10% of the GPU processor use. It would be nice if Adobe actually put some effort into performance optimizations. While many tasks don't parallelize well, many do. It was fine in the past, but it's been obvious for years now that multiple cores are the future, and now we're seeing 32 cores in *consumer* level processors like the Ryzen.
 
Last edited:
Thank you for posting your PugetBench score. Finally, some baseline comparisons! Can you also post up your full configuration?

Something is kinda weird about your Photomerge 45MP number, but I doubt it's Photoshop itself. I ran the test on my work PC using the latest version of PS (21.0.2) and got the results below. As you can see the Photomerge score comparison between our computers just doesn't make any sense, given the other score deltas.

It can't be a GPU issue since Puget's testing shows that GPU has no bearing on the result: https://www.pugetsystems.com/labs/articles/Photoshop-CC-2018-NVIDIA-GeForce-GPU-Performance-1139/

Maybe someone else with a 7,1 MP can run the bench and see if it's reproducible.

(I built this thing for about 2300 CAD [1770 USD] two years ago from consumer parts and it's been serving me well.)
 

Attachments

  • BenchResults.jpg
    BenchResults.jpg
    250.2 KB · Views: 118
  • Annotation 2019-12-31 145806.png
    Annotation 2019-12-31 145806.png
    24.1 KB · Views: 140
  • Annotation 2019-12-31 145814.png
    Annotation 2019-12-31 145814.png
    19.7 KB · Views: 110
  • Annotation 2019-12-31 145817.png
    Annotation 2019-12-31 145817.png
    37.7 KB · Views: 127
Thank you for posting your PugetBench score. Finally, some baseline comparisons! Can you also post up your full configuration?

Something is kinda weird about your Photomerge 45MP number, but I doubt it's Photoshop itself. I ran the test on my work PC using the latest version of PS (21.0.2) and got the results below. As you can see the Photomerge score comparison between our computers just doesn't make any sense, given the other score deltas.

It can't be a GPU issue since Puget's testing shows that GPU has no bearing on the result: https://www.pugetsystems.com/labs/articles/Photoshop-CC-2018-NVIDIA-GeForce-GPU-Performance-1139/

Maybe someone else with a 7,1 MP can run the bench and see if it's reproducible.

(I built this thing for about 2300 CAD (1770 USD) two years ago from consumer parts and it's been serving me well.)

Yes, I personally suspect it might be an error in the scripts of the Mac OS version of their test. I can't find any way to report the results or give them feedback though.

As for the full configuration, it's actually embedded in the results image.
 
Please show me a benchmark that shows a 40-50% increase in render speed for one of the new W-32xx series Xeons vs. an AMD processor. Also please make sure it's consistent across the board, and not just one application.

If you want to extensively benchmark a series of systems, knock yourself out.

I am moving from a pair of Westmere CPUs in my flashed 4,1 (see sig). As shown below, I'll see a 40% increase with my temporary CPU (and over a 300% increase in my primary CPU; when I can get it, most likely April).

The heavy lifter for my workflow is tile based rendering, therefore Cinebench scores matter to me. This is what has been currently reported:

Cinebench R20 - Intel X5650 x2 score 2705 (AKA my Mac Pro)
Cinebench R20 - AMD 2700 score 3488 (temp CPU until I can get my hands on a 3950X)
Cinebench R20 - AMD 3950X score 8788.

At the end of 2020, I should be able to drop in a 4950X and see another double digit uplift in performance (right now Zen 3's engineering samples are showing a 17% IPC increase. AMD's CTO has stated that they are looking for a 15% IPC increase every generation, with a 12 - 18 month release cadence.).

So, realistically, I should be able to manage 48 months on a stable platform, and be ready to jump to the AM5 socket in 2022 (2nd Gen AM5).

Which should be around the time the 8,1 on 10nm (and PCIe 5.0) rolls out.
 
If you want to extensively benchmark a series of systems, knock yourself out.

I am moving from a pair of Westmere CPUs in my flashed 4,1 (see sig). As shown below, I'll see a 40% increase with my temporary CPU (and over a 300% increase in my primary CPU; when I can get it, most likely April).

The heavy lifter for my workflow is tile based rendering, therefore Cinebench scores matter to me. This is what has been currently reported:

Cinebench R20 - Intel X5650 x2 score 2705 (AKA my Mac Pro)
Cinebench R20 - AMD 2700 score 3488 (temp CPU until I can get my hands on a 3950X)
Cinebench R20 - AMD 3950X score 8788.

At the end of 2020, I should be able to drop in a 4950X and see another double digit uplift in performance (right now Zen 3's engineering samples are showing a 17% IPC increase. AMD's CTO has stated that they are looking for a 15% IPC increase every generation, with a 12 - 18 month release cadence.).

So, realistically, I should be able to manage 48 months on a stable platform, and be ready to jump to the AM5 socket in 2022 (2nd Gen AM5).

Which should be around the time the 8,1 on 10nm (and PCIe 5.0) rolls out.

Remember, you were the one claiming 40-50% better performance for an AMD system. Maybe vs. your old outdated system, but that's the case with any new system vs. an old one. This is a Mac Pro 7,1 benchmark thread, not "compare my old Mac Pro to a theoretical system based on an AMD processor thread." I've posted a real-world benchmark showing the performance of the new Mac Pro 7,1 and you deflect away from benchmarks apparently because they don't support your preconceived notions.

Look, if you feel an AMD system is better for you then great, go buy one. I don't care and pretty much no one else here does. If all you care about is a certain subset of tasks that the Mac Pro isn't good at then, oh well, not every machine is everything to everyone--that's always been the case. That said, there's no reason to keep polluting this thread with useless information or worse yet, false claims that don't hold up to real benchmark data. Real benchmark data is showing the new Mac Pro is a powerhouse at many tasks. Cheap? No. But that's never been Apple's market, and quite frankly in the workstation space it's actually quite price competitive.
 
Wrong. Performance freaks care about performance above all else, not pragmatists. For example: the TB3 stability issues that are well-known when you are off the pure Intel platform. A pragmatist needs things to WORK, and not have to worry about compatibility just to eke out a tiny bit more performance or save a little money. Also, there is nothing pragmatic about endlessly speculating about a processor option that doesn't exist; it's just wasting time.

So no, this isn't about pragmatism or a "reality distortion field." It's actually about reality: The reality that the only processor options that exist. Speculation about an AMD processor in a Mac Pro is a total waste of time--that's the true "reality distortion field" here.

i'd say a $5500 computer outperforming a $16.5k mac pro is something else than "a tiny bit more performance or save a little money". but yes it all depends on what you do and need.

Cinebench R20 - AMD 3970X score 17150.

anyway, windows or mac, red pill or blue pill, who cares...
after 20 years of macos i took the red, and i dont say all is bright :)
cooling and keeping such a system silent is a challenge in progress
 
Last edited:
  • Like
Reactions: ssgbryan
i'd say a $5500 computer outperforming a $16.5k mac pro is something else than "a tiny bit more performance or save a little money". but yes it all depends on what you do and need.

Cinebench R20 - AMD 3970X score 17150.

anyway, windows or mac, red pill or blue pill, who cares...
after 20 years of macos i took the red, and i dont say all is bright :)
cooling and keeping such a system silent is a challenge in progress

"Horses for Courses" the Cinebench score alone is just that, one score, on a synthetic benchmark for one task. You don't need a Mac Pro for that, great, don't get one. Why is this so hard for the AMD enthusiasts here to wrap their heads around, and why do all of you feel like you have to weigh in a Mac Pro 7,1 benchmark thread?
 
  • Like
Reactions: tommy chen
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.