MP 7,1 Has anyone seen a benchmark?

[radus]

just to compare to amd top of the line
https://gadgetversus.com/processor/amd-ryzen-threadripper-3970x-vs-intel-xeon-w-3275/
oh this is the right one
https://gadgetversus.com/processor/amd-ryzen-threadripper-3970x-vs-intel-xeon-w-3275m/

 

[H. Flower]

Thats a bit low for multi I would say, no?

Just put in 192 gigs of the Samsung RAM, score is now:

1146, 15822

 

[chfilm]

Just put in 192 gigs of the Samsung RAM, score is now:

1146, 15822

Ha that’s awesome! Can’t wait for my machine! Thanks 😊

 

[Macsonic]

Real world speed test on the 2019 new Mac Pro compared to the 2013 trashcan and 2019 MacBook Pro.

 

[The Other One]

Real world speed test on the 2019 new Mac Pro compared to the 2013 trashcan and 2019 MacBook Pro.

I'm really thrown by those results. I'm chiefly interested in performance in Premiere and After Effects... and if those results are accurate, the 16" MacBook Pro is largely keeping up with the Mac Pros. What am I missing?

 

[deconstruct60]

I'm really thrown by those results. I'm chiefly interested in performance in Premiere and After Effects... and if those results are accurate, the 16" MacBook Pro is largely keeping up with the Mac Pros. What am I missing?

Shouldn't be much thrown much. Much of this isn't particularly surprising.

The 8 core option for the new Mac Pro only Turbo's up to 4.0 GHz. ( the rest turbo up to 4.4 GHz. ) . In contrast, the maximum CPU option on the MBP 16" Turbo max is 4.5 GHz. So these entry models 'give up' approximately 0.5 GHz on highly single threaded critical section portions of applications.

The moving of large files in and out of the storage system will have an impact with the MBP 16" having 16GB more of RAM ( a substantive portion can either be used by the file system for cache or as workings space for the app).

The GPUs across the MBP 16" is about even if not pushing very hard on OpenCL ( or not much direct Metal). The baseline VRAM here s 8GB and the Navi based GPU in the MBP 16" is the 'newest' of the bunch. (so not particularly greatly hobbled by being a 'mobile' GPU. )

Bump the Mac Pro to 12c and even up (or surpass) the RAM capacities of the MBP 16" and the edge will probably shrink on most of these differences. ( and gap grow where the multiple CPU cores and/or GPU can get some traction on the workload).

Adobe's targeting the lowest common denominator is a bit of a boat anchor here also.

But yes ... in part this is one factor as to why Apple isn't deeply stressed in addressing the "old" $2,500-4,000 range of the Mac Pro. Attach an external display ( and storage and perhaps a eGPU) to the MBP 16" and covering much of the old workload space. ( iMac Pro's 10 core ( Turbo 4.4Ghz ) will do well against the 8 core Mac Pro . Same with iMac Turbo i9 for as long as it can hold the 5.0 Ghz there).


The entry Mac Pro configuration's main value point is the stuff that isn't there ( empty slots to be filed.). bumping CPU , Storage , and RAM one step past those initial setting all get to better performance workstation starting point on general workloads. The entry is a better match to where adding in a card or two allocated to a narrow workload is where there is better "bang for the buck".

 

[gazwas]

I'm really thrown by those results. I'm chiefly interested in performance in Premiere and After Effects... and if those results are accurate, the 16" MacBook Pro is largely keeping up with the Mac Pros. What am I missing?

I watched that and thought the same?

I’m a FCPX user so the Premiere/After Effects numbers don’t worry me as I know there has been an update for then Mac Pro in FC.

I just think its early days as I expect software will be optimised for such an important machine like the Mac Pro.

 

[chfilm]

I watched that and thought the same?

I’m a FCPX user so the Premiere/After Effects numbers don’t worry me as I know there has been an update for then Mac Pro in FC.

I just think its early days as I expect software will be optimised for such an important machine like the Mac Pro.

Honestly I didn’t like that video. It didn’t seem to me like a very thorough real world work example, also his Mac Pro isnt particularly powerful specced, of course it doesn’t compete very well.

Still, a big part seems to be Adobe not being optimized again for the new horse power and also it just shows that the MacBook is a beast, which is awesome.

 

[deconstruct60]

....
I’m a FCPX user so the Premiere/After Effects numbers don’t worry me as I know there has been an update for then Mac Pro in FC.

I just think its early days as I expect software will be optimised for such an important machine like the Mac Pro.

It isn't just the Mac Pro. Adobe apps ignoring the GPU are bad in the MBP 16" too. That MP 2013 in the benchmark line up had D300 in it. 2GB of VRAM is relatively improvised in the current era, but could rationalize why apps from 2-5 years ago would avoid it as being "too small". [ Note: the MBP 16" GPU VRAM options start off at 4GB. ]

If Apple does something reasonable with the rest of the Mac line up in terms of GPUs in 2020 (especially in the rest of the desktop line up), software which deeply underutilizes the GPUs for appropriate workloads will increasingly underperform the other titles that do. That is only way going to drag Adobe into the present reality of hardware abilities.

 

[Macsonic]

I'm really thrown by those results. I'm chiefly interested in performance in Premiere and After Effects... and if those results are accurate, the 16" MacBook Pro is largely keeping up with the Mac Pros. What am I missing?

Yeah, you could be right. More benchmark results may surface in the coming weeks and we can compare. I just shared this video as reference to this thread topic. Thanks

 

[danwells]

New MBP is, in fact, a beast... I've got one I've been reviewing on another site, and I really can't find any benchmark where it's not right in the same range as all of the 8 core desktops. I'm a little surprised that the 12-cores the Photoshop Cafe guy had couldn't run away from it, but it is REALLY surprisingly quick, and it can sustain its power in a way that laptops often can't.

 

[Nugget]

I'm really thrown by those results. I'm chiefly interested in performance in Premiere and After Effects... and if those results are accurate, the 16" MacBook Pro is largely keeping up with the Mac Pros. What am I missing?

Make sure to factor in the price of the active noise-cancelling headphones you’ll want to be wearing while that MacBook Pro is doing the work and the fans are spun up to 5500rpm.

 

[Passingby]

Bootcamp GPU results are now published on Geekbench.

OpenCL Benchmarks - Geekbench

Bootcamp is the only way to compare this AMD card to Nvidia. In general openCL the Radeon Pro Vega II is 10% slower than the Quadro RTX 5000 Max-Q or GeForce RTX 2080 Max-Q . The irony is that Razer put these fantastic GPUs in laptops that cost about the same as the Radeon Pro Vega II and they run quite cool and silently in that thin and light laptop form factor. Amazing achievements by Nvidia and Razer.

Apple should have gone for Titan RTX if they were going to charge that much instead of a variation of the failed Radeon VII/Instinct MI60.

 

[Passingby]

Cinebench scores.

Base model 8 core 2019 Mac Pro is about the same as the 12 core X5690 cMP.

For extra context an 8 core 9900K @ 5Ghz gets 5400 matching the 12 core 2019 Mac Pro.

An overclocked Cascade Lake i9/Xeon or Threadripper 3970X would make the bars below look very short.

 

[gazwas]

I’m so tempted by the 16” MacBook Pro but just can’t bring myself to spend so much (£4000) on a laptop that I can’t ever upgrade the RAM, SSD, Battery, GPU or simply just clean it internally.

So far, all the tests I’ve seen point to the 16 and 24 core machines to be the best of the range. 8 and 12 core Mac Pro’s are too similar to high end regular Macs and the 28 core not much faster the the 24 core.

Regardless of any of these machine performance, a major reason people want a Mac Pro (including me) is the ability to upgrade components in the future and the Mac Pro is the ONLY option available.

 

[Passingby]

I’m so tempted by the 16” MacBook Pro but just can’t bring myself to spend so much (£4000) on a laptop that I can’t ever upgrade the RAM, SSD, Battery, GPU or simply just clean it internally.

Yeah hard. Apple should see how PC laptops are sticking to their guns. Some have two m.2 drives. The Razer Blades have upgradable RAM+storage with desktop class high performance graphics running cool. Also optical switch keyboards now. Apple could do all this but they are obsessive about taking freedoms away from users, put a walled garden around everyone and pressure customers into more regular full computer upgrade. Imagine if Apple was a government and treated citizens in this obsessive way. Another tech firm gone mad.

 

[pierre1610]

Geekbench 5 scores for the 8-, 12-, and 16-core Mac Pro are now available on the Mac Benchmark Chart:

https://browser.geekbench.com/mac-benchmarks

Why is the 28 core is much higher than the 12 core when it’s a single core raiting and the clock speed is lower on the 28

 

[zzzachi]

Mac Pro 2010 2x6 Core - CB20 3412 - 3800$ (my old Mac Pro)
Mac Pro 2019 8 Core - CB20 3660 - 6400$ (32gb of RAM, RX580 8GB)
Mac Pro 2019 12 Core - CB20 5490 - 7400$ (32gb of RAM, RX580 8GB)
Threadripper Ryzen 3970X - CB20 16660 - $5500 (64GB of RAM, RTX2080 11GB)

... this made me leave MacOS and buy a Threadripper System ...
(this and the security that I can always upgrade and change things as I like)

I also bought an RX580 to upgrade my Mac Pro 2012, to use it aside the PC.
PC is for 3D Work, Mac Pro for everything else.

 

[MGrayson3]

Why is the 28 core is much higher than the 12 core when it’s a single core raiting and the clock speed is lower on the 28

All these numbers are nearly identical. Since 5 different runs is all they need to include a machine, I suspect the standard errors are larger than the score difference. The iMac turbo boost is to 5 GHz, and the Xeons only go to 4.6 GHz. That's what you're seeing on single processor scores.

 

[gazwas]

Why is the 28 core is much higher than the 12 core when it’s a single core raiting and the clock speed is lower on the 28

Noticed that very wierd result also.

I think until more machines are in the wild these results need to be taken with a pinch of salt.

 

[smbu2000]

Cinebench scores.

Base model 8 core 2019 Mac Pro is about the same as the 12 core X5690 cMP.

For extra context an 8 core 9900K @ 5Ghz gets 5400 matching the 12 core 2019 Mac Pro.

An overclocked Cascade Lake i9/Xeon or Threadripper 3970X would make the bars below look very short.

Are those scores correct? They seem quite low. My 12core Ryzen R9 3900X scores over 7200 in CB20 at stock clocks.

 

[Passingby]

Are those scores correct? They seem quite low. My 12core Ryzen R9 3900X scores over 7200 in CB20 at stock clocks.

These Xeons are only 4 to 4.4 ghz and don't have as good IPC as the Ryzen/i9/Cascade Lake/Threadripper

 

[gazwas]

These Xeons are only 4 to 4.4 ghz and don't have as good IPC as the Ryzen/i9/Cascade Lake/Threadripper

While I don’t doubt you, it still doesn’t explain why the 2.5 GHz 28 core Mac Pro’s single core score is so much higher than all the others with same Gen processors but higher clock speeds?

Is this a result of people who buy the 28 core machine spend more on RAM, GPU etc that may be effecting the scores somehow Or is Catalina not yet playing nice with the new machine?

 

[thisisnotmyname]

Honestly I didn’t like that video. It didn’t seem to me like a very thorough real world work example, also his Mac Pro isnt particularly powerful specced, of course it doesn’t compete very well.

Still, a big part seems to be Adobe not being optimized again for the new horse power and also it just shows that the MacBook is a beast, which is awesome.

Right, I just saw a poorly designed experiment. As I saw the results I was just thinking, "OK, you've found a single threading limit here and a RAM bottleneck there (with double the RAM in one system)," etc... It wasn't so much a shootout of architectures but rather proving that some tasks are throttled by CPU/RAM/whatever; we all knew that going in.

 

[Snow Tiger]

The only numbers I've seen published online are in Marquess Brownlee's YouTube video (28 cores, 384 GB RAM)...

Geekbench 5 Single-core ~1200 (very similar to all other fast Macs - iMac 8 core, iMac Pro, MBP16)
Geekbench 5 Multi-core ~21500 (about 3x as fast as the logjam of fast Macs, 1.5x as fast as 18-core iMac Pro)
Disk Speed Test - just under 3 GB/second (a tiny bit faster than iMac Pro and MBP16, the fastest other Macs)

Has anyone seen any other configuration, or any other test?

Here are scores from two different installed processors in the same Mac Pro 7,1 with 32 GB ( 4 x 8 GB ) .

W-3223 Eight Core @ 3.5 GHz GB5 ( 64 bit ) Single Core = 1061 .
W-3223 Eight Core @ 3.5 GHz GB5 ( 64 bit ) Multi Core = 8220 .

Gold 6212U Twenty Four Core @ 2.4 GHz GB5 ( 64 bit ) Single Core = 1052 .
Gold 6212U Twenty Four Core @ 2.4 GHz GB5 ( 64 bit ) Multi Core = 15305 .

The Gold 6212U System was not able to have its NVRAM refreshed . Had I been successful , that score should have been higher . System firmware is too immature for such a prototype processor installation at this time .

Both processor configurations had memory configurations installed as advised by Apple . Unoptimized memory installations will adversely affect scores .