ZBrushCentral

Benchmarking in Zbrush

build a new workstation and it looks like Zbrush doesn’t work with double CPU’s (Xeon e5 v4) ? or is it just me ?
loads only one of them and I get 50% load in the task manager on heavy loads
Max threads I can set in preferences is 40 as well

We just encountered something like this with another user who had an incredible number of cores. He set ZBrush to use a max of 16 cores per CPU and said it then worked like a champ.

Attachments

I’m looking to upgrade from my laptop to a proper desktop.
Anyway here are my current (old) machine’s specs and score:

Medion x7817
intel i7 3610 QM @2,3 Ghz (3,3 Ghz on turbo)
12GB RAM 1333Mhz
500GB hdd @7200 RPM

I get
Single Thread Timer 3.35
Multi Thread Timer 1.05

I’d love to see any Ryzen rigs score while waiting for Threadripper :smiley:

From what I remember, doesn’t ZBrush only use about 8 threads effectively, and after that, it’s a sharp falloff? That is probably why you don’t see ZBrush saturating your entire system.

Hiya!

Just finished building a new Workstation Rig.

AMD Ryzen7 1800x
64GB DDR4
AMD FirePro W9100 (32gb)

I get…

Single = 2.742
Multi = 0.617982
Multithreading = 443%

:slight_smile:

Paul L. Ming

@pming

Strange… your single thread performance is on the same level as CarterTG’s Surface Pro 4 i7 6650U 2.20Ghz (16GB Ram, Windows 10 x64, ZBrush 4R7 Patch3 x64)

Multi is much better but that’s to be expected.

Your scores are also pretty much in-line with what Invertex posts here about his i7 5820k at stock speeds.
That’s not bad but I was expecting better.

Anyway, thanks for the numbers!

I wonder how this benchmark reflects real performance when sculpting.
Maybe we should share a ZPR (that would make sure we had the same project settings) and test some common taxing operations like zremeshing, dynameshing etc.
Is there a way to accuratemy measure fps in ZBrush? That would be useful - knowing how laggy it gets at what point count…

cheers

I built this workstation about a year ago. Thought I’d post results with zbrush 4r8p1
Windows 7 x64
Dual Xeon E5-2690 @ 2.9Ghz (3.8GHz Turbo Boost) - 8 hyper threaded cores per CPU = 32 cores total
64 GB ram quad channel ram

single thread - 3.519
multi thread - 0.4429
multi thread performance - 794%

For me, zbrush uses all cores on both of my CPUs when doing intensive things like dynamesh (only if project is turned on), project all and rotating around a multi million poly model.

Hmm…

On a core i980x/24GB RAM I get pretty decent scores

Single Threaded Timer = 3.1
Multi Threaded Timer = 0.68
Multithreading = 452%

Considered upgrading (after almost 8 years) to a Ryzen 1800x but seeing ZBrush performance on it is only like a few percent better on it I’m not so sure any more.
Not even sure though what this benchmark exactly does measure. To me brush performance would be the most important part.

[quote="“RafalZ,post:325,topic:204544”]

Strange… your single thread performance is on the same level as CarterTG’s Surface Pro 4 i7 6650U 2.20Ghz (16GB Ram, Windows 10 x64, ZBrush 4R7 Patch3 x64)

Multi is much better but that’s to be expected.

Your scores are also pretty much in-line with what Invertex posts here about his i7 5820k at stock speeds.
That’s not bad but I was expecting better.[/QUOTE]

Yeah this is a lot worse than I was expecting… The 5820k came out in 2014 and is cheaper, yet beats the 1800X which just came out.

Given such results, I would highly recommend anyone coming across this thread to consider an Intel i7 8700k which is around 20% faster than a Ryzen 1800x in general work, and around $90 cheaper. But, if ZBrush performance is your primary focus, then go with the Intel i7 7800X, which has quad channel memory for higher memory bandwidth which ZBrush needs as much as it can get (use 4 or 8 memory sticks to utilize quad channel) and is more than $100 cheaper than the AMD 1800X.

Thats very interesting seems like the ryzen really aren’t performing well in zbrush, thats disappointing. Does anyone know what tasks uses single thread and what uses multi thread. im looking to build a new rig that could handle as much tris in a ztool as possible. i have the 5820 right now, if i can find something that can improve the performnce of zbrush by at least 50% id be happy.

Hiya!

My new Maingear system is awesome. :slight_smile: Expensive as a new car…but awesome!

Ran several checks on different days. Average is about…

Single = 2.4
Multi = .3
MT Performance = 770%

One thing interesting…I can’t up the “Max Threads” past 44. :frowning: Oh well…still rocks on toast!

Dual Xeon E5-2699 v4, 3600 MHz (36 x 100)
128gb ECC DDR4-2400
AMD FirePro W9100 (32gb)

So far I’ve found Keyshot, MODO, Maya, Houdini and Blender utilize all 88 threads to render. And, I must admit, I do get a little bit har…“excited” :wink:
Now my Ryzen build, with an 11gb 1080Ti I use for gaming on my 65" QEHD 4k TV. Sweeeeeet! :smiley:

:slight_smile:

Paul L. Ming

Single:3.202
Multi:1.78199998
Performance: 179%
Kaby Lake G4560
8Gb Ram 2000Mhz
Radeon R9 200
Win 10
ZBrush 4R7 in trial

Stick to the 5820k for now, no current upgrades will give you the performance boost you want unless you’re willing to pay thousands. If you haven’t overclocked your 5820k though, there is a lot of performance to gain there without you paying more money.

hi all,

I was for awile looking for information to decide between 7900x and a threadripper.
Seems that Zbrush likes INTEL cores better than AMD. at leas for now or there is some kind problem with AMD XFR. (hyperthread)

end up with a I9 7900X

and here are the Zbench redsults @ 4.3ghz.

Single Thread Timer= 1.934999
Multi Thread Timer= 0.31

Multithreading performance= 624%

25659767_10156562720661002_3636368649217707752_n.jpg

also found some:

AMD Threadripper 1950x result to compare:

Single Thread Timer= 2.624
Multi Thread Timer= 0.490997

Multithreading performance= 534%

I think the issue comes from the fact that AMD is prioritizing core count over per-core performance. Intel’s individual cores are some 20-30% faster than AMD’s similarly priced chip cores depending on workload, and not everything in most programs is or can be made fully multi-threaded and thus can be held up by too slow of single-core performance, so finding that right balance between single-core performance and core count is important and I think Intel has it right for most content creator workloads, since for highly-parallel workloads, using the GPU is much more preferable, we need the CPU to process those long serial operations.

Hi,
Here’s the result I get on 1950x with 32gigs of RAM in 4 slots (3200mhz, timings 14-14-14-34)
(no manual overclock, just xmp for ram to work on advertised speed)

Single Thread Timer= 2.461
Multi Thread Timer= 0.364998

Multithreading performance= 674%

cheers
Rafal

oh thats better fot the TR.

( I edited my post , the multy thread was 0.31) @ 4.3ghz

@ stock I get
2.1 single
0.34 multy

so is about th same as the threadripper

my ram 32gb at 3200 mhz CL 16. T: 16 18 18 36

Intel Core i7 8700K 3.7 GHz 12MB

Single Thread Timer= 1.94
Multi Thread Timer= 0.54
Multithreading performance= 353%

I would say good value for money

How is the new performance compared to 4r8?
Do you measure similar values?

No, Actutaly im getting totally different values in ZB 2018 about 20% worst numbers

4r8 multy= 3.1

2018 multy= 3.6

, no idea why.