Question V-ray 6 / Corona benchmarks

igor_kavinski · Feb 3, 2025

Lion Cove and Skymont performance: http://www.portvapes.co.uk/?id=Latest-exam-1Z0-876-Dumps&exid=threads/arrow-lake-builders-thread.2622775/post-41387565

Lion Cove per core performance: 2,253.833 vsamples

Skymont per core performance: 1,777.875 vsamples

Feel free to post your per core performance in this thread, especially if you have a Zen 4c or Zen 5c core in your CPU.

Alder Lake and Raptor Lake owners also encouraged.

To isolate the process of the wimpy cores, use Process Lasso and apply process affinity on the child render process of V-ray. Then keep Task Manager open side by side to ensure that only the wimpy cores are being utilized, for the sake of accurate results.

thigobr · Feb 3, 2025

Lenovo Thinkpad T16 - 7840U - 32GB LPDDR5 6400 RAM

Code:

Microsoft Windows 11
V-Ray 6.10.03 [155820e5]
[0]: AMD Ryzen 7 PRO 7840U w/ Radeon 780M Graphics

Testing V-Ray... 100% 01:00/01:00
V-Ray score: 10453 vsamples

Code:

Linux 5.15.167.4-microsoft-standard-WSL2
V-Ray 6.10.03 [155820e5]
[0]: AMD Ryzen 7 PRO 7840U w/ Radeon 780M Graphics

Testing V-Ray... 100% 01:00/01:00
V-Ray score: 10622 vsamples

Custom 5950X - 64GB RAM DDR4 3800 RAM

Code:

Linux 6.13.0-1-default
V-Ray 6.10.03 [155820e5]
[0]: AMD Ryzen 9 5950X 16-Core Processor

Testing V-Ray... 100% 01:00/01:00
V-Ray score: 35395 vsamples

Code:

Linux 6.13.1-xanmod1
V-Ray 6.10.03 [155820e5]
[0]: AMD Ryzen 9 5950X 16-Core Processor

Testing V-Ray... 100% 01:00/01:00
V-Ray score: 36354 vsamples

igor_kavinski · Feb 3, 2025

Question: what is the second score for? Both those CPUs have cores of the same type.

Det0x · Feb 4, 2025

Maybe i will update my score when i got the time

This was done on older lower performing agesa

V-Ray 6 Benchmark

benchmark.chaos.com

And a newer Corona 10 benchmark-run as a bonus
(this benchmark seems to scale with memory amount)

Corona 10 Benchmark

benchmark.chaos.com

poke01 · Feb 4, 2025

CPU test - M4 4P+6E. An increase of ~30% from M3. These new cores are awesome at software RT. M4 CPU used less than 20 watts to get this score.

7700X 8C/16T - stock

igor_kavinski · Feb 4, 2025

@poke01 , any way for you to isolate the performance of your M4 P and E cores?

Also, you and Det0x could try to pin the workload only on the virtual SMT threads. I wonder if it decreases the performance in any way or if the CPU, seeing the main thread idle, just assigns all core resources to the virtual thread.

poke01 · Feb 4, 2025

igor_kavinski said:
@poke01 , any way for you to isolate the performance of your M4 P and E cores?

I don't think macOS has that feature. Maybe once Asahi Linux is supported on M4, I could try again.

Det0x · Feb 4, 2025

igor_kavinski said:
@poke01 , any way for you to isolate the performance of your M4 P and E cores?

Also, you and Det0x could try to pin the workload only on the virtual SMT threads. I wonder if it decreases the performance in any way or if the CPU, seeing the main thread idle, just assigns all core resources to the virtual thread.

There are no "primary threads" and "secondary threads" when your running with SMT enabled

"thread 1" and "thread 2" on core0 will perform the same, if they are benchmarked aginst each other in ST workloads

Only running thread 1-3-5-7-9-11-13-15 will have the same performance as only running thread 2-4-6-8-10-12-14-16, if done with affix selection on a 8 core CCD

Guess you can view both threads from a core with SMT enabled as "virtual"

igor_kavinski · Feb 4, 2025

Thanks, Det0x. Didn't know that Corona bench was similar. Quick to report results and portable!

Memory can't help Ice Lake it seems. Less than half of 9950X's score.

511 · Feb 4, 2025

igor_kavinski said:
Lion Cove and Skymont performance: http://www.portvapes.co.uk/?id=Latest-exam-1Z0-876-Dumps&exid=threads/arrow-lake-builders-thread.2622775/post-41387565

Lion Cove per core performance: 2,253.833 vsamples

Skymont per core performance: 1,777.875 vsamples

Feel free to post your per core performance in this thread, especially if you have a Zen 4c or Zen 5c core in your CPU.

Alder Lake and Raptor Lake owners also encouraged.

To isolate the process of the wimpy cores, use Process Lasso and apply process affinity on the child render process of V-ray. Then keep Task Manager open side by side to ensure that only the wimpy cores are being utilized, for the sake of accurate results.

Where to download it from?

igor_kavinski · Feb 4, 2025

511 said:
Where to download it from?

V-Ray Benchmark – test your CPU/GPU rendering power | Chaos

Test free the render performance of your hardware with V-Ray® Benchmark. See how fast V-Ray renders on your machine using our free standalone app.

www.chaos.com

https://www.chaos.com/corona/benchmark

511 · Feb 4, 2025

PL1 = PL2 = 45W

511 · Feb 4, 2025

Pl1 = Pl2 = 45W HT ON

Pl1 = Pl2 = 65W HT ON

thigobr · Feb 4, 2025

igor_kavinski said:
Question: what is the second score for? Both those CPUs have cores of the same type.

It was using a different Linux install/kernel. First entry was running on OpenSuse Tumbleweed and second entry on Gentoo with custom Xanmod kernel.

igor_kavinski · Feb 4, 2025

511 said:
View attachment 116285

Pl1 = Pl2 = 65W HT ON

Can you do 65W HT off please?

CropDuster · Feb 4, 2025

511 · Feb 5, 2025

igor_kavinski said:
Can you do 65W HT off please?

PL1=PL2=65W HT OFF

igor_kavinski · Feb 5, 2025

511 said:
PL1=PL2=65W HT OFFView attachment 116350

Interesting that HT on that CPU is more effective at 45W than 65W.

511 · Feb 5, 2025

igor_kavinski said:
Interesting that HT on that CPU is more effective at 45W than 65W.

3% difference between
at 45W 14271/11995 = 1.18 or 18% more
at 65W 17188/14886 = 1.15 or 15%

CropDuster · Feb 5, 2025

CropDuster said:
View attachment 116321

HT off

LightningZ71 · Feb 5, 2025

511 said:
3% difference between
at 45W 14271/11995 = 1.18 or 18% more
at 65W 17188/14886 = 1.15 or 15%

Maybe memory subsystem contention issues? L3 down to dram will be under more pressure.

511 · Feb 5, 2025

LightningZ71 said:
Maybe memory subsystem contention issues? L3 down to dram will be under more pressure.

Perhaps

CropDuster said:
View attachment 116395

HT off

What were the PL1 /2?

igor_kavinski · Feb 17, 2025

Core Ultra 9 285K und Co. im Neu-Test: Intel bricht AMD-Dominanz

PCGH testet den Core Ultra 9 285K, Core Ultra 7 265K und Core Ultra 5 245K erneut mit frischen Updates. Kann Intel die AMD-Dominanz brechen?

www.pcgameshardware.de

IPC test with all cores fixed at 4.0 GHz

But not enough P-cores:

MarkPost · Feb 17, 2025

igor_kavinski said:
Core Ultra 9 285K und Co. im Neu-Test: Intel bricht AMD-Dominanz

PCGH testet den Core Ultra 9 285K, Core Ultra 7 265K und Core Ultra 5 245K erneut mit frischen Updates. Kann Intel die AMD-Dominanz brechen?

www.pcgameshardware.de

IPC test with all cores fixed at 4.0 GHz

View attachment 117387

But not enough P-cores:

View attachment 117388

wtf?

whats up with these pcgameshardware little boys?

This is the pefomance of a real zen5 (4 cores, 4.0GHz):

igor_kavinski · Feb 18, 2025

MarkPost said:
This is the pefomance of a real zen5 (4 cores, 4.0GHz):

Can you try with DDR5-5600 CL36 (assuming they chose sedate RAM timings)?

EDIT: Also, they seem to be doing a 10 min stress test.

Question V-ray 6 / Corona benchmarks

Lifer

Senior member

Lifer

Golden Member

Diamond Member

Lifer

Diamond Member

Golden Member

Lifer

Golden Member

Lifer

Golden Member

Golden Member

Senior member

Lifer

Senior member

Golden Member

Lifer

Golden Member

Senior member

Platinum Member

Golden Member

Lifer

Senior member

Lifer