# CPU Benchmark

**URL:** <https://forums.chaos.com/t/cpu-benchmark/42329>\
**Category:** Hardware\
**Created:** [August 13, 2012, 1:05pm UTC](https://forums.chaos.com/t/cpu-benchmark/42329 "2012-08-13T13:05:14Z")\
**Posts on this page:** 12\
**Page:** 1

<div class="post-metadata">

**Author:** ![v-cube](https://dub1.discourse-cdn.com/chaosprd/user_avatar/forums.chaos.com/v-cube/32/71276_2.png) [@v-cube](https://forums.chaos.com/u/v-cube)\
**Post date:** [August 13, 2012, 1:05pm UTC](https://forums.chaos.com/t/cpu-benchmark/42329/1 "2012-08-13T13:05:14Z")

</div>

Hello Newsgoup,

is there a **up to date** CPU benchmark to overview the performance of different CPUs? I think I remember some table I saw a while ago but was unable to find it…

can anybody help?

thanks

Andy

---

<div class="post-metadata">

**Author:** ![sandykoufax](https://dub1.discourse-cdn.com/chaosprd/user_avatar/forums.chaos.com/sandykoufax/32/49770_2.png) [@sandykoufax](https://forums.chaos.com/u/sandykoufax)\
**Post date:** [August 13, 2012, 4:08pm UTC](https://forums.chaos.com/t/cpu-benchmark/42329/2 "2012-08-13T16:08:23Z")

</div>

If you wanna just CPU benchmark, visit here

> **[Socolive Trực Tiếp World Cup 2026 - Xem Bóng Đá HD Hôm Nay](https://socolivetvttbdg.tv/)**
>
> Socolive trực tiếp bóng đá chất lượng cao ở tất cả giải đấu, cùng đội ngũ BLV chuyên nghiệp, hình ảnh HD, âm thanh sống động cùng nhiều chuyên mục hữu ích.

[http://www.maxwellrender.com/index.php/benchwell](http://www.maxwellrender.com/index.php/benchwell)

---

<div class="post-metadata">

**Author:** ![v-cube](https://dub1.discourse-cdn.com/chaosprd/user_avatar/forums.chaos.com/v-cube/32/71276_2.png) [@v-cube](https://forums.chaos.com/u/v-cube)\
**Post date:** [August 13, 2012, 4:48pm UTC](https://forums.chaos.com/t/cpu-benchmark/42329/3 "2012-08-13T16:48:40Z")

</div>

thanks sandykoufax, if these numbers are equivalent to the to expect vray performance, than this is great information, however I think in the past we always had “our own” vray benchmark scene (which a some point wasn’t working so good anymore, since it rendered too fast, so the faster machine could not process it in an optimal way).

best regards

Andy

---

<div class="post-metadata">

**Author:** ![TutanMark](https://avatars.discourse-cdn.com/v4/letter/t/e19adc/32.png) [@TutanMark](https://forums.chaos.com/u/TutanMark)\
**Post date:** [September 6, 2012, 2:42pm UTC](https://forums.chaos.com/t/cpu-benchmark/42329/4 "2012-09-06T14:42:51Z")

</div>

Hi Andy.

You can have a look at this also:  
[http://www.cpubenchmark.net/high\_end\_cpus.html](http://www.cpubenchmark.net/high%5C_end%5C_cpus.html)

---

<div class="post-metadata">

**Author:** ![cmjohncheng](https://avatars.discourse-cdn.com/v4/letter/c/3bc359/32.png) [@cmjohncheng](https://forums.chaos.com/u/cmjohncheng)\
**Post date:** [October 27, 2012, 2:34am UTC](https://forums.chaos.com/t/cpu-benchmark/42329/5 "2012-10-27T02:34:46Z")

</div>

y vray dont hv own benchmark tools? like cinema4D got cinebench…can help user decide which processor/graphic card is better for vray.

---

<div class="post-metadata">

**Author:** ![vlado](https://avatars.discourse-cdn.com/v4/letter/v/779978/32.png) [@vlado](https://forums.chaos.com/u/vlado)\
**Post date:** [October 27, 2012, 11:46am UTC](https://forums.chaos.com/t/cpu-benchmark/42329/6 "2012-10-27T11:46:12Z")

</div>

> [@Rotational Motion Blur and Overlapping Geometry Samples...](https://forums.chaos.com/t/rotational-motion-blur-and-overlapping-geometry-samples/38466/20):
>
> y vray dont hv own benchmark tools? like cinema4D got cinebench…can help user decide which processor/graphic card is better for vray.

We have been thinking about this, but it won’t be a very good indication for anything. As processors started getting more and more cores, we found issues with the V-Ray code - things that worked fine on 4 or 8 threads, no longer worked well for 16 or 32 threads. So we had to make specific optimizations to make sure things run as fast as possible. In that sense, the benchmark results would constantly change from V-Ray release to V-Ray release which would make them rather pointless.

Best regards,  
Vlado

---

<div class="post-metadata">

**Author:** ![cmjohncheng](https://avatars.discourse-cdn.com/v4/letter/c/3bc359/32.png) [@cmjohncheng](https://forums.chaos.com/u/cmjohncheng)\
**Post date:** [October 28, 2012, 3:07pm UTC](https://forums.chaos.com/t/cpu-benchmark/42329/7 "2012-10-28T15:07:26Z")

</div>

> [@gfxeye](#):
>
> We have been thinking about this, but it won’t be a very good indication for anything. As processors started getting more and more cores, we found issues with the V-Ray code - things that worked fine on 4 or 8 threads, no longer worked well for 16 or 32 threads. So we had to make specific optimizations to make sure things run as fast as possible. In that sense, the benchmark results would constantly change from V-Ray release to V-Ray release which would make them rather pointless.
> 
> Best regards,  
> Vlado

since vray each stable release not often too(more than half year maybe?), if each release can come with benchmark tool(or come with benchmark tools hotfix/update) sure not bad, at least we know current release use which processor is better.

---

<div class="post-metadata">

**Author:** ![add101](https://dub1.discourse-cdn.com/chaosprd/user_avatar/forums.chaos.com/add101/32/34011_2.png) [@add101](https://forums.chaos.com/u/add101)\
**Post date:** [October 29, 2012, 10:43am UTC](https://forums.chaos.com/t/cpu-benchmark/42329/8 "2012-10-29T10:43:04Z")

</div>

This site has some more specific rendering benchmarks : [CPU Benchmarks and Hierarchy 2026: CPU Rankings | Tom's Hardware](http://www.anandtech.com/bench/CPU/2)  
Doesn’t look like they have workstation processors though, and no-vray benchmarks, but it gives a good overview of relative rendering performance.

---

<div class="post-metadata">

**Author:** ![vlado](https://avatars.discourse-cdn.com/v4/letter/v/779978/32.png) [@vlado](https://forums.chaos.com/u/vlado)\
**Post date:** [October 29, 2012, 10:47am UTC](https://forums.chaos.com/t/cpu-benchmark/42329/9 "2012-10-29T10:47:52Z")

</div>

From what I remember, some of these benchmarks use the default scanline renderer, which has very poor multithreading.

Best regards,  
Vlado

---

<div class="post-metadata">

**Author:** ![Crayox13](https://dub1.discourse-cdn.com/chaosprd/user_avatar/forums.chaos.com/crayox13/32/64242_2.png) [@Crayox13](https://forums.chaos.com/u/Crayox13)\
**Post date:** [October 30, 2012, 9:38am UTC](https://forums.chaos.com/t/cpu-benchmark/42329/10 "2012-10-30T09:38:28Z")

</div>

Vlado, can you tell me if that 16 core issue was fixed, and that vray now scales much better to many cores?  
It’s the difference for me, if I should get 2 4core CPUs, or 2 8core ones.  
thanks

---

<div class="post-metadata">

**Author:** ![Dariusz\_Makowski\_Dadal](https://avatars.discourse-cdn.com/v4/letter/d/c4cdca/32.png) [@Dariusz\_Makowski\_Dadal](https://forums.chaos.com/u/Dariusz_Makowski_Dadal)\
**Post date:** [October 30, 2012, 11:32am UTC](https://forums.chaos.com/t/cpu-benchmark/42329/11 "2012-10-30T11:32:54Z")

</div>

> [@Light cache Fly-Through, Singel frame or full sequence?](https://forums.chaos.com/t/light-cache-fly-through-singel-frame-or-full-sequence/39478/7):
>
> Vlado, can you tell me if that 16 core issue was fixed, and that vray now scales much better to many cores?  
> It’s the difference for me, if I should get 2 4core CPUs, or 2 8core ones.  
> thanks

I think vray currently is optimized up to 16 or even 32 threads… not cores please note.

---

<div class="post-metadata">

**Author:** ![vlado](https://avatars.discourse-cdn.com/v4/letter/v/779978/32.png) [@vlado](https://forums.chaos.com/u/vlado)\
**Post date:** [October 30, 2012, 4:41pm UTC](https://forums.chaos.com/t/cpu-benchmark/42329/12 "2012-10-30T16:41:02Z")

</div>

> [@Light cache Fly-Through, Singel frame or full sequence?](https://forums.chaos.com/t/light-cache-fly-through-singel-frame-or-full-sequence/39478/7):
>
> Vlado, can you tell me if that 16 core issue was fixed, and that vray now scales much better to many cores?

We fixed a few issues; there are a few more to work on, but we’ll get to them too.

Best regards,  
Vlado
