THREAD CLOSED!!! Post Your AMD RyZen Blender Benchmarks at 200 Samples!

[Melvis]

All I know is Im getting horrible scores with my 8350 compared to others in this thread and im only putting it down to different OS? as Im on 7 and the best score is on Linux.

 

[basco]

49.74s
win8.1-64, 5960x-3500mhz-16gb-2133c12

 

[silentbogo]

All I know is Im getting horrible scores with my 8350 compared to others in this thread and im only putting it down to different OS? as Im on 7 and the best score is on Linux.

That's strange too, because most of my HWBot scores showed a different trend in cinebench - most of Win10 scores are below Win7 @ same clocks (at least for my old X5650, which I no longer have )

 

[hiepgia]

Update my score with AMD Settings in this (http://cdn.overclock.net/f/f1/f1e01303_48054.jpeg) with Render Sample 100.

Intel E3-1231v3@ 3.6Ghz, DDR3-2133Mhz: 0:48.94

 

[basco]

if i do the same as hiepgia i get 24,65

 

[Feklar]

Re-run with no other programs running. 6900K stock clocks.

 

[silentbogo]

Yep, with Render samples reduced to 100 I get a time of 1:19.
So, the mystery of abnormal scores is solved

 

[xkm1948]

I am too lazy to update every score. Lets just keep it going at 200 settings then.

 

[the54thvoid]

The point of this thread is so we can get some data points of certain tiers of processors and compare it to AMD's claim.

The point of my post was to point out the 6900 scores from forum users aren't close to the scores of the 6900 AMD used, therefore querying the very validity about the Blender test itself. People are suggesting the Blender download from AMD is not the same version used in the demo. If you want to compare current CPU's against Zen, you need a verifiable test and that would be a clean OS install using a stock 6900 getting the same as AMD. Nobody has managed that.

 

[hiepgia]

I am too lazy to update every score. Lets just keep it going at 200 settings then.

Please keep both score 200 and 100

 

[Pickoff]

5960X @4.5 1.244V

HT OFF, Samples 100
28.57

HT-ON, Samples 100
20.18

HT-ON, Samples 200
38.46

 

[xkm1948]

Time for someone else to take over the mantle. Ain'tgot enough time!

 

[ZoneDymo]

1.40.06

Intel 2600k @ 4.5 Ghz

(how do I post the screenshot here?)

http://i.imgur.com/SfOGZSy.png

 

[the54thvoid]

@Feklar - can you rerun at 100 samples with stock clocks? Be a good reference point.

 

[Feklar]

3rd run. Samples 100 instead of 200. 6900k stock clocks. 00.24.62

 

[ZoneDymo]

Update my score with AMD Settings in this (http://cdn.overclock.net/f/f1/f1e01303_48054.jpeg) with Render Sample 100.

Intel E3-1231v3@ 3.6Ghz, DDR3-2133Mhz: 0:48.94

Oh wow, OP take note of this.
This is some good spotting, setting it to 100 like in the AMD test makes a huge difference.
I went down from 1.40.06 to 00.48.57....

 

[Melvis]

That's strange too, because most of my HWBot scores showed a different trend in cinebench - most of Win10 scores are below Win7 @ same clocks (at least for my old X5650, which I no longer have )

Indeed, and here is my scores of my 8350 at stock clocks on Windows 7, seems alot slower then others on different OS's?

3:25:60

 

[biffzinker]

Yep, with Render samples reduced to 100 I get a time of 1:19.
So, the mystery of abnormal scores is solved

Same for me, down to 36.68 compared to my old score of 1.12.38

00:36.68
Samples 100:

1:12.38
Samples 200:

 

[DeathtoGnomes]

2.33.60
stock settings.
200 Samples

1.17.42 @100 Samples

 

[Melvis]

2.33.60
stock settings.
Cant seem to find where to change sample size.

View attachment 81968

Click the + symbol top right then go down three quarters and its there. Took me awhile to lol

Correction! Once you apply the file it should be on the right hand side a third of the way down the page, you might have to reopen the program again, start over to see it.

 

[DeathtoGnomes]

Click the + symbol top right then go down three quarters and its there. Took me awhile to lol

Correction! Once you apply the file it should be on the right hand side a third of the way down the page, you might have to reopen the program again, start over to see it.

thanks, edited my post to include image with 100 samples

 

[dyonoctis]

Wait, i'm confused. Ryzen is apparently doing the blender test in 36 sec. Wich is matched by an haswell quad core i7 at 100 samples, which is apparently the setting used in amd's demo. And the 6900k is 14 second faster than zen. That's not what happened in the demo at all...

 

[Countryside]

I believe RyZen finishes around 35 seconds with 3.4GHz

Correct me if im wrong but i think it was 54sec for RyZen and 59sec with i7

 

[ShurikN]

Correct me if im wrong but i think it was 54sec for RyZen and 59sec with i7

That was handbrake.

 

[DeathtoGnomes]

Wait, i'm confused. Ryzen is apparently doing the blender test in 36 sec. Wich is matched by an haswell quad core i7 at 100 samples, which is apparently the setting used in amd's demo. And the 6900k is 14 second faster than zen. That's not what happened in the demo at all...

$300 chip vs $1100 chip.