VRBench from Chaosgroup.

Wraith

Bettyswollocks
In march of 2016 our very own Newsguru WYP pulled us altogether and we went nuts benching our CPUs for props and huge swelling epeen with a benchmark called Corona as you can see we even got mad props for being the first tech site to showcase it.



Now while looking into more benches and why not we all love to bench right, we discover a shoutout to the Guv himself via an Instagram post to take a look at a new bench.. well I couldn't help myself and dove right on in and to be honest it's not half bad. Now rather than pass this by and just keep browsing we figured you guys of the forums might just want to have a look at it and once again get the tech testosterone flowing.

So here it is, VRaybench benchmark for CPU & GPU by a company called Chaosgroup. While it is free to download you will have to register with their website to access the downloads it's not that big of a deal with only a couple of details to fill in though.

Here are our first run bench results to kick off the board, you'll find a direct download link at the bottom.

kd3MIvn.jpg


cSY1vP8.jpg

CPU results​

AlienALX - 01:05 (XEON @ 2.8Ghz)
DiceHunter - 01:16 (1800X @ 4.0Ghz - Mem 3200MHz)
WYP - 01:16 (1700X @ 4.0Ghz - Mem 3200MHz)
Wraith - 02:43 (4790K @ 4.0Ghz - Mem 1866MHz)​

GPU results​

Dicehunter - 01:01 (GTX1080Ti 2050/6000)
AlienALX - 01:09 (Titan X Pascal Stock)
Wraith - 01:33 (GTX1070FE Stock)
WYP - 01:38 (GTX1080FE Stock)

For our results we need RAM speed, Overclocks (if OC'd) or just state stock speeds.

Download link for VRBench
 
Last edited:
Not terribly impressed with either of my scores tbh. Also not terribly impressed that a 1070 scores the same as a 1080.

fTU3ph5.jpg
 
It's more about the fact the 1070 & 1080 are the same architecture and have the same amount of VRAM OK the 1080 is GDDR5X but from our preliminary tests it seems it makes little difference.

EDIT: Which CPU are you using? @2.2Ghz
 
I'll try and pump out some more results later. See how Ryzen works when the memory is run slower etc.

I'll also try the Intel Rig.
 
It's more about the fact the 1070 & 1080 are the same architecture and have the same amount of VRAM OK the 1080 is GDDR5X but from our preliminary tests it seems it makes little difference.

EDIT: Which CPU are you using? @2.2Ghz

It's 2.8ghz. In my sig dude (the 14 core Xeon).
 
Very strange. I closed MSI AB and it finished at 1:01. Then nothing happened, it skipped, and I got 1:06.
 
Yeah, we're just running this to see what the results look like from a community aspect and see if there is any validity to the bench. So far it's got us scratching our heads.
 
I've tried it stock (2.8ghz CPU and stock Titan XP) and with 150mhz on the TXP. Other than that no change, apart from closing MSIAB. It does load all of the cores up though.
 
I'm going to kill all non essential programs and unplug all my other monitors and see if that makes a difference.
 
It is impressive that scores from myself and Dice are that close. Not exactly unexpected for the same CPU at the same clocks and memory speeds.

Will see if any other programs were harming my scores and if I can replicate this variability.
 
Interesting.. I've run this bench 8x now and the CPU score stays the same every time. It's only the GPU side that seems to take a hit, slightly better only running 1 monitor and zero background programs to scupper the results.

lnZ5VZa.jpg
 
That is what I find odd. I can not get the same exact score in Cinebench no matter how hard I try. There is always some variance, and more than 1 second. Having said that I score at absolute max around 2100 in Cinebench, which is around 20% faster than a 1800X at full clocks. If I really crunched the times the result is at least similar in this.

I desperately want to install VMWare and OSX Sierra and basically split my PC in two, but I have no bloody internet speed.
 
Interesting.. I've run this bench 8x now and the CPU score stays the same every time. It's only the GPU side that seems to take a hit, slightly better only running 1 monitor and zero background programs to scupper the results.

Does your GPU always prefer maximum performance? Is the GPU always clocking to max?
 
Back
Top