• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Blender release official benchmark

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
I got ebb to run up to Mr. Drukers for me,
got some tweaking to do on the 8086K, it's right there with the 6800k.........big suprize there.

I'll have to stop by the junction and see uncle joe, he's moving mighty slow but, his cameras were installed by Chuck Berry.
 
Thank you, OCF men, but what is the purpose of this benchmark? What does it accomplish? I don't understand. To get a score? Isn't there already a lot of different other benchmarks that do the same?

I mean, if your CPU is faster in cinebench or whatever, it'll also be faster in blender:shrug:


son, son, here, we are men, we HAVE to compete, doesn't matter what with, but I promise you, 2 minutes after the second car on earth was sold was when the first motor race took place.
 
I'm using v388.71.
I guessing that time is because I have two GTX 970's. One is the FTW+ model & the other is a standard ACX. They are different EVGA cards that can't run SLI, but I bet they shared the workload running that test. I keep them both in there for F@H, I don't really need SLI.

Oh, did Mr. Drucker get your fitting in yet? Stay away from that crap Mr. Haney sells... :D


there is no sli in blender, each scene is broken down into tiles and each card works on their own tiles, one tile at a time, you just throw what ever cards you have in the rig .
you can choose to use which ever cards you want to use in the rig, one card or two or how ever many you want. this file set is locked to all the cpu cores and only one gpu can be used with it.
you can download the demo files of these at blender.org and render each one in the unlocked, blender download, get the Barcelona pavillon, that guy did some awesome work!!!!
 
So two non-SLI cards in the PC won't make a difference vs one, interesting.
Does the benchmark just use the CPU cores when selecting the CPU test?

I'll checkout the demo files, sounds interesting.
 
That was bone stock. I can't imagine an overclocking taking minutes off... that makes zero sense... wonder if its because two monitors or a bunch of stuff up? It aint the HOF that's doing it, :rofl:. Its still a 1070.
 
Last edited:
One monitor. OCF and HWMonitor running for all runs. And my stock time was still faster than 10:15 (9:30.53).

edit: The Pascals respond well to vRAM speed. I ran mine at 4552 MHz. Try bumping up your vRAM on that card.I picked up seconds with clock speed but vRAM OC gave it a big boost.
 
Last edited:
I just wish my cooling and ambient temp was sufficient to run the CPU short test. :( All I have left is swapping the BIOS on my VGA and going for big clocks, but it's already late and my day is winding down.

edit: I'm curious to see some more GPU scores. So far mine seem like a glitch, but they're a consistent glitch if that's the case. It's not like I cherry picked an outlier or anything.
 
Last edited:
And I'm still impressed with my GPU scores. I'm still within what, 20% of a 1080 Ti, maybe less? Obviously I don't know how the bench scales, but for now I'll stick with smug. LOL Besides, I don't expect it to last long. Where's Neb with his 1070?

edit: Something just occurred to me, even though the render is happening in the background, display resolution may be a factor. That would explain a lot, wouldn't it? Just spitballin' here. I don't know squat about how Blender or their benchmark works.
 
Last edited:
"your" screen resolution does not matter, even in a regular render, blender only displays it in the resolution that it is rendering in, and in the bench it's not displayed at all.
 
Back