Vray benchmark results summary (performance cost)

[TRGraphics]

No, running it stock speed (2182 mhz) on win64 and 1.48.03 with 4 gig of memory.

[jujubee]

Do you suppose your extra 2 gigs of memory is gaining you 36 seconds over us on a slightly slower processor? What kind of card/drives are you running? Strange... Maybe I need more RAM.

[TRGraphics]

The hard drive is just a sata drive 7200 rpm. I did notice a speed increase when I switched to WIN64 and 4 gig of memory. Even on files that don't use that much memory. Maybe win64 just allocates 4gig of memory better on dual core sytems. I really don't know, however.

I also don't think max and vray uses the extra L2 cache on the 4400. At least it doesn't appear to make a difference. I can overclock a 3800 to the same clock speed and get almost identical times on the this file. Within 2 seconds. I've switched all my computers to WIN64 and 4 gig of ram.

2182 mhz is what vray reports. Windows shows it as 2.22 mhz.

Maybe it's 1.48.03 that makes a difference. I didn't compare times before I switched to the newer version.

[jujubee]

Interesting. I'm on 64 as well. So the only other differences that 'would' matter is motherboard and RAM.

I wouldn't have thought increased RAM past 2 gigs would have played much of an issue on rendertimes - just mostly viewport previews... I'll have to run a test in the coming weeks unless someone else can test this out.

[TRGraphics]

My motherboard is the Asus A8N-E. The memory is Corsair XMS3200 CL 3-3-3-8. Not the fastest ram available, for sure. But, I could afford it for 5 computers.

[jow]

i think the reason is cos TRGraphics is using 1.48. ,hasnt it got a reflection bug or something...

[jujubee]

Same motherboard - I have a 'supposed' step up with the ASUS A8N SLI Premium as it was the only one left in the store. But I've disabled all the built-in cool'n'quiet, NOS, and other overclocking tools which may be enabled by default. I'm also officially using/testing the Sun/Sky system.

[jow]

that is weird,i just tested it on mine (1.48.02-x2 4200,2gig ram) and i got a time of 5.13secs,and ive got bits running in the background.

ps.just to make clear,its official aswell.

[TRGraphics]

This is strange! We are all over the place with times. I don't have any of the board OC'ing enabled on this computer either.

I must have missed the post about a bug in 1.48.03. I haven't noticed any problems with it here. What is suppose to be the problem with reflections?

[jow]

maybe the bug is with just 1.48.02,i dont know....
check your results against one of the ones done on 47.03,here on the forum.

im not OCing mine either,and im using xp pro and a gigabyte mobo..

[jujubee]

that is weird,i just tested it on mine (1.48.02-x2 4200,2gig ram) and i got a time of 5.13secs,and ive got bits running in the background.

What were your times with other versions?

It's not just me - daforce is clocking in around the same times on similar setup. It would be ironic if the 4200s are somehow faster than a 4400. Well have to wait till this render is over until I can test.

[DaForce]

its bnecause you other peoples are using 1.48.03... if you render with 1.47.03 your times will be the same as mine an jujubee's. I have rendered with 1.48.03 and it came in around 4m50s i believe

TR try using 1.47.03 and see what times you get

Basically if people are not going to render with 1.47.03 there is no point in the benchmark. Unless we all Re-render with 1.48.03

[DaForce]

People if you are going to partake in this benchmark test you must use 1.47.03 for the results to be comparable.

[n6]

People if you are going to partake in this benchmark test you must use 1.47.03 for the results to be comparable.

true!

[TRGraphics]

I thought the test was in the other thread? I was simply wanting to know more about the dual 265 opteron setup and LightCache rendering. To me, this would be the ideal setup for a primary computer. If you use LC for stills.