[ivox3]

I can see it, ..but I have to put on X-ray glasses ...

[Bubbaloo]

It's up now.

[numerobis]

i hope the 64bit bug can be fixed and we see similar results with XPx64 and OS X sooon...

[-Adrian]

Please try to use the available list; I find myself correcting about every 3rd entry as people enter dupes. Thanks

[Bubbaloo]

Sorry, I wanted to make sure it listed them as dual quad cpu's.

[monochrome]

How is it that a stock e6600 dual core is getting nearly the same rendering times as QX6700? Of course I don't really care, I did the tests for my own verification of price/performance ration...but it's annoying.

To the liars, what's the point? You're like one of those people with a 10 year old BMW with M-badges glued on at some chop shop.

[b-kandor]

monochrome: I'm not sure anyone's lying - your time with that chip is too long for some reason? Are you using win32 with 32bit maxwell? Also, what did you set the thread count to?

btw: I have an e6400 (dual core)oc'd to 3.2ghz which ran the benchwell test in 35:38. Meanwhile, my q6600 oc'd to 3.0ghz ran the benchwell in 22 min. and numerbois' system was even quicker....

[numerobis]

yes, monochrome, you're right... there are some curious entries in the list

jdp
Apple MacPro Quad 2.66 3 1.5 (32bit) 25m 22s 423.08
jdp Apple MacBookPro 2.16 3 1.5 (32bit) 25m 22s 423.08
mistyped values? - they look the same

these values are not possible ...with this clocks ...neither on osx
pmore78 Apple MacBookPro 2.33 2 1.5 (32bit) 32m 26s 215.1
pmore78 Apple MacBook 2.00 2 1.5 (32bit) 39m 11s 178.11
compared to other macbookpro's
3dworks Apple MacBookPro 2.16 2 1.5 (32bit) 58m 39s 183.05

and this one is much too slow...
AI_Intel Intel Core 2 Extreme QX6700 2.66 2 1.5 (32bit) 43m 59s 244.11 [/quote]
wrong thread number? the test is set to 2 by default

[-Adrian]

Sorry, I wanted to make sure it listed them as dual quad cpu's.

Good point, i'll see that i fetch the amount of cpus as well to remove any uncertainty. edit: //fixed

@others: I agree that some entries are a bit fishy. I've had several obviously false entries, but it's generally hard to judge what's realistic. I don't wanna kick anyone out without being certain. People can always contact me when they accidentally mistyped something, it's easily fixed.

[-Adrian]

How is it that a stock e6600 dual core is getting nearly the same rendering times as QX6700? Of course I don't really care, I did the tests for my own verification of price/performance ration...but it's annoying.

I think you might have run it using 2 threads, making it virtually equal to an E6600, which has the same cores.

[jdp]

yes, monochrome, you're right... there are some curious entries in the list

jdp
Apple MacPro Quad 2.66 3 1.5 (32bit) 25m 22s 423.08
jdp Apple MacBookPro 2.16 3 1.5 (32bit) 25m 22s 423.08
mistyped values? - they look the same

wrong thread number? the test is set to 2 by default

I apology, did it in a rush as always, and got a firefox crash while doing it, restored the session but the entry was already submitted. the correct one is the 1st one... BTW, an option to login and correct mistakes would be nice...

[monochrome]

Just overclocked to 3ghz and my time dropped to 37 min.

Render settings:

- render core : RS1
- render version : Maxwell Render 1.5
- desired rendering time : 125h00m
- desired sampling level : 15
- render resolution : 960 x 768
- using 2 threads
- illumination layers:

I just used the default...I'm trying it right now with 4 threads. Thanks for the heads up...

[-Adrian]

I apology, did it in a rush as always, and got a firefox crash while doing it, restored the session but the entry was already submitted. the correct one is the 1st one... BTW, an option to login and correct mistakes would be nice...

As i said, you can always contact me via mail (quite a few did) and i'll change it. I decided against user administration, because in my experience nobody wants yet another password/account to remember. These entries are only done once in several months and i think this open approach is best suited for that. Haven't encountered any spam, i assume no bot or spammer has the patience to enter sensible values

@monochrome: if you get different results with 4 threads (you can bet on that), just enter them like before and i'll remove your old entry
The scene defaulting to 2 threads is a bug in Maxwell Studio, i wish i could help it.

[monochrome]

Apologies for the accusations. I was under the impression that Maxwell auto detected CPUs and allocated threads accordingly.

I ran the test again, over clocked @3ghz, and now I got 20.5 minutes.
By the way, I'm curious about the heat gains by over-clockers and stability issues.

Mine is running @ ~48 Cidle and ~70C under load. So far so good but Orthos stress test returns memory errors.

[b-kandor]

Orthos should not error, you may need to up your vcore a little - but 70c seems hot? Are you using a stock cooler?

My q6600 is also at 3.0ghz but has a zalman 9700cnsp (I think that's the number). And it's only 54c when rendering in maxwell, orthos or prime runs hotter than that, but only 60c.