Page 1 of 1
Lowest bm
Posted: Sat Mar 17, 2007 2:01 am
by Thomas An.
Hi all,
We all talk about how to improve this and that in order to get higher benchmarks in Maxwell ... but ... out of curiosity, what is the lowest benchmark you have ever gotten on your fastest machine ?
Posted: Sat Mar 17, 2007 2:03 am
by 4 HeRo
Amd 2800 1.5GBram 2.2Bm
Posted: Sat Mar 17, 2007 2:16 am
by Thomas An.
4 HeRo wrote:Amd 2800 1.5GBram 2.2Bm
Cool
... How did you do that ?
Posted: Sat Mar 17, 2007 7:38 am
by b-kandor
I'm thoroughly confused by this question
Posted: Sat Mar 17, 2007 2:46 pm
by kizo
sometimes i get the BM 0.something and it keeps like that as long as i let it..
i haven't figured out why this heppens but when it does the computing is done by 1% of proc. power.
the same scene with exactly the same settings renders normally when started again an BM goes up to 70 cca.
Posted: Mon Mar 19, 2007 6:29 pm
by Tim Ellis
0.2 is my lowest. Not a good result. 4kx4k render.
Tim.
Posted: Mon Mar 19, 2007 11:08 pm
by ivox3
Thomas, ....this is a very unscientific inquiry --- I'm a little surprised.
Posted: Wed Mar 21, 2007 10:55 am
by Hybaj
ivox3 wrote:Thomas, ....this is a very unscientific inquiry --- I'm a little surprised.
Exactly
Posted: Mon Oct 01, 2007 5:09 pm
by Tim Ellis
My new record:-
Benchmark of 0.009
8 core 4gb x64bit render node. 5000x2810 ml render.
3.5GB mxi.
Tim.
Posted: Mon Oct 01, 2007 5:23 pm
by Thomas An.
Tim Ellis wrote:My new record:- Benchmark of 0.009
ouch
Posted: Mon Oct 01, 2007 6:36 pm
by sandykoufax
Tim Ellis wrote:My new record:-
Benchmark of 0.009
8 core 4gb x64bit render node. 5000x2810 ml render.
3.5GB mxi.
Tim.
I want to know the result of it.
Posted: Mon Oct 01, 2007 6:51 pm
by Tim Ellis
sandykoufax wrote:Tim Ellis wrote:My new record:-
Benchmark of 0.009
8 core 4gb x64bit render node. 5000x2810 ml render.
3.5GB mxi.
Tim.
I want to know the result of it.
I sacked the render as it took toooooo long to even see a result other than a black screen, other nodes had 0.008 with the highest one at 0.01. pftt!
Result was due to 15 very high poly emitters, some with 7900 tris per object.
Have now changed these to single poly emitters shining through the high poly versions and will post resulting BM when rendered.
Tim.
Posted: Mon Oct 01, 2007 7:05 pm
by Fernando Tella
Other thing that reduces BM a lot is crossing clipped planes. Some time ago I was trying to make a tree crossing around 300 clipped planes at the same point and the benchmark falled down to 1-2 in a 640x480 image.
Posted: Tue Oct 02, 2007 3:14 pm
by Leonardo
kizo wrote:sometimes i get the BM 0.something and it keeps like that as long as i let it..
i haven't figured out why this heppens but when it does the computing is done by 1% of proc. power.
the same scene with exactly the same settings renders normally when started again an BM goes up to 70 cca.
I get the same thing some times 0.0000000000000something
I imagine it happen when Maxwell consumes all the ram in voxelization
Next time I render at a lower resolution it works fine
EDIT- Hey, It just happen... I got a 0.02
