View Full Version : eVGA e-GeForce 6600 GT PCI-E
Ok, this has been odd. There must be something that I'm unaware of to cause this. Here's what happened.
- I created my own timedemo in the admin map
- I benched my timedemo and the average was 42fps
- Then I thought I'd try benching the demo1 timedemo again...
- My results were then as they should be, 33fps
I know for certain that my past benchmarks of demo1 were at the correct AA/AF and resolution in my autoexec.cfg (and verified via the in-game menus and console)
The complete results for my timedemo are below:
0xAA/0xAF @ 1600x1200: 89.5fps
4xAA/8xAF @ 1600x1200: 42.0fps
I'll be updating my review with my (now correct) results for demo1 though I cannot explain why I was getting ~73fps prior to now. Clearly, the AA/AF settings were not being retained previously yet somehow they now are. I'll just chalk it up to the 71.24 drivers for now.
- Then I thought I'd try benching the demo1 timedemo again...
- My results were then as they should be, 33fps
Exactly what i expected :D .. still not bad for such high settings .
this just show how if the game is properly optimized for the NV40 will benefit from the SLI .
one need to think about the upcoming games that are supported by nVIDIA like Stalker & BF2 .. they will result in an exiting experiance for the SLI owners .
i hope i'll be one of them in that time :)
Yes, I'm glad I finally received the results that would be expected and they are still respectable considering.
Grapes0Wrath
01-26-05, 03:17 AM
Anyone else notice that AA/AF had absolutely no impact on the other resolutions in D3? :screwy:
http://img190.exs.cx/img190/8647/nvnewsaaaf4fx.jpg
I think something else is a little amiss other than just the 16x12 and SLI numbers. If the system were CPU limited I'd expect to see the results as seen in SLI with there being little impact going from Resolution to resolution, but the GF6600GT single card experiencing no impact from AA/AF, but experiencing resolution hits at the same time would indicate that it's not strictly CPU limited throughout, and therefore you should see AA and AF impact your results. Especially when every other reviewer has experienced performance hits at those same resolutions and similar setup (CPU/memory);
http://www.ixbt.com/video2/sli-d2.shtml#p26
Now if somone can explain that one, that would be nice, then I can get back to skiing. (snowcool)
But then again, as was pointed out to me today by the author, I'm not a reviewer myself and therefore unworthy of questioning the results. :nono:
Wow, yeah Clay, that was worth registering for. No, wait no it wasn't! :thumbdwn:
Ciaola!
This is an error on my part. I was not aware of this particular article that you just now pointed out to me via email: http://www.overclockercafe.com/Articles/NVidia_6600GT/NVidia_Anomaly.htm But it sheds some light on this. Your twisting of my response to your email is incorrect. I never said you were unworthy of questioning results. I'm sure that if you ask around you'll find that I do not treat others that way. I was questioning your heated tone and how you seemed to not have any appreciation or understanding for mistakes that are inevitiable at times as we've all seen on countless other reviews.
I'm now working on an update to clarify this error in my review.
saturnotaku
01-26-05, 08:59 AM
I was questioning your heated tone and how you seemed to not have any appreciation or understanding for mistakes that are inevitiable at times as we've all seen on countless other reviews.
Maybe when he's skiing he'll fall and break his neck.
Maybe when he's skiing he'll fall and break his neck.Hey now, we don't want that.
I'll be the first to admit any mistake I make and then do my best to correct it. I've always appreciated this quote along these lines:
If you're not making mistakes, then you're not doing anything. I'm positive that a doer makes mistakes. -- John Wooden
The guy is correct , those are obvious mistakes .. i didnt mention them because i thought Clay is going to correct them anyways after correcting the 1600x1200 results .
The guy is correct , those are obvious mistakes .. i didnt mention them because i thought Clay is going to correct them anyways after correcting the 1600x1200 results .Clearly, the onus is not on you. However, you did overlook them too, you never mentioned anything but the demo1 results. Anyone could claim otherwise after the fact.
Clearly, the onus is not on you. However, you did overlook them too, you never mentioned anything but the demo1 results. Anyone could claim otherwise after the fact.
I have to admit that i overlooked them . thats my mistake & i dont know why i did !! :|
maybe because i thought you were 100% sure about them or simply because i was interested on other games scores like FarCry since Doom3 is well known for its superior performance on the NV4x series .
In way or another that was a stupid mistake i did by overlooking them .
I have to admit that i overlooked them . thats my mistake & i dont know why i did !! :|
maybe because i thought you were 100% sure about them or simply because i was interested on other games scores like FarCry since Doom3 is well known for its superior performance on the NV4x series .
In way or another that was a stupid mistake i did by overlooking them .
Well, like I said, the onus is squarely on my shoulders so you have nothing to apologize for. I just wasn't going to let you get away with claiming that you noticed them earlier. :angel:
saturnotaku
01-26-05, 09:44 AM
It's not the fact that the mistake was pointed out that is the problem, it's the manner in which it was done. It's called constructive criticism. There's a difference between:
1) Hi Clay, I just noticed in your review some data that appears to be inconsistent. If you take a look at your Doom 3 results, I see some of the numbers are exactly the same between no AA/AF and 4x AA/8x AF. Can you explain those results or was that an error on your part?
and
2) omfg u suX0r luk @ ur doom3 resultz ur a retard
See what I'm getting at? ;)
Yes, I do. But considering the likely source it's now apparent to me why the e-mail was in the flippant tone that it was. It doesn't excuse that approach but it does make more sense now. ;)
Hmm i wonder what happened with those invalid results ... its been almost 2 weeks & i still see nothing changed/fixed ! :p
Hmm i wonder what happened with those invalid results ... its been almost 2 weeks & i still see nothing changed/fixed ! :pHmm...
short memory (http://www.nvnews.net/vbulletin/showthread.php?t=44467#post522141)? :p
I do not have the motherboard and 6600GT cards and have not had them since shipping them out on the morning of January 26th.
Hmm...
short memory (http://www.nvnews.net/vbulletin/showthread.php?t=44467#post522141)? :p
I do not have the motherboard and 6600GT cards and have not had them since shipping them out on the morning of January 26th.
Yeah i remember that .. but i thought you just needed sometime to look on the problem & fix it .
anyway i believe Chris's new "SLI Investigation" should fill-in those missing blanks for Doom3's benchies ;)
Yeah i remember that .. but i thought you just needed sometime to look on the problem & fix it .
anyway i believe Chris's new "SLI Investigation" should fill-in those missing blanks for Doom3's benchies ;)
I'm a busy guy, I had time allotted to that review and then simply ran out under the circumstances. But yes, Chris's work looks great and goes into more detail.
EDIT: BTW, I discussed this anomoly with Chris via IM and he it stumped him as to what would cause it and/or a workaround.
ChrisRay
02-07-05, 04:40 PM
It happened to me actually. But rebooting restarting the game fixed it. Or reloading the map.
It happened to me actually. But rebooting restarting the game fixed it. Or reloading the map.After we'd discussed it I assume? I tried the obvious of rebooting and reloading but I was always getting the same results regardless. I was using the 71.24 drivers as well so perhaps that had something to do with it.
ChrisRay
02-07-05, 07:56 PM
After we'd discussed it I assume? I tried the obvious of rebooting and reloading but I was always getting the same results regardless. I was using the 71.24 drivers as well so perhaps that had something to do with it.
Yes after we discussed it. I spoke to rage about it the other night. I didnt try 1600x1200 though. It only occurred at 1280x1024 with 4xAA. But I was able to get it to work.
Yes after we discussed it. I spoke to rage about it the other night. I didnt try 1600x1200 though. It only occurred at 1280x1024 with 4xAA. But I was able to get it to work.
Interesting. Well I was only testing (initially) at 1600x1200 and that's when the issue was first noticed and this was at 4xAA/8xAF.
vBulletin® v3.7.1, Copyright ©2000-2012, Jelsoft Enterprises Ltd.