Benchmarks and comparisons needed from the 3d gurus :)

Zarich

Newcomer
I have a geforce 3 ti500. I have been hearing that the 23.12s are the best drivers for the ti500 or any of the geforce 3s.
I was hoping someone could do some comparisons to the new 40s both speed and IQ wise to show which set really is the best for the geforce 3 series.
 
Just a word of warning, the new 40's are quite buggy at the moment. Probably wouldn't be that great of an idea to upgrade, anyway.

However, I would expect that the 40-series will prove to be quite good for the GF3's once they mature...
 
I suck at doing this kind of stuff doom. I also dont have all the resources many of the people here do.
 
Zarich said:
I have a geforce 3 ti500. I have been hearing that the 23.12s are the best drivers for the ti500 or any of the geforce 3s.
I was hoping someone could do some comparisons to the new 40s both speed and IQ wise to show which set really is the best for the geforce 3 series.
you want others to tell you, whats best for you ?

[of category: the thread doesnt belong to this forum]
 
im trying to find out what works and what doesnt.
The crew here has always seemed to been good and benchmarking and image comparisons. I on the other hand dont have a clue.
I can run 3dmark, but everytime I run it I get different results. So I dont have a clue what that means.

Currently what I can tell the 23.12s run faster then the 40.41s and everything in between, but that doesnt seem to make sense. That would indicate that for the geforce3 series that nvidia has not improved the drivers since the 23.12s. I have been asking around and finding most people come up with the same results. So I was hoping the gurus around here who seem to be able to do things in a controlled environment could do speed and IQ comparisons to tell me what the real deal is. Apparently this was to much to ask.
 
Zarich I think you need to look at what you are asking here...Benchmarking is time consuming..

Load a driver
Run the games
Take Screen Shots
Upload them to a Server
Collect the Data

Then uninstall that driver and

Load a driver
Run the games
Take Screen Shots
Upload them to a Server
Collect the Data

Thats alot to ask, why not ask Mike Chambers @ Nvnews to do a review ..he has got the banwdith and probably access to all the hardware. :-?
 
okay doom.. maybe thats what I need to do.

The comparisons everyone is doing now is for the geforce 4 and based on recent drivers. Very few have said.. but hey.. the geforce 3 works better at this level instead.

Ill send mike and email and see if he thinks that would be workable. I really think geforce 3 owners would be in for a bit of a supprise overall.
 
hmm the only thing is game compatibility, really older Dets may be faster on the TNT2 for instance, but up to date Det's may fix issues with new games. It isnt just all about performance is it?
 
Not all about speed? Agreed!

It certainly isn't all about speed. What about reliability, image quality, absence of artifacts (stray polygons, mis-drawn textures etc).

People bang on about framerates as if they were their life's blood! Obviously it's important but does it really matter if a game runs at 1280x1024 at 130fps or 150fps? IMHO the difference in quality between the different levels of Anisotropic Filtering / Quincunx / FSAA are so minor that they should be largely ignored when considering the overall performance hit.

If you're running Detonator drivers with a GF3 and a P4 / XP then the chances are you can run absolutely any game on the market with acceptable (80+) framerates and special effects enabled. After this, surely it is the quality of the frames that is paramount?? Not to mention driver stability... what good is 3% faster framerate if you have to reboot your machine every 30 mins????

It would be interesting to hear other views... bring them on!!!!
 
People bang on about framerates as if they were their life's blood! Obviously it's important but does it really matter if a game runs at 1280x1024 at 130fps or 150fps? IMHO the difference in quality between the different levels of Anisotropic Filtering / Quincunx / FSAA are so minor that they should be largely ignored when considering the overall performance hit.

Personally speaking I would not agree with the above. The difference in quality is noticable and at times the performance is too low to cope with the higher quality modes. The overall performance hit should be IMO playable or unplayable and that is subjective from person to person - as is IQ to a degree.
 
This is exactly why I came here and said.. he can someone do the research.. because all I can do is framerates and numbers.. and based on that I see the 23.12s are better.. and yet theres more to it. This is why I need someone with expertise to do this study.
:)
 
Back
Top