Brent....what a fine review you have posted.....

[Brent]

Brent, let me get back to you on your UT2003 demo. There were some significant bugs with maps like Oceanic (has to do with the shark entity, used to cause crashes, now it might just cause huge performance drops). The crash one was fixed, I know, but the performance one might not be.

yeah let me know on that, first time i've heard of any problems with that map

 

[Snyder]

Regarding the Image Quality comparison with FS2004:
Is this really 8xAF with the 5700U? If yes than it's worse than I thought - looks more like 2x or 4x to me...

 

[Brent]

Regarding the Image Quality comparison with FS2004:
Is this really 8xAF with the 5700U? If yes than it's worse than I thought - looks more like 2x or 4x to me...

8XAF was selected from the nvidia control panel in the drivers

that is what is seen in game with 8XAF selected from the control panel of the drivers

it's been that bad with every driver release since the GFFX cards have been out, never has been fixed or improved

so until it is, i'll keep right on showin it

 

[Ailuros]

Regarding the Image Quality comparison with FS2004:
Is this really 8xAF with the 5700U? If yes than it's worse than I thought - looks more like 2x or 4x to me...

I was just thinking the same thing. Granted I never touched FS2k4, just FS2k3 (ok the first has more clouds heh...), yet I never recall the latter to look as bad on my former GF4Ti.

Brent,

Any chance of additional feedback if 8xS with even just 2xAF is playable on the 5700U? FS2k3 was very CPU bound and I doubt it's that much different with FS2k4.

 

[THe_KELRaTH]

I thought the review was very well put together and informative. I like the emphasis on best gameplay w/ quality rather than just fps.

When you mention gameplay I assume you still test with Vsync Off therefore do you find any one card better or worse with pan / screen tearing problems or are both much the same and do both cards feel the same with Vsync On.

Cheers Kel

 

[Blastman]

Besides your review, lately I’m seeing more comments that with NV cards some of the lighting is being rendered too bright in some games. It was another 5700U article. I’m left wondering if this is a result of NV lowering the precision of their drivers too much to gain additional performance so they’re lost dynamic range and some settings will be over-rendered.

http://www.techwarelabs.com/reviews/video/inno3d_5700ultra/index_4.shtml

Lighting when using the 5700 Ultra was unusual and not correctly presented. In Halo, NFS Underground, and at times in Quake 3 where lighting from lights, headlights, and gun mounted flashlights was pointed at objects, the transition from light to dark was very prounounced and not even. It looks artifical and captures your attention easily. This could be due to drivers or GPU rendering of the lighting, either way this effect was present with any level of AA and AF, on or off.

 

[vb]

Besides your review, lately I'm seeing more comments that with NV cards some of the lighting is being rendered too bright in some games. It was another 5700U article. I'm left wondering if this is a result of NV lowering the precision of their drivers too much to gain additional performance so they're lost dynamic range and some settings will be over- rendered.

http://www.techwarelabs.com/reviews/video/inno3d_5700ultra/index_4.shtml

Quote:

Lighting when using the 5700 Ultra was unusual and not correctly presented. In Halo, NFS Underground,

and at times in Quake 3 where lighting from lights, headlights, and gun mounted flashlights was pointed at objects, the transition from light to dark was very prounounced and not even. It looks artifical and captures your attention easily. This could be due to drivers or GPU rendering of the lighting, either way this effect was present with any level of AA and AF, on or off.

Maybe it is related to this

The basic premise of UltraShadow is that it allows developers to program a depth bounded region that will restrict the level of lighting calculation to a specified range of the Z Buffer, which corresponds to the area that the lighting source is most likely to affect. The net effect of this is that it minimises the shadow calculation to just that region, thus removing some unnecessary work calculating regions that will not be seen anyway.

 

[Tim Murray]

there are no games that use UltraShadow, and the only one might be Doom 3 (and D3 engine games)

 

[vb]

there are no games that ask for brilinear also. It is not imposible for to NV to set a default range for a light to be active and, as long as the app doesn't change it... there you go.

It would be interesting to see if 5200/5600/5800 have the same issues.

 

[Brent]

Regarding the Image Quality comparison with FS2004:
Is this really 8xAF with the 5700U? If yes than it's worse than I thought - looks more like 2x or 4x to me...

I was just thinking the same thing. Granted I never touched FS2k4, just FS2k3 (ok the first has more clouds heh...), yet I never recall the latter to look as bad on my former GF4Ti.

Brent,

Any chance of additional feedback if 8xS with even just 2xAF is playable on the 5700U? FS2k3 was very CPU bound and I doubt it's that much different with FS2k4.

fs2002 you mean, there was no fs2k3

as far as i know af is fine on the gf4, though slow

the bad af problem in fs2k4 only happens with the gffx series

also, i've noticed other games now with poor af now as well, ut2k3, and ut2k4 namely, i'll have screenshots and comparisons in an upcoming review i'm working on now

 

[Brent]

I thought the review was very well put together and informative. I like the emphasis on best gameplay w/ quality rather than just fps.

When you mention gameplay I assume you still test with Vsync Off therefore do you find any one card better or worse with pan / screen tearing problems or are both much the same and do both cards feel the same with Vsync On.

Cheers Kel

Yes, i play with VSYNC off

note though i've always played with vsync off on my primary computer with every game i play

so i'm kinda use to having vsync always off

i wasn't really looking to see which card had worse tearing or not because its just a given that that will happen at times with vsync off

i know this much, with the settings I used in the games the fps were low enough that tearing was not a huge problem, i think the one game that it was most noticeable in was max payne 2 because it got up to some high fps, but i didn't notice either card being better or worse for tearing

 

[digitalwanderer]

Yes, i play with VSYNC off

BLASPHEMY!!!



HOW can you handle the tearing? Seriously? I can't play with v-sync off ever since I noticed the difference, now there is no going back for me.

I'm actually trying to decide if I should bench games with v-sync on or not, I'm leaning towards not just 'cause it shows performance better but if I do that I'll be including some charts with v-sync on too since I feel it's so essential to the gaming experience anymore.

 

[Tim Murray]

tearing is pretty rare if you don't run at some crappy refresh rate like 60Hz. or, at least, it's not really noticeable.

 

[Brent]

Yes, i play with VSYNC off

BLASPHEMY!!!



HOW can you handle the tearing? Seriously? I can't play with v-sync off ever since I noticed the difference, now there is no going back for me.

I'm actually trying to decide if I should bench games with v-sync on or not, I'm leaning towards not just 'cause it shows performance better but if I do that I'll be including some charts with v-sync on too since I feel it's so essential to the gaming experience anymore.

tearing isn't that bad

i run at 85Hz and i always have some level of AA and AF on at a high resolution which keeps the framerates down below 85 most of the time

so it isn't really a problem

i'd rather do that then have vsync on + triple buffering since i'm using a 128MB video card on my primary computer

 

[Pete]

Heh, I avoid vsync + triple buffering because I'm packing only 64MB, but I guess 128MB can get constricting once you start laying on some AA.

 

[Sazar]

yah I'll chime in a bit here...

:thumbsup: on the review...

I am curious as to whether you have received any further word from gigabyte vis a vis hl2... might help for other people with a similar problem...

also is there a reason you only posted screenshots in certain areas? ie only used to point out differences and anomalies?

the lighting issues and the flight sims shots show some very clearly defined boundaries on the nv card... I'll wait/see wrt what blastman said about the lighting issues but for the AF... when playing the flight sim are you actually seeing the mipmap boundaries as clearly as is shown in the caps?

just curious

 

[Brent]

yah I'll chime in a bit here...

:thumbsup: on the review...

I am curious as to whether you have received any further word from gigabyte vis a vis hl2... might help for other people with a similar problem...

also is there a reason you only posted screenshots in certain areas? ie only used to point out differences and anomalies?

the lighting issues and the flight sims shots show some very clearly defined boundaries on the nv card... I'll wait/see wrt what blastman said about the lighting issues but for the AF... when playing the flight sim are you actually seeing the mipmap boundaries as clearly as is shown in the caps?

just curious

haven't recieved any more word from gigabyte on the hl2 thing

i posted screenshots in only certain areas, i.e. cropping, for bandwidth reasons on the site, but also so you can easily compare image side by side

oh yes, playing the game it is actually that bad, here, see for yourself

here are the full screenshots that i cropped the sections out of

http://home.comcast.net/~b.justice/fs2004_5700u.jpg

http://home.comcast.net/~b.justice/fs2004_9600xt.jpg

the images are large in file size

 

[Sazar]

I was actually thinking of bandwidth reasons when I posted about the screenies but had a brain crap and forgot to include that in me (is it this or that) commentary

btw... have you noticed the mip-map boundaries to be as evident with other driver sets ? out of curiosity...

ta for response btw...

 

[Brent]

I was actually thinking of bandwidth reasons when I posted about the screenies but had a brain crap and forgot to include that in me (is it this or that) commentary

btw... have you noticed the mip-map boundaries to be as evident with other driver sets ? out of curiosity...

ta for response btw...

53.03 and 52.16

possibly others, i don't remember prior to 52.16

 

[martrox]

Jeez, Brent, the 5700 on FS2004 absolutely sucks...... Where's that famous much better ansotropic filtering......., or am I missing something here?