first r420 review leak

[trinibwoy]

Let's see...... you make some dumbass comments that are so off the wall stupid that it pisses people off then you post telling them to calm down?!?!

They weren't off the wall, they were off my keyboard.

Looks to me like you are trying to piss people off just because NVDA's 6800 is bigger, hotter, louder, less OEM friendly, more PSU crippling, and heavier than the X800.

Did I mention it is also slower, has worse IQ, and requires hacked-out-the-ass drivers to stay competitive?

WOW!!! get a life dude. you're trying to brand the proud owner of a 9800PRO (and maybe X800PRO) as an nvidia fanb0y.....shameful

Wow to bad about that NVDA stock going down 34% these last few weeks. Get out now before you lose another 20% on top of that.

How LAME was that. The entire market is suffering and 'the last few weeks' arent indicative of jack shit. So STFU ok?

 

[BigGamer X]

Why not both then I would be even happier no ? Really though if ati kills nvidia then nvidia will have to sell the 6800 cheaper, I am much more excited about doom3 than halflife so it would be to my interest to have the 6800 reduced in price if it performs better even in that select application.

Looking at how all of these cards killed the 5950u and 9800xt, I would think even the x800 pro will run Doom 3 just fine. It's the games about a year from now that are even more demanding that I am thinking about. The x800xt and 6800u seem a little more future proof, depending upon which games you like. (yes the pro beat or tied the 6800u in some games, but overall 6800u looks stronger) If the x800pro is ONLY 299, I think that's the way I will go. If it is 399.....I figure I might as well get the XT then :?

 

[ChrisW]

To all the guys saying that all TWIMTBP games are crap benchmarks please provide an ATI approved list for a 'fair' comparison

Wish I was around when Nvidia was on top and they were the ones moaning about the 9700PRO

Who said they were crap benchmarks? I'm only pointing out that Tom's Hardware is the only site to review 100% nVidia optimized games (look at the 6800 reviews). The other sites had no problem finding games to benchmark that were not nVidia optimized to benchmark. There is nothing wrong with benchmarking a few games that are nVidia optimized, but when you are obviously using the list of games provided by nVidia to benchmark everything, it shows your bias.

 

[trinibwoy]

Why not both then I would be even happier no ? Really though if ati kills nvidia then nvidia will have to sell the 6800 cheaper, I am much more excited about doom3 than halflife so it would be to my interest to have the 6800 reduced in price if it performs better even in that select application.

Hmmmmm Twisted Logic but I like

 

[Blastman]

Even Halo is using hand written shaders from nVidia themselves. This is why it's important to take Tom's Hardware with a grain of salt and wait for some more credible sites to release their reviews before making any kind of judgement here.

I’m suspecting Halo is running a …“lowered”… shader path for NV cards just like Farcry. In this Portuguese review … oczone … the 5700U is not rendering the ice-gun properly (second set of pictures) compared to the 9600XT. The 5700U is also reducing 8AF down to about 2AF compared to the 8AF of the 9600XT.

 

[Doomtrooper]

To all the guys saying that all TWIMTBP games are crap benchmarks please provide an ATI approved list for a 'fair' comparison

Wish I was around when Nvidia was on top and they were the ones moaning about the 9700PRO

I would not be getting cocky here, you have seen little. Keep a comparison of THG numbers vs. other sites. THG was the only site that showed a 5800 U beating a 9700 Pro. Now, if he is benchmarking properly how could the entire internet be wrong

 

[dr3amz]

Let's see...... you make some dumbass comments that are so off the wall stupid that it pisses people off then you post telling them to calm down?!?!

They weren't off the wall, they were off my keyboard.

Looks to me like you are trying to piss people off just because NVDA's 6800 is bigger, hotter, louder, less OEM friendly, more PSU crippling, and heavier than the X800.

Did I mention it is also slower, has worse IQ, and requires hacked-out-the-ass drivers to stay competitive?

WOW!!! get a life dude. you're trying to brand the proud owner of a 9800PRO (and maybe X800PRO) as an nvidia fanb0y.....shameful

Wow to bad about that NVDA stock going down 34% these last few weeks. Get out now before you lose another 20% on top of that.

How LAME was that. The entire market is suffering and 'the last few weeks' arent indicative of jack shit. So STFU ok?

can you two take it to PM? don't need any fanboi/related arguments ta (futuremark boards are good for that, full of nvidiots

keep it happy - its a good day... today!!! 00:25 - 14hrs to go w00t!

 

[Ailuros]

No, my math is not bad. Running temporal AA at 90fps requires your framerate to the same as your monitor refresh rate most of the time.

When framerate fluctuates at higher resolutions, the images will still look good. When it fluctuates with temporal AA, you'll get noticable flickering.

Frankly, I'd rather play at 1024x768 @ 90fps without temporal AA and vsync off. (and before I get accused of this being an anti-ATI issue, it's not. Temporal AA can be implemented on other cards too.)

Temporal AA looks usable for Call of Duty, UT2003, and a bunch of other 100+fps (with minimum FPS > 70fps) games. I think you're kidding yourself if you think it will be great on FarCry.

I'd exclude UT2k3 also from that list, unless you run around shooting walls in indoor maps. I tested it thoroughly in Rustatorium because the countless of poles sticking out makes it a lot easier to see differences between patterns. I get rid of edge crawling in 1024*768 with 4xTaa, as long as I enter "killpawns" in the console

 

[BetrayerX]

Only on PS2.0. On DX8 games, it wasn't. Point is, if someone could design a 3D pipeline that was 4 pipes, but ran at 2Ghz vs one that was 16 pipes but ran at 500Mhz, and they had equivalent performance, I don't think you could say one was better than the other. Hell, the 2Ghz version might even use far less transistors.

Yes but why would u design a dx 9 card to run dx 9 like crap and dx 8.1 very fast .

You make me laugh with your defence of the nv30. It was crap. They messed up.

You make me laugh with your inability to read and think.

Defense? The point is, a card with a smaller number of pipelines but higher clocks could have just as good performance with a card with more pipes but lower clocks. I use the NV3x as an example, because until the X800 is released, we don't have any hard data except the NV3x vs R300, and because in many scenarios, it did perform as well as an R300. Yes, there were mistakes in the design, but that's not relevant to the general argument. NV3x's poor PS2.0 performance had little to do with its clocks or # of pipelines, and had more do with with register storage and Nx2 organization.

You forgot to mention the NV magic driver pixie dust.

 

[DemoCoder]

It would be interesting to get DT, HB, dr3amz, et al, to committ to a list of approved games prior to tommorow, so they couldn't complain if the ones they picked turned out to show ATI worse. That's even assuming that they have NDA info which HB in particular has been beating his chest about.

DT already added Max Payne2 on the list (curious, a non-PS2.0 game)

 

[Slappi]

Wow to bad about that NVDA stock going down 34% these last few weeks. Get out now before you lose another 20% on top of that.

How LAME was that. The entire market is suffering and 'the last few weeks' arent indicative of jack shit. So STFU ok?

I nailed you dead on!

NVDA stock owner...... whoever would of thought it.

Don't worry.... it will bounce off 18. For a bit.

 

[dr3amz]

No, my math is not bad. Running temporal AA at 90fps requires your framerate to the same as your monitor refresh rate most of the time.

When framerate fluctuates at higher resolutions, the images will still look good. When it fluctuates with temporal AA, you'll get noticable flickering.

Frankly, I'd rather play at 1024x768 @ 90fps without temporal AA and vsync off. (and before I get accused of this being an anti-ATI issue, it's not. Temporal AA can be implemented on other cards too.)

Temporal AA looks usable for Call of Duty, UT2003, and a bunch of other 100+fps (with minimum FPS > 70fps) games. I think you're kidding yourself if you think it will be great on FarCry.

I'd exclude UT2k3 also from that list, unless you run around shooting walls in indoor maps. I tested it thoroughly in Rustatorium because the countless of poles sticking out makes it a lot easier to see differences between patterns. I get rid of edge crawling in 1024*768 with 4xTaa, as long as I enter "killpawns" in the console

i guess it depends on the game.

for some reason i dont see any flickering in Farcry using 2xTempAA @ 1024*768. If I use 3xT then I do.

however, watch 3dmark03 with 2xT and I see flickering on the rocks for example on mother nature.

at 3xT there is also quite a lot of `odd/black` pixels in places they shouldnt be - especially in game test 3 (orcs) :?

 

[trinibwoy]

To all the guys saying that all TWIMTBP games are crap benchmarks please provide an ATI approved list for a 'fair' comparison

Wish I was around when Nvidia was on top and they were the ones moaning about the 9700PRO

I would not be getting cocky here, you have seen little. Keep a comparison of THG numbers vs. other sites. THG was the only site that showed a 5800 U beating a 9700 Pro. Now, if he is benchmarking properly how could the entire internet be wrong

Sheesh....cocky? What is there to be cocky about? No matter which card is faster I win. Not everybody here has a fave IHV you know...... 8)

 

[Ailuros]

I actually can't run lower than 1280x1024 or 1600x1200, because of the LCD monitors I use. Running in lower resolution invokes the monitor to use upscaling interpolation which frankly looks bad compared to native resolution.

Albeit this being a CRT it's pretty much the same here too. Anyway I see this thread has turned into a potential flamewar and I'm not really fond of that kind of crap.

Carry on kids.

 

[trinibwoy]

Wow to bad about that NVDA stock going down 34% these last few weeks. Get out now before you lose another 20% on top of that.

How LAME was that. The entire market is suffering and 'the last few weeks' arent indicative of jack shit. So STFU ok?

I nailed you dead on!

NVDA stock owner...... whoever would of thought it.

Don't worry.... it will bounce off 18. For a bit.

HAHAHAHAHA. Nah I was thinking of buying it for real. But I ditched MSFT last Tuesday and split the money between AMD and ATYT. You're dead wrong dude

 

[dr3amz]

It would be interesting to get DT, HB, dr3amz, et al, to committ to a list of approved games prior to tommorow, so they couldn't complain if the ones they picked turned out to show ATI worse. That's even assuming that they have NDA info which HB in particular has been beating his chest about.

DT already added Max Payne2 on the list (curious, a non-PS2.0 game)

i don't necessarily dislike the games, i just don't think they are right for showing off the very latest next gen hardware.

this happened with the nv30 launch too - benchmarks showed it faster than ati's card in dx7/8 games... wow? this was meant to be a dx9 part. also these games are very cpu limited, so what's the point? this isnt a game review or cpu review - its a graphics card review... why not post some benchmarks about how fast windows updates its desktop with some IQ shots for good measure?

again - to me it looks like nv are gonna suffer in the dx9 side of things, thats all

 

[Rican]

i wonder how much more headroom does the x800 have for o/c

 

[termikk]

now one of the big sites always leaks almost all the review...hmmm...
looks to be on purpose
so what does thg gain/lose? what about ati/nvda?

for the ones waiting for the text, wait for dave, thg's is usually moronic, so no help there!

Since it's a (p)review I'll discount the leaked drivers (by nvda as usual), but really shipping drivers should be used, and after that official ones.
the story isn't told yet...

 

[dr3amz]

I actually can't run lower than 1280x1024 or 1600x1200, because of the LCD monitors I use. Running in lower resolution invokes the monitor to use upscaling interpolation which frankly looks bad compared to native resolution.

Albeit this being a CRT it's pretty much the same here too. Anyway I see this thread has turned into a potential flamewar and I'm not really fond of that kind of crap.

Carry on kids.

yer, someone slap them

 

[Chisholm]

The Spanish site Meristation has some more info:

http://www.meristation.com/sc/articulos/articulo.asp?c=HARD&cr=5218