Hexus.net offers first glimpse of RV360 vs. NV36

[GTaudiophile]

Kyle from HardOCP likes to post over at Rage3D from time to time. According to a recent post of his, Hexus.net's benches between RV360 and NV36 are accurate and even use new WHQL drivers from nVidia.

Hexus.net's Benches Begin Here

Looks like nVidia had to quickly make a new core to keep up with RV360, but they were smart to do it for a main-stream card. The catch, from what I've read, is that NV36 will not compete at the same price point at RV360, though. NV36 may debut at $299 compared to RV360, which will debut at $199 (and comes with HL2). I hope ATi or their OEMs counter with at least faster memory.

 

[NFS4]

Niiiiice...too expensive though.

 

[boyRacer]

not bad... 🙂 i dunno about that pricetag though...

 

[GTaudiophile]

NV36 is far more competitive but still doesn't win every bench.

nVidia should not forget, that if they decide to sell NV36 at $299, that's like $50+ more expensive than a 9800 NP or 9700 PRO.

 

[rbV5]

NV36 may debut at $299 compared to RV360

Geez, that can't be true.

 

[DaveSimmons]

Originally posted by: rbV5

NV36 may debut at $299 compared to RV360

Geez, that can't be true.

Maybe nVidia management hasn't put down the pipe yet --
- "no one will notice if we cheat in 3DMark"
- "no one will notice if we degrade image quality for a few extra FPS"
. . .
- "reviewers won't notice the det51's are fast but don't really work"
. . .
- "people will pay 9800 PRO prices for a card that's half as fast for DX9 games"

 

[Johnbear007]

Originally posted by: DaveSimmons

Originally posted by: rbV5

NV36 may debut at $299 compared to RV360

Geez, that can't be true.

Maybe nVidia management hasn't put down the pipe yet --
- "no one will notice if we cheat in 3DMark"
- "no one will notice if we degrade image quality for a few extra FPS"
. . .
- "reviewers won't notice the det51's are fast but don't really work"
. . .
- "people will pay 9800 PRO prices for a card that's half as fast for DX9 games"

 

[yhelothar]

who still gives a sh1t about 3dmark?!? doesn't everyone agree it's the most useless benchmark ever?
who cares about nvidia's buggy beta det 51s?
what dx9 game is it still only half as fast with the fixed dx9 performance in the det 52s?
and don't tell me the dx9 is inferior just cuz nvidia did something to the drivers, cuz i don't think anyone could see a difference..
and i seriously doubt this would debut at $299... the geforcefx 5900 are already down to $235... I can't see this going over $200

 

[DaveSimmons]

> who still gives a sh1t about 3dmark?!? doesn't everyone agree it's the most useless benchmark ever?

cheating is still wrong, mmkay?

> who cares about nvidia's buggy beta det 51s?

nVidia pushed hard to get reviewers to use 50-51 for benchmarks, before actually fixing rendering for v52.

> what dx9 game is it still only half as fast with the fixed dx9 performance in the det 52s?

9800 pro vs. fx5700 not 5900.

> and don't tell me the dx9 is inferior just cuz nvidia did something to the drivers, cuz i don't think anyone could see a
difference..

Yes, the fake trilinear filtering is supposed to be pretty close to the real thing, only being noticeable on some maps in some games, and the dx9 reduced-quality shaders won't be noticeable all the time either.

> and i seriously doubt this would debut at $299... the geforcefx 5900 are already down to $235... I can't see this going over $200

We'll see. But if they do try to push it at $299 it will be a good sign that nV is still under the influence.

 

[yhelothar]

cheating is still wrong, mmkay?

and they didn't even hide it.... in fact they publically announced the cheat to try to discredit futuremark, who tried to discredit nvidia

nVidia pushed hard to get reviewers to use 50-51 for benchmarks, before actually fixing rendering for v52.

nahh just for half life 2

9800 pro vs. fx5700 not 5900.

yeah and you seriously believe that fx5700 will cost $50-$75 more than it's sister card?

Yes, the fake trilinear filtering is supposed to be pretty close to the real thing, only being noticeable on some maps in some games, and the dx9 reduced-quality shaders won't be noticeable all the time either

and has it not been fixed in the det 52s? true that nvidia needs to use FP16 to remain competitive performance wise.. but I personally can't see much of a difference. You can always opt for higher precision with lower performance.
I agree that this is a big downside on nvidia cards... but I can't really see the difference anyways, at least not now..
I personally went from a 9700pro to a geforcefx5900.. the only image quality difference I CAN ACTUALLY SEE is that the 9700pro has far superior AA quality.. which then that doesn't even bug me that much since I use 1600x1200 in most of my games...