NVIDIA GeForce 6600GT and 6600 (NV43): Part 1 - Performance.
|
Contents
- Official specifications
- Architecture
- Video cards' features
- Testbed configurations, benchmarks,
2D quality
- Synthetic tests in D3D RightMark
- Synthetic tests in 3DMark03:
FillRate Multitexturing
- Synthetic tests in 3DMark03:
Vertex Shaders
- Synthetic tests in 3DMark03:
Pixel Shaders
- Test results: Quake3 ARENA
- Test results: Serious Sam:
The Second Encounter
- Test results: Return to Castle
Wolfenstein
- Test results: Code Creatures
DEMO
- Test results: Unreal Tournament
2003
- Test results: Unreal II: The
Awakening
- Test results: RightMark 3D
- Test results: TRAOD
- Test results: FarCry
- Test results: Call Of Duty
- Test results: HALO: Combat
Evolved
- Test results: Half-Life2(beta)
- Test results: Splinter Cell
- Test results: DOOM III
- Test results: 3DMark03 Game1
- Test results: 3DMark03 Game2
- Test results: 3DMark03 Game3
- Test results: 3DMark03 Game4
- Test results: 3DMark03 MARKS
- Conclusions
FarCry, demo01
GeForce 6600GT vs. OTHER CARDS |
GeForce 6600GT vs. GeForce 6800/6800GT |
GeForce 6600 |
|
|
|
|
|
|
|
|
|
|
|
|
The easiest modes without AA and anisotropy: This game is very difficult for all video cards, so there is a point even in this mode to pay attention to the comparison with R9800PRO/6800LE. More powerful core in 6600GT obviously led it to victory over R9800PRO, though the 256-bit bus in 6800LE still disavowed the NV43 advantage resulting in parity. The victory over X600XT and PCX5900 is obvious and obligate. It's interesting to note that 6600 is also a terrific success. But the battle with the "emulators" is lost for 6600GT. The question is why. Comparison with 6800LE demonstrates a serious role of the memory bandwidth. Considering more than one texture in this test, this is very strange. Again mysterious little differences between NV43 and NV40 in caches, etc? The question remains unanswered.
With enabled AA: The problem is still not in the memory bandwidth. Look, with AA the loss of 6600GT to the "emulators" is less! If it had been up to the specifics of memory operation, it must have been vice versa. It means that NV43 has its minor peculiarities, which make it not simply 1/2 of NV40 but with some modifications (most likely by cutting down some units and upgrading the others).
With enabled anisotropy: the picture is similar to "pure speed", complete victory of NV43.
The final heaviest mode with AA and anisotropy: The same picture. There is no need in comments.
Thus, on the whole:
- GeForce 6600GT versus RADEON 9800 PRO (ATHLON64 3400+) – victory(!)
- GeForce 6600GT versus GeForce PCX5900 – victory (!)
- GeForce 6600GT versus RADEON X600XT – victory (!)
- GeForce 6600GT versus GeForce 6800LE (ATHLON64 3400+) – approximate parity with a little advantage of 6600GT
- GeForce 6600GT 325/700 MHz versus GeForce 6800 (8/3) 325/350 MHz (ATHLON64 3400+) – ambiguous picture, minor defeat of NV43
- GeForce 6600GT 350/1000 MHz versus GeForce 6800GT PCX (8/3) 350/500 MHz – the same picture
- GeForce 6600 300/600 MHz versus GeForce PCX5900 – victory!
- GeForce 6600 300/600 MHz versus GeForce PCX5750 – victory!
- GeForce 6600 300/600 MHz versus RADEON X600XT – victory!
Write a comment below. No registration needed!
|
|
|
|
|