Fillrate Tester Results
Source: Tom's Hardware – Keywords: nvidia, gtx, 280
Fillrate Tester Results
Geometry Shader
The Geometry Shading performances of the previous Nvidia Direct3D 10 GPU weren’t especially impressive, due to under-dimensioned internal buffers. Remember that according to Direct3D 10 a Geometry Shader is capable of generating up to 1,024 single-precision floating-point values per incoming vertex. Thus, with significant amplification of geometry, these buffers were quickly saturated and prevented the units from continuing calculation. With the GT200 the size of these buffers has been multiplied by a factor of six, noticeably increasing performance in certain cases, as we’ll see. To make the most of the increase in the size of these buffers, Nvidia also had to work on the scheduling of Geometry Shading threads.
On the first shader, Galaxy, the improvement was very moderate – 4%. On the other hand, it was no less than 158 with Hyperlight – evidence of the potential improvement with this type of shader, everything being dependent on their implementation and their power consumption (number of floating points generated per incoming vertex). So the GTX 280 has closed the gap and edges the 3870 X2 for this same shader.
Now let’s look at the Rightmark 3D Point Sprites test (in Vertex Shading 2.0).
Why are we talking about this test in the section on Geometry Shaders? Simply because since Direct3D 10, point sprites are handled by Geometry Shaders, which explains the doubling of performance between 9800 GTX and GTX 280!
Various Improvements
Nvidia has also optimized several aspects of its architecture. The post-transformation cache memory has been increased. The role of this cache is to avoid having to retransform the same vertex several times with indexed primitives or triangle strips by saving the result of the vertex shaders. Due to the increase in the number of ROPs, performance in Early-Z rejection has been improved. The GT200 is capable of rejecting up to 32 masked pixels per cycle before applying a pixel shader. Also, Nvidia announces that they’ve optimized communication of data and commands between the driver and the front end of the GPU.
- Previous page ROPs
- Next page Vertex/Pixel Shading Performance
- Quad SLI Vs. 3-Way SLI
- Comparative Component Charts
- Best Graphics Cards for the Money: June 08
- Radeon 3850 AGP Plus Single-Core CPU
- GPU vs. CPU Upgrade: Extensive Tests
- NForce 780a SLI Debuts Hybrid SLI
- How To Overclock Your Graphics Card
- PCI Express 2.0 Graphics Cards: How Much Extra Performance Do They...
- Nvidia GeForce 9800 GTX Review
- Nvidia GeForce 9800 GX2 Review
hmm not as big an improvement as i thought. will have to wait and see on the drivers improving the cards , but the 260 gtx seems to be the much better option given the price. still , will have to see what ati bring to the fray first. patience will be reflected in price i have no doubt.
frankly depressing, Me WANTS MRAW POWER!!!!
I am so disappointed. Now if AMD delivers on the dual GPU single memory rumour (2 GPUs on a single card but without the Crossfire problems) NVidia could have a serious problem.
Why have they tested this system with only 2Gb of RAM? If you're testing a GPU with 1Gb of VRAM, surely you'd have more installed?
They also have 2 conflicting prices on page 28.
For the 280GTX- $846 and $650;
For the 260GTX- $450 and $400
Wouldn't it have been more prudent to test against a 8800gtx ultra as this is still the single most powerfull card.
It might just be me but 66.5dBa is unbearable unless you have your PC locked away in a cupboard somewhere. This business of supplying substandard fans on very expensive cards is intolerable. Why don't they strike a deal with Zalman / Thermalright for example, and ship cards that are quiet / silent? I'm sure that people who have the money to buy a £500 GPU could afford £10 more for a better cooling solution that's included.
where is that 20W to 30W idle you are talking about? The least in the graph is 199W!
mi1ez: Probably the reason for just 2GB RAM was that it allowed Tom's to stick with 32-bit OS architecture. If they tried using more RAM they'd be stuck with 64-bit Bindows which would not be pretty - aside from really needing 8GB to give a big difference over 2GB in 32bit Vista, there's the slight issue of stable signed drivers, which these cards probably won't have for a while. Good luck trying to get Vista 64 to even "see" the cards! XD
jhoravi: that idle power would only come up on newer nVidia mobos as the card would be shut down entirely when idle and hand over to the integrated chip.
And was it me or was the Noise text copypasted over the Temperature text on the next page? Oops.
Lets try again Mr THG (uhhhm try getting your fraking website working plz)...
Now lets see this puppy in action:
http://www.evga.com/products/pdf/01G-P3-1289-AR.pdf
!!
Bob