GPU Shoot-Out – Part II – Setting New Benches

Conclusion

The “Feel” of the Video cards:

Again, this gets very subjective.  As mentioned previously in Part I, the Nvidia control panel seems to be stuck in a time warp while ATi’s has appeared to progress.  “Centered timings” still could not be enabled easily in the Nvidia CP, and a convoluted workaround was necessary.  Both sets of drivers are very mature, except for ATi’s CrossfireX-3 drivers.  This reviewer gives a slight edge to Nvidia’s IQ over ATi’s although you have to be really nit-picky to notice.  The cards themselves can easily be ranked in terms of performance, 1-3 – top-to-bottom:

1. 4870×2

2. 280GTX

3. HD4870

CF-x3 is a little faster than x2 and becoming less of a curiosity than with previous drivers.  We expect the situation will improve as we are testing with Catalyst 8-12 now for future articles in our series.

The 4870×2 is very smooth and minimizes any microstutter that is usually apparent in multi-GPU solutions.  It is ideal for 19×12 resolution in this reviewer’s opinion.  If you must have a single-GPU solution, the GTX280 stands a notch above the HD4870.  Even where the frame rates were very close, the Nvidia GPU felt more “solid” than 4870-512MB, and it would be a good choice for 19×12.  4870-512MB just falls short and would be excellent for playing the newest and most demanding PC games at 16×10.  In the next parts of our review, we will compare the 1GB versions of HD4870 to the 512MB versions and explore possible 512MB VRAM limitations using 4870-512MB paired with 4870X2-2GB in crossfireX-3.

We will be back very soon with Part III of our testing – still using Catalyst 8-9  and Geforce 178.13 – for an exact P35 vs. X48 motherboard comparison.  That will let you know you if our new X48 motherboard’s PCIe 2.0 specification makes any difference performance-wise over our [now current] PCIe 1.0 spec’d P-35 motherboard.  We will also be exploring X48 motherboard’s crossfireX-3  performance – [then] unlimited with 16x +16x full bandwidth –  over our [current P35 motherboard’s] 16x + 4x crossfire solution.  Thus, we will be able to compare 4870-512MB with 4870-1GB and different crossfireX-3 combinations with 4870×2-2GB to see if the 512MB it is now forced to share with a 4870-512MB card, limits performance over “true” crossfireX-3 with 4870X2 and 4870-1GB in the second slot.  Of course, we will also watch Nvidia’s drivers evolve over the same period to see if they managed to get more performance out of their Tesla architecture.  We expect to also explore Nvidia’s propriatory PhysX with their second Big Bang Drivers (180.48) vs. Catalyst 8-12 and ATi’s now-enabled “Stream” drivers which is meant to take on Nvidia’s CUDA.

Here is our planned “GPU-Shootout” series:

Part I – Cat 8.8 (8/20/08) vs Geforce 177.41 (06/26/08) (p35 MB platform)[done]

Part II – Cat 8.9 (9/17/08) vs Geforce 178.13 (09/25/08) (p35 MB platform)[done]

Part III – Cat 8.9 (9/17/08) vs Geforce 178.13(09/25/08) (x48 MB platform from now on)[benches completed]

Part IV – Cat 8.10 (10/11/08) vs Geforce 178.24 (10/15/08)  [benches completed]

Part V – Cat 8.11 (11/13/08) vs Geforce 180.48 (11/19/08) [benches completed]

Part VI – Cat 8.12 (12/12/08) vs. Geforce 180.48 (11/19/08)  (Stream vs BigBangII; benching in progress)

We expect to also have Intel’s new 65w,  low-power QuadCore Q9550 in for testing shortly and we will also give you stock and overclocked comparisons of it vs. our Core2Duo e8600 which we can now get to well over 4.0Ghz.

–After that, we expect to compare the maturing Intel I-7 CPU platform with our current maxed out Penryn system … and we expect to also explore Nvidia GTX280/290 SLi on an X58 motherboard.

Stay tuned.  All five parts of our planned testing has been completed – with Part VI testing in progress – and we think we will have some very interesting articles for you to read very shortly as you plan your own coming upgrades.

Mark Poppin

ABT Editor