There are
3DMark 2001SE and
3DMark 2005 by MadOnion /
Futuremark to serve the purpose of synthetical benchmarking. 3DMark 2001SE build
330 is almost 5 years old but still has some interesting tests to offer, first
of all, those synthetical completely Fill Rate and High Polygon Count, though
Nature is a good one as well. 3DMark 2005 is able to use pixel and vertex
shaders v3.0, and they happen to be a must have for modern 3D gaming titles. One
must note that shaders of 3DMark 2005 are written in HLSL and compiled at
run-time rather than written in assembly language and pre-compiled, so they may
lack optimisation. Unfortunately, many 3D software developers follow this
popular trend. Both 3DMark 2001SE and 3DMark 2005 operate through Direct3D.
This test illustrates how fast a scene can be filled, i. e. texturing
and rasterising speed. It constains almost no geometry data, so there is a very
little to zero dependence on processor speed or video memory size.
As it has been expected, BFG GeForce 7800GS OC is far ahead with its 256-bit
memory data bus. After the overclocking it has advanced even more: +19% in
single-texturing and +25% in multitexturing. Compared to the non-overclocked
leader, ASUS GeForce 6600GT Top is about 2 times slower in single-texturing
though not that much in multitexturing. Sapphire Radeon X1600 Pro has performed
poorly, it's exactly 3 times slower than BFG GeForce 7800GS OC and about 2 times
slower than ASUS GeForce 6600GT Top. What to say, a slow video memory and 4
texture pipelines have made it a reality.
This test renders a scene with several objects moving dynamically and either
1 or 8 active lights. There is almost no texturing (background only), so it
doesn't depend much on video memory bandwidth. It's a matter of vertex pipelines
mostly to deliver high results in this test.
BFG GeForce 7800GS OC with its 6 vertex pipelines is still a leader, but the
others get closer. Sapphire Radeon X1600 Pro has managed to prevail over ASUS
GeForce 6600GT Top this time (5 vs. 3 vertex pipelines) and to camp not so far
away from BFG GeForce 7800GS OC in 8 lights mode.
Well, Nature is considered fairly as the most beautiful game test of the
3DMark 2001SE suite. It employs vertex shaders v1.0 to produce a wind effect on
grass and trees' leaves, also to calculate movements of the fisherman and
butterflies. Pixel shaders v1.0 together with multitexturing are involved in
generation of the water surface.
BFG GeForce 7800GS OC has left both competitors far behind even without
overclocking. In fact, after the overclocking it appeared over than 2 times
faster than ASUS GeForce 6600GT Top no matter what mode to look at, and that's
an impressive result. At the same time, Sapphire Radeon X1600 Pro has lost all
bets to ASUS GeForce 6600GT Top once again. It seems that low texturing speed
and moderate 1-light vertex performance have played a bad joke with Sapphire
Radeon X1600 Pro, especially if to take into account that Nature isn't abundant
in pixel shaders unlike many modern tests, so texture pipelines of RV530 appear
overloaded while pixel pipelines feel underloaded.
The other game tests of the 3DMark 2001SE suite (Car Chase, Dragothic and
Lobby) don't use pixel shaders at all, so they're of a little value nowadays
from the point of game design. However, they depend well on texturing and
rasterising speed. Yes, somewhere back in the past it has been of key importance
for 3D graphics hardware. Anyway, refer to the following tables.
3DMark 2001SE (1280x1024x32bpp), Low detail, average FPS |
|
Car Chase |
Dragothic |
Lobby |
BF NoAA |
16xAF NoAA |
16xAF 4xMSAA |
BF NoAA |
16xAF NoAA |
16xAF 4xMSAA |
BF NoAA |
16xAF NoAA |
16xAF 4xMSAA |
BFG 7800GS OC o/c |
334 |
325 |
277 |
482 |
437 |
292 |
311 |
308 |
268 |
BFG 7800GS OC |
325 |
307 |
246 |
425 |
380 |
256 |
309 |
303 |
246 |
ASUS 6600GT Top |
261 |
229 |
146 |
308 |
283 |
159 |
280 |
234 |
141 |
Sapphire X1600 Pro |
245 |
149 |
106 |
216 |
176 |
123 |
210 |
128 |
94 |
3DMark 2001SE (1280x1024x32bpp), High detail, average FPS |
|
Car Chase |
Dragothic |
Lobby |
BF NoAA |
16xAF NoAA |
16xAF 4xMSAA |
BF NoAA |
16xAF NoAA |
16xAF 4xMSAA |
BF NoAA |
16xAF NoAA |
16xAF 4xMSAA |
BFG 7800GS OC o/c |
109 |
109 |
109 |
244 |
230 |
190 |
145 |
145 |
140 |
BFG 7800GS OC |
108 |
108 |
107 |
221 |
203 |
166 |
145 |
144 |
135 |
ASUS 6600GT Top |
106 |
104 |
85 |
180 |
165 |
115 |
139 |
129 |
91 |
Sapphire X1600 Pro |
106 |
82 |
62 |
146 |
120 |
91 |
118 |
80 |
64 |
There are two modes of detalisation, High and Low. While High detail mode
helps 3DMark 2001SE to prove its processor-dependent reputation, Low detail one
delivers interesting results. When it comes to anisotropic filtering, BFG
GeForce 7800GS OC and ASUS GeForce 6600GT Top have lost 11% and 12% on maximum
respectively while Sapphire Radeon X1600 Pro — 39%. Remember, 16x
anisotropic filtering (AF) requires 128 samples per texel to be processed while
bilinear (BF) and trilinear (TF) take only 4 and 8 respectively. Of course, it
makes 16xAF dependent very much on video memory bandwidth. Finally, RV530
features only 4 texture pipelines, so there is absolutely no reason why Sapphire
Radeon X1600 Pro having not so fast memory should deliver much performance with
16xAF enabled. The situation gets different with 4x multisampled
anti-aliasing (MSAA) activated in addition to 16xAF: BFG GeForce 7800GS OC
and ASUS GeForce 6600GT Top have dropped their performance by 33% and 44% on
maximum respectively while Sapphire Radeon X1600 Pro — only by 24%.
Anti-aliasing is handled by raster operators, so there is no wonder: BFG GeForce
7800GS OC and ASUS GeForce 6600GT Top accommodate half as many operators as
texture pipelines while Sapphire Radeon X1600 Pro sticks to 1:1 ratio.
Having first conclusions made, let's continue with 3DMark 2005.
Return to Proxycon features advanced geometry (the hangar with many
high-detailed objects including those moving dynamically) and advanced lighting
(up to 8 sources). Pixel shaders are also used widely. The view is limited by
walls of the hangar and corridor which is typical for most 3D action games.
BFG GeForce 7800GS OC holds the top while Sapphire Radeon X1600 Pro gains
the second place. ASUS GeForce 6600GT Top hasn't performed well especially with
4xMSAA enabled.
Firefly Forest depends on vertex performance mostly because of trees and
grass: branches of the first move with the wind, and the second just needs to be
generated dynamically. In a matter of fact, there are only 2 lighting sources:
the moon which is a statical one and the flying light which is a dynamical one.
The situation remains the same mostly, though Sapphire Radeon X1600 Pro has
helped ASUS GeForce 6600GT Top to reduce the gap by losing a lot of performance
on 16xAF.
Canyon Flight employs advanced very much pixel shaders to generate walls of
the canyon and the water surface. There are different fog effects to take place
as well. Geometry of objects is moderate, the sun serves as the only lighting
source.
BFG GeForce 7800GS OC is still the leader. ASUS GeForce 6600GT Top and
Sapphire Radeon X1600 Pro share the second place except of 16xAF +
4xMSAA modes where Sapphire Radeon X1600 Pro has prevailed definitely.
Overall, BFG GeForce 7800GS OC has won this competition. Considering
1024x768 with trilinear filtering and no anti-aliasing enabled, it has scored
about 5800 marks before the overclocking and about 6900 after. Sapphire Radeon
X1600 Pro has managed to possess the second place with about 4100 marks, and
ASUS GeForce 6600GT Top has got about 3700 ones. With bilinear filtering which
is set by default ("Optimal") the scores would be approx. 2% higher. BFG GeForce
7800GS OC hasn't shown any weak points, but Sapphire Radeon X1600 Pro loses a
lot of performance while performing 16x anisotropic filtering, and ASUS GeForce
6600GT Top isn't a horse to bet on when 4x multisampled anti-aliasing is
enabled.