shadermark 2.1 released

[DSN2K]

Ive updated mine with PS2.0a results and listed drivers, performance is worse with PS3.0

 

[hovz]

lol.....the reason the kid on rage 3ds xtpe is faster than your modded vivo is because the idiot is benching with performance and balanced driver settings and claiming wins for ati...

 

[Skinner]

Sorry to ask, but how does I run this prog

 

[Broken Hope]

Sorry to ask, but how does I run this prog

Start it up, hit run then when it loads press P to start the benchmark.

 

[Skinner]

Well, the shadermark2.1 exe doesn't run for me, it don't respond, maybe something went wrong during the install.

 

[hovz]

Well, the shadermark2.1 exe doesn't run for me, it don't respond, maybe something went wrong during the install.

do u have 9.0c?

 

[Skinner]

Yes, and I have done many SM3.0 and SM2b tests with both of my cards in Far Cry and never had any problems with both paths.

I'll down the test again, maybe something is corrupt..

 

[Apoc]

video mode / device info
(1024x768) X8R8G8B8 (D24X8) vram used 110100480
HAL (pure hw vp): RADEON 9500 (Omega 2.5.67) (Anti-Detect-Mode: off, gamma correction: DAC)

pixel shader version: Pixel Shader 2_0

2 209 fps
3 147 fps
4 146 fps
5 120 fps
6 146 fps
7 135 fps
8 94 fps
9 342 fps
10 228 fps
11 146 fps
12 53 fps
13 86 fps
14 108 fps
15 57 fps
16 77 fps
17 94 fps
18 8 fps
19 33 fps
20 Not Supported
21 Not Supported
22 43 fps
23 Not Supported
24 Not Supported
25 24 fps
26 24 fps

 

[jvd]

i can't get it to render past 85fps for some reason .

 

[Apoc]

i can't get it to render past 85fps for some reason .

It's a stupid suggestion but... have you checked v-sync or temporal AA?

 

[Dave Baumann]

    X800 XT PE     6800 Ultra            
    PS 2.0 PS2.0b  PS2.0   PS3.0 PS3.0 FP16 PS3.0 FP32
2   1001   1000     890     888    889       720
3    732    732     720     720    720       540
4    732    732     660     720    719       533
5    606    606     585     585    585       360
6    732    732     650     650    649       509
7    675    675     600     600    600       480
8    373    373     360     380    413       300
9   1268   1267    1006    1008    958       997
10   817    817     800     801    802       806
11   698    698     650     650    650       510
12   250    250     450     450    450       330
13   333    333     360     285    270       240
14   402    402     420     419    419       360
15   288    288     252     290    300       255
16   346    346     300     300    300       255
17   444    444     390     390    380       300
18    40     40      50      50     50        37
19   139    139     148     160    163       136
20    46     46      48      48     51        48
21                   79      79     78        79
22   205    205     110     112    115       105
23                  120     126    129       116
24   143    143      75      75    77         72
25   118    118      84      77    90         77
26   116    116      80      73    87         75

Both boards early reference samples, same system as the AGP AMD system used here. WHQL Cat 4.9 w CCC/.NET 1.1 SP1 for ATI, Beta 65.76 for NVIDIA.

 

[jvd]

i can't get it to render past 85fps for some reason .

It's a stupid suggestion but... have you checked v-sync or temporal AA?

yes in my drivers thats all off

Dunno why .

 

[ellingsen1]

Here are my results.

 XFX 6800 Ultra Stock  425/1.15  66.31 Drivers
  PS2.0 PS3.0   
2   962    962
3   781    781
4   781    781
5   712    709
6   721    721
7   672    672
8   409    385
9  1077  1076
10  841   841
11  721   721
12  481   481
13  418   401
14  481   480
15  277   337
16  330   330
17  450   450 
18   55    57
19  165   179  
20   55    55  SM3.0 Both
21   88    88  SM3.0 Both
22  119   120
23  132   134
24   80    80  SM3.0 Both
25   94    95
26   90    90

[/code]

 

[jvd]

so dave according to your numbers the 6800ultra is slower than the x800xt pe ? Also the x800xt pe does not loose or gain speed going to p.s 2.0b .

 

[kemosabe]

Forgive my newbieness to this particular topic, but how are we to expect any significant performance gains on the NV4X series in games that support SM3.0 in light of these results? :?

 

[ERP]

Forgive my newbieness to this particular topic, but how are we to expect any significant performance gains on the NV4X series in games that support SM3.0 in light of these results? :?

Test 21 is the same as test 20 but uses dynamic flow control to basically double the frame rate.

ps3.0 isn't a panacea, but there are cases where it cn be a win.

 

[trinibwoy]

Both boards early reference samples, same system as the AGP AMD system used here. WHQL Cat 4.9 w CCC/.NET 1.1 SP1 for ATI, Beta 65.76 for NVIDIA.

Dave, I'm sure you've noticed but people are getting much higher nvidia scores with the 66.31 driver that is recommended for use on 2.1. Why didn't you use that one?

 

[Lezmaka]

so dave according to your numbers the 6800ultra is slower than the x800xt pe ? Also the x800xt pe does not loose or gain speed going to p.s 2.0b .

According to Dave's results, the XT PE is faster in a little more than 2/3 of the tests, and the Ultra is faster in the other 1/3. It's be interesting to see if the 6800's gain much speed with the 66.31 recommended drivers on dave's system, but I think he's got something pretty big coming up this week and may not have the time

 

[jvd]

we'll see soon i guess

 

[Xmas]

Dave, if you have the time, please do a run with ps_2_a on the 6800 as well.