• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

X1900XT VS HD3870

E

Eradan

[H]ard|Gawd
2FA
Joined
Apr 7, 2006
Messages
1,188
Here are a few comparison performance numbers at 1680x1050, all medium settings for Crysis. I'm running Vista Business, 2GB RAM, P965 mobo, and a Yorkfield Q9550 CPU (stock).

Here's the X1900XT (stock speeds)

11/28/2007 6:04:25 PM - Vista
Beginning Run #1 on Map-island, Demo-benchmark_gpu
DX9 1680x1050, AA=No AA, Vsync=Disabled, 32 bit test, FullScreen
Demo Loops=3, Time Of Day= 9
Global Game Quality: Medium
==============================================================
TimeDemo Play Started , (Total Frames: 2000, Recorded Time: 111.86s)
!TimeDemo Run 0 Finished.
Play Time: 84.32s, Average FPS: 23.72
Min FPS: 16.15 at frame 1957, Max FPS: 32.41 at frame 995
Average Tri/Sec: 20555382, Tri/Frame: 866596
Recorded/Played Tris ratio: 1.06
!TimeDemo Run 1 Finished.
Play Time: 78.20s, Average FPS: 25.58
Min FPS: 16.15 at frame 1957, Max FPS: 38.44 at frame 79
Average Tri/Sec: 22390054, Tri/Frame: 875448
Recorded/Played Tris ratio: 1.05
!TimeDemo Run 2 Finished.
Play Time: 78.06s, Average FPS: 25.62
Min FPS: 16.15 at frame 1957, Max FPS: 38.44 at frame 79
Average Tri/Sec: 22426450, Tri/Frame: 875268
Recorded/Played Tris ratio: 1.05
TimeDemo Play Ended, (3 Runs Performed)
==============================================================

Completed All Tests

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

11/28/2007 6:04:25 PM - Vista
Run #1- DX9 1680x1050 AA=No AA, 32 bit test, Quality: Medium ~~ Overall Average FPS: 25.6

And the freshly installed HD3870 (stock speeds):

11/28/2007 7:22:47 PM - Vista
Beginning Run #1 on Map-island, Demo-benchmark_gpu
DX9 1680x1050, AA=No AA, Vsync=Disabled, 32 bit test, FullScreen
Demo Loops=3, Time Of Day= 9
Global Game Quality: Medium
==============================================================
TimeDemo Play Started , (Total Frames: 2000, Recorded Time: 111.86s)
!TimeDemo Run 0 Finished.
Play Time: 53.19s, Average FPS: 37.60
Min FPS: 27.39 at frame 276, Max FPS: 55.25 at frame 894
Average Tri/Sec: 32591840, Tri/Frame: 866784
Recorded/Played Tris ratio: 1.06
!TimeDemo Run 1 Finished.
Play Time: 46.12s, Average FPS: 43.37
Min FPS: 27.39 at frame 276, Max FPS: 58.48 at frame 61
Average Tri/Sec: 37935368, Tri/Frame: 874772
Recorded/Played Tris ratio: 1.05
!TimeDemo Run 2 Finished.
Play Time: 45.50s, Average FPS: 43.96
Min FPS: 27.39 at frame 276, Max FPS: 58.48 at frame 61
Average Tri/Sec: 38473416, Tri/Frame: 875226
Recorded/Played Tris ratio: 1.05
TimeDemo Play Ended, (3 Runs Performed)
==============================================================

Completed All Tests

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

11/28/2007 7:22:47 PM - Vista
Run #1- DX9 1680x1050 AA=No AA, 32 bit test, Quality: Medium ~~ Overall Average FPS: 43.665

Hope this is helpful to anyone on the fence about a potential upgrade. I think I now enough overhead with the new card to increase the quality settings a bit in Crysis. I'm going to do some testing later with AA enabled.
 
impressive jump. Now i (somewhat) know how my 1950xt fares against the newest revision
 
may you please try AA with both cards.... i want to see how AA fares with the 3870..

THANKS PLEASE AND THANK YOU :)
 
Thanks for this!

Iam gaming on 1650x1050 too and OP system is quite near to mine (except CPU). I was mainly going for 3870 for lower power consumption and more silent stock cooler, but its nice to see there will be a significant speed jump also :)
 
Wow nice. I've been on the fence about upgrading my X1900XT. Looks like it's time to spend some money.
 
I wish someone would buy my cards...been wanting to sell them so I can upgrade, but doubt it'll evr happen!
 
Hm, now I'm a bit intrigued by the HD 3870 as an upgrade to my 7950GT instead of just thinking 8800GT. :)
 
At work now but I'll get some more info online this evening, including Crysis with AA enabled.


E
 
Here we go, same setup as before but with 2X AA enabled (which sets shader quality to High):

12/1/2007 10:00:59 AM - Vista
Beginning Run #1 on Map-island, Demo-benchmark_gpu
DX9 1680x1050, AA=2x, Vsync=Disabled, 32 bit test, FullScreen
Demo Loops=3, Time Of Day= 9
Global Game Quality: Custom
Custom Quality Values:
VolumetricEffects=Medium
Texture=Medium
ObjectDetail=Medium
Sound=Medium
Shadows=Medium
Water=Medium
Physics=Medium
Particles=Medium
Shading=High
PostProcessing=Medium
GameEffects=Medium
==============================================================
TimeDemo Play Started , (Total Frames: 2000, Recorded Time: 111.86s)
!TimeDemo Run 0 Finished.
Play Time: 66.77s, Average FPS: 29.95
Min FPS: 19.59 at frame 138, Max FPS: 42.43 at frame 929
Average Tri/Sec: 25985886, Tri/Frame: 867531
Recorded/Played Tris ratio: 1.06
!TimeDemo Run 1 Finished.
Play Time: 59.02s, Average FPS: 33.88
Min FPS: 19.59 at frame 138, Max FPS: 42.58 at frame 976
Average Tri/Sec: 29708000, Tri/Frame: 876750
Recorded/Played Tris ratio: 1.05
!TimeDemo Run 2 Finished.
Play Time: 58.60s, Average FPS: 34.13
Min FPS: 19.59 at frame 138, Max FPS: 42.58 at frame 976
Average Tri/Sec: 29923816, Tri/Frame: 876717
Recorded/Played Tris ratio: 1.05
TimeDemo Play Ended, (3 Runs Performed)
==============================================================

Completed All Tests

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

12/1/2007 10:00:59 AM - Vista
Run #1- DX9 1680x1050 AA=2x, 32 bit test, Quality: Custom ~~ Overall Average FPS: 34.005
 
Just noticed the result say NO AA. Going to run it again to make sure I didn't screw something up.


E
 
Accurate now. I'm need to practice my cutting and pasting. Duh...


E
 
3DMark 06:

X1900XT: 6519
HD3870: 11231

I'm actually going to have time to game this weekend and will provide some more feedback on my experiences. Any game requests, please let me know.


E
 
zone_86 said:
Hm, now I'm a bit intrigued by the HD 3870 as an upgrade to my 7950GT instead of just thinking 8800GT. :)
Click to expand...

I just saw a review where they found that Nvidia is cheating on Crysis performance by "optimizing" (turning off features) Crysis to make it run faster.

They renamed the Crysis EXE and got a big performance drop putting the 3870 on top, I think, I have to read it again to be sure exactly...
 
gibber said:
I just saw a review where they found that Nvidia is cheating on Crysis performance by "optimizing" (turning off features) Crysis to make it run faster.

They renamed the Crysis EXE and got a big performance drop putting the 3870 on top, I think, I have to read it again to be sure exactly...
Click to expand...

I think we need a new thread about that, and the [H] needs to do their own test to expose Nvidia's cheating "optimized" driver, and exactly what effect it has on performance.

Back on the thread topic, I see that Circuit City has the X1950Pro for $92 after rebate and I have a $15 off $100 coupon to use too, I might just get that so I can hold out until whatever the next gen of cards that comes out in 6 months from now is and get one of those, which, hopefully will be finally able to play Crysis maxed out :D

Either that, or hopefully the 3870 will be down to $150 by then.
 
Yeah I'm waiting on a few things. 8800GT's to get the new bioses with better fan control (shipped that way so won't have to flash) and better HSF's to begin with on the 8800GT.

The thing that really bothers me about NV cards are the drivers. So many beta drivers and so many of them cause issues you really have to test too many of them to find a good one in any config. I'm still using 158.27's because they simply run better than any other driver that I have tried for the 7950GT But in the mean time I will also closely watch the 3870 for a few months and make a decision then, but if I can get a 3870 for right @ $200.00 USD I may go for it for the better HSF's and no constant driver fiacso (like NV god it's a mess).
 
You must log in or register to reply here.
Back
Top