3dmark 05

[Cantacuzene]

Its out and its amazing. Even with a 6800GT I was only pulling like 15fps max. Its visually stunning. I'm eagerly anticipating the day I see it running at 60fps.

My score was 3047, but I had other programs open and hadnt rebooted. I'll do a proper benchmark later.

[Bires]

Wow...that was fast...screenies were released just a week ago.


DLing now...


It's a biggie-280MB!


I'm eagerly anticipating the day I see it running at 60fps.

By then, 3DMark06 will be the standard.

Nothing like an artificial benchmark to keep the parade of hardware upgrades coming! :kaching:

[Cantacuzene]

http://www.lemonizer.com/upload/uploadsSeptember/benchie.jpg

Thats my school after a reboot and an overclock to 6800 Ultra speeds. All I need is an NV Silencer and I could break 5,000... but that would void my lifetime warranty. And no benchmark is worth voiding a lifetime warranty. I'll be satisfied with this score for now.

[Bires]

VERY VERY CPU intensive. My 2.4C just crawled during the CPU tests....ack

5-20 FPS in the games and 1-3 in the CPU tests!
http://www.lemonizer.com/upload/uploadsSeptember/3dm05.jpg
http://service.futuremark.com/compare?3dm05=22586

[CynJon]

My system (AMD64 3000+, 1GB PC3200, 9700Pro, all watercooled) got 2049 3DMarks at stock video card settings. When I squeezed the video card a bit ;) (421MHz core/376 MHz memory), my score climbed to 2558. It's a pretty cool slideshow, though!


<starts looking for video card upgrade> DAMN YOU FUTUREMARK!

[PrObLy]

Doesn't look like I'll be able to run it as my "ancient" :rolleyes: GF4 Ti-4600 doesn't support pixel shaders 2.0 :neartears

[zippyjuan]

This will bring your system to its Knees! I think mine is still pretty decent, but only got a score of 1952. The frame rates in the first part only averaged 10.0 on the last segment (serpent) and I was a mere 1.0 to 2.5 frames per second on the CPU test. Does this version include any audio, pixel shader, fill rate, or vertex shader tests like the last one? I did not see any performed (maybe because my system could not keep up). ( Edit: apparently these are only available with the pro version)

I have an AMD Athlon XP 2400+ with 512 mb Corsair XMS memory (but a 2x 512 pack for a gig total on the way) and a Radeon 9700 non pro 128mb video card. Windows XP Home with SP2 and Directx 9C installed.

If you want an excuse to upgrade- try running this. It will convince you that you have to spend a couple thousand more on your computer. Over the past year, just with video driver updates, my 3D Mark 03 score has gone up over 100 points. There will probably be driver updates aimed at this one as well.

As far as the size of the download (which took me like 45 minutes to an hour to download) it seems like you get the full version- you just cannot access parts of it unless you register and get a code to activate it.

[Cantacuzene]

This is not cpu intensive at all. Both me and my roommate have ge-gorce 6800GT. I have an AMD 2500+ and he has an AMD64 FX51, which is significantly faster, yet his 3dmark score was only 30 points higher, less than 5%, despite havig a cpu thats nearly 50% faster.

[zippyjuan]

There is already a look at 3D Mark05 which includes a look at how various video cards performed on the test here: http://www.xbitlabs.com/articles/video/display/3dmark05.html

I think in my case it is the fact that my video card is only 128mb. I do have the latest drivers from ATI installed.

[Bires]

This is not cpu intensive at all. Both me and my roommate have ge-gorce 6800GT. I have an AMD 2500+ and he has an AMD64 FX51, which is significantly faster, yet his 3dmark score was only 30 points higher, less than 5%, despite havig a cpu thats nearly 50% faster.

What were your CPU test scores? My 2.4C scored only 1.6 and 2.7. The accelerated scores were almost 10x that. That seems like a cpu bottleneck to me.

[Cantacuzene]

The cpu score doesnt matter in the accelerated tests because they are pushing two different graphics. When identical video cards get the same score but each has a radically different processor that means the bottleneck is the video card and not the cpu.

[Bires]

The cpu score doesnt matter in the accelerated tests because they are pushing two different graphics.

That's why there is a nonaccelerated CPU test.



When identical video cards get the same score but each has a radically different processor that means the bottleneck is the video card and not the cpu.


Right.

And that's why I'm curious what your actual scores are. You didn't post your session details. I want to know where the point is that the CPU stops being the the bottleneck. It's hard to believe that the cpu alone couldn't push the polygons in the serpent test beyond 1.6fps

[bachviet]

My stock score is 2353 with a R9800Pro and A64 3200+ (crappy RAM though).

My slightly modded score (GPU to 416/351) is listed in my sig.

[Showtime]

This test was supposed to be very gpu intesive and my 9700np showed it with a 1250. I need to clean up all the crap running on my system or do a fressh install soon. Maybe I should even buy a new card so I can get a nice score. ;)

-j