would be interesting to see what X1900XTX doesQuote:
Originally Posted by perkam
Printable View
would be interesting to see what X1900XTX doesQuote:
Originally Posted by perkam
UT has always been amd biased. Bah it sucks how the x1300p out performs the 6600gt im getting that does 11k in 03 :-/ Oh well, I dont really card, all I play is ut and css. When 07 comes out Ill go 1900Quote:
Originally Posted by RAMMAN
Probably more accurate of late 07 Unreal 3 engine based games at high detail.
So is it released yet???
interesting
Quote:
Originally Posted by tyellbee
Apparently 12pm est :(
Hm.. 6800U's whipping X800XT's because of SM3 support
One of the screenshots looks very nice
i reckon thats going to change with ut2007.plus its sm3 and sm2 only so sm2.0b hardware is going to take a plunge performance wise just like what we are seeing in 3dmark06.Quote:
Originally Posted by sabrewolf732
hah, futuremark coders making code as efficient or representative of code done by epic? Hah :slap:Quote:
Originally Posted by afireinside
Why would it change? Because futuremark said so? I don't see how this 1 "benchmark" is making people say intel p4's are the new good gaming cpu's :rolleyes:Quote:
Originally Posted by RAMMAN
I don't know what the problem is. I mean it seems fair to me. In that comparison that Perkam just posted the Intel platform was OCed, but the AMD one was not, so it's only fair that the Intel one has a better score.
As for the CPU, I would expect dual core to get 2x the performance because graphics is very parallisable.
Also, from EVA's chart it looks like dual core gets you a 75% boost, which is perfectly reasonable.
no because intels involved.hence why intel cpus will mysteriously perform better in multithreaded apps starting with quake4,s latest patch :slap: .Quote:
Originally Posted by sabrewolf732
6800 GT + Intel Pentium 4 3005 MHz
3DMark Score 2178 3DMarks
SM 2.0 Score 960 Marks
SM 3.0 Score 893 Marks
CPU Score 651 Marks
6800 GS + AMD Athlon(tm) 64 1800 MHz
3DMark Score 1933 3DMarks
SM 2.0 Score 807 Marks
SM 3.0 Score 766 Marks
CPU Score 705 Marks
Upon further review it looks like amd is better in 06
Um, the quake patch is multithreaded, thus AMD X2's also got boosts :rolleyes: :stick:Quote:
Originally Posted by RAMMAN
yes but intels boosts were bigger than amd,s.something to do with sse instructions...anyway enough of this ot discussion.
Link?Quote:
Originally Posted by RAMMAN
Also, is it just more or does 06 not seem that graphically impressive? Futuremark can suck a nut.
One thing I think is ridiculous, the best CPU available can't even get 1 FPS in the CPU test. I mean what is this supposed to be indicating? Future performance for 2006? Try future performance for 10 years from now. This isn't telling you anything about current hardware performance. It would be one thing if this had a functional use, but now I think people are really starting to question the purpose 3DMark, especially in light of the fact that it was released so suddenly. Now it just looks like its whole purpose to is to keep hardware looking crappy.
There better be an insane amount of AI and physics (rivaling that of the PPU tech demos) for it to be performing that poorly. It's obvious the whole reason this released so fast is they didn't bother optimizing it, or making it widely compatible. Makes me wonder if holes are gonna be found later on, like 3DMark05's CPU-limit, because of the lack of testing and such.
The lighting is much better, but it looks like they didn't even touch the poly count! Maybe 10% more. Also, where's that big city thing?! That's what I want to see.
qft. 06 is very dissapointing imo. This is no where near indicative of game performance. And it for sure wont be indicative of ut07Quote:
Originally Posted by Cybercat
After seeing the scores, this has to just be a ploy to sell more hardware...Quote:
Originally Posted by Cybercat
We all run resolutions over 1200x1024, and none of us will even buy games that cant even get ~30fps minimum at that low of a resolution, let alone the ~14 fps our rigs will do, if this is really a good tool.
I feel sorry for you dual core Fx-60 guys. Only dual dual, or quad dual, or a serious SMP system is going to do well in this.
This is blasphemy, let's boycott futuremark until they can release a DECENT benchmark :toast:
anyway quake4,s bias is barely noticeable compared to what will be seeing in ut2007 and what were seeing in 3dmark06.Quote:
Originally Posted by sabrewolf732
for obvious reasons ut2007 will perform much, much better than 3dmark06 but it will be almost as biased towards nvidia/intel.
and yes the only real difference visually between 3dmark05 and 06 is because of the res and hdr :stick: .
It takes only 600$ of gfx cards to even get over 25fps in any of the tests.
1) 25fps @ 1200x1024 is too low to game on.
2) Half-Life 2 uses HDR and has playable framerates.
3) Every SM2.0 benchmark or game also gets better framerates...by about a factor of 2.
http://www.pcper.com/images/reviews/...ark06-1280.gif
from PCPer using FX-55 , 2 x 512 MB Corsair 3200XL @ 2-2-2-5
OK, I just looked at screenshots. In Return to Proxycon, they added more stationary light sources, and bloom (for brighter impact). In the Firefly Forest, all they did is add another pixie! Canyon Flight, added HDR, gave the captain a spanish face, added more junk in the background. It's like one person said in the description of the one of the tests: Yay for 3DMark05 + HDR!
But wait, what's this? Two new tests! At least we got that!
Definitely looks half-assed. No wonder they didn't call it 3DMark07. I'm surprised they're still charging to register this thing.