u change the driver settings for each game, other than the OGL forcing triple and Z buffering i dont change anything on the performance settings, and i assume that most do the same. your speaking of games this thread is about the driver control panelQuote:
Originally Posted by PaganII
what's more interesting is why you're having so much trouble putting together a coherent sentence, yet you're posting this on a >9000$ machine.Quote:
Originally Posted by E30M3
can not we just agree it is wrong to cheat
and when they do we Protests
this time it was amd next time it's maybe nVidia doing it
they just should not start lowering quality in a race about who is fastest
it only affects us and gaming in a bad way.
Hmm you can try to write in a language,Quote:
Originally Posted by biohead
which is not your native language.
it is not so easy
not for me at least
:shrug:
you can keep telling this over and over again yet i enjoy a higher IQ on my 6850 than what was possible on the 5850 which was proven over and over againQuote:
Originally Posted by E30M3
it's also proven that the difference between HQ (better than 5xxx series and my old 8800) and normal quality (worse quality in older games, same quality in new games) is around 5% which still puts the 68xx series in front of the gtx460 in terms of performance and price/performance
i don't see how amd lost this round, they own a 3-4 times higher market share than nVidia in the DX11 markt, they dethroned NV in overall market share and their higher priced products make up a significantly higher proportion of their revenue than nvidias...
sure nvidia is still pretty competitive right now but unless you want to buy a card above 400€ (GTX580) amd is the way to go due to output option, video playback quality, performance, price and power consumption (that is from 50€ to 300€) if you don't have any brand preferations at all...
this might change in the next round, and certainly was the other way round in the 8800 times but to claim that amd lost this round is ignorance (just as ignorant as some AMD fanboys continue to claim that AMD didn't loose against i7, they hold on to certain price points and certain workloads but overall they lost just like nvidia...)
In before nvidia people cite ~70% marketshare leftover from 79xx and 88xx series.
HAHAHA! Remember good old GeForce FX?Quote:
Originally Posted by E30M3
I too would like an answer to this question.Quote:
Originally Posted by Frontl1ne
I have seen the minor flickering mentioned in the articles within Just Cause 2, CoD: MW2 and Dragon Age: Origins with the 10.10 drivers. Videos and pics to come after I get my main system up and running again.Quote:
Originally Posted by Eastcoasthandle
And this forum proves once again that an objective opinion is hard to come by.
ive been saying it for a while... direct x should include a default render mode that has to pass a series of tests, so it does NOT get tweaked/optimized...
so people can chose to use this mode or the optimized settings ati and nvidia offer that boost performance by SLIGHTLY reducing image quality...
if im on a mainstream gpu ill appreciate the latter, but if im playing an older game i want max image quality and no optimizations at all...
and how can you sell 500$+ videocards to people that dont render the games the way they were supposed to look but blur things to boost the performance by a few percentage points?
thats just plain stupid...
btw, i dont get those videos... whats the right side of the video?
the flickering on the left side of the 6000 videos is def worse than in the 5000 videos.
BUT, on the nvidia videos is a slight flicker both left and right... the left side looks better than the 6000 but not really better than the 5000.
Quote:
Originally Posted by saaya
Left is a GPU rendering the scene. At the right is the software renderer which my its very nature doesn't have ANY optimizations. ;)
This is useless.Quote:
Originally Posted by saaya
The drivers from AMD and NVIDIA can detect the "default quality test" and use standard algorithm.
Not in min FPS. In most reviews i seen 5970 drops way below 580. And you know that's whats makes or brakes a game. I don't care if i cant get 10 fps more on high. If FPS fluctuate like crazy then game is much more unplayable. When i play any first person shooter i want stable FPS and definitely dont want them to drop into single digits. I wont post any slides just check Anand's review. And btwy most crossfire setup suffer from that. You ether agree with me that min FPS is more important then high for good smooth gaming or you a fanboy. That's why I'm not so found of dual gpu cards, and would never get card like 5970 for gaming. Would definitely pick up 5870 instead.Quote:
Originally Posted by Dimitriman
5970 has higher minimum (and avg) than 580 in AvP (dx11), BFBC2 (dx11), Just cause 2 (dx10).
580 wins in Dirt2 (dx11), Lost planet (dx11) and SC2 (dx9)
in Metro (dx11) theire about even in minimum fps....
This is from Hardware Canucks (1 of few review site i trust)
So it depends on wich games you play really.
Edit: Also looked at Anands 580 review. They only show minimum framrates in 2 games, 1 wich is is crysis warhead wich looks like it need more than 1gb ram (normal 5970 has only 1 gb ram) for thoose settings at 2560x1600, so ofc the 580 will do much better there since it got 1,5gb ram (5970 was better than 580 @ 1920x1200)
Edit2: Lab501 wich is also an excellent site wich has avg and minimum framerates.
Theire 580 vs 5970:
5970 wins in SC2, CoD4, just cause 2, AvP, Mafia2, Medal of honor, Cod:black ops
580 wins in Farcry 2, Crysis warhead, Dirt2, Hot pursuit, Metro 2033, Hawx 2 (tho same minimum), warhammer 4000, Battleforge, BFBC2, Darksiders
5970 was 1 fps ahead in resident evil 5 but so close i call that even. Same minimum in Hawx too, but 580 is slightly ahead in avg.
580 quite ahead in in avg on lost planet 2, but 5970 still had higher minimum.
580 slightly ahead in min on Batman Arhkam Asylum, but slightly behind in avg.
You can call anyone who dont agree with you fanboy, but the black and white picture you are painting makes you look way more fanboyish than others. Both card got theire strengths and weaknesses. Looking at just the 2 reviews i looked at now, ive would say 580 pulls the longest straw. Unless you only play certain games where the 5970 shines...
(and just for the record, i wouldnt buy either of them)
It does that with new tweaked driver which supposedly gives 10% boost? That probably has nothing to do with that eh:rolleyes: When you look at a review where they used full IQ picture changes some what. 6800's look pointless as in most cases 5800's takes the cake. Even in tessellation benchmarks which supposedly 6800's were redesign for they are far from shining.
And it's not about d** measuring contest. You cant compare dual gpu to single anyway. And Nvidia has the fastest single gpu card on the marker today.
99% would never purchase 5970 or upcoming 595. Most will probably go Xfire or SLI first. \
Oh and i hear that 5970 is not so smooth gaming card anyway. It has its own share of problems. That was my point. SLI is not perfect ether so i still outright dismiss anything like dual gpu, xfire, sli as a gaming platform. For benchmarking sure but for gaming give me single gpu card under 230W full load.
They both use 10.10 and 262.99 (so does anand)Quote:
Originally Posted by railer
I dont understand your point, shouldnt you use the latest and best drivers?
You are not making any sence...
Dont let the 68XX naming confuse you, they are not really ment to replace the 58XX. But they still do better than 58XX in tesselation....Quote:
Originally Posted by railer
If it where, i would only have showed you the numbers from the card i own.Quote:
Originally Posted by railer
But i dont own either of them. So moot point...
(And by doing so i would only look un-objective and like a huge fanboy)
Now THIS is true fanboy talk :clap:Quote:
Originally Posted by railer
Ofc you can compare them, they are both graphics card right? They also happen to cost roughly the same and both are power hogs and they are also the best card each maker has atm...
You may ofc do all that if you want, no one is forcing you to do anything.Quote:
Originally Posted by railer
way to blow things out of proportions, it took a month to find the differences between both parts, not a single person who switched from NV to AMD or an older AMD card noticed a big difference...
I can't tell a difference among any of the videos on that page. Each video card under each quality setting looks the same as the others and the ALU reference as well to me. Is there a better test that shows an easily discernible difference?Quote:
Originally Posted by Amorphous
Oher sites with quality tests:
http://www.3dcenter.org/artikel/amds...ilterqualitaet
http://ht4u.net/reviews/2010/amd_rad...ded/index9.php
Funny how nVIDIA makes a scene about this, considering they use FP16 demotion too since Rel.260 ForceWare drivers and that you can't turn off in the control panel... Though you can turn it off with a utility that is not available to the public.
Got to love marketing BS. :eh:
Nice try...
http://www.geeks3d.com/20100916/fp16...e-says-nvidia/Quote:
If you wish to test with Need for Speed: Shift or Dawn of War 2, we have enabled support for FP16 demotion – similar to AMD – in R260 drivers for these games. By default, FP16 demotion is off, but it can be toggled on/off with the AMDDemotionHack_OFF.exe and AMDDemotionHack_ON.exe files which can be found on the Press FTP.
:rofl: I was going to skip this thread but this made me go reply :DQuote:
Originally Posted by slaveondope
I am a NvidiA fanboy simply because of F@H ...
It is on and was on for a couple of releases, for at least the games mentioned in the article you've linkend as far as I am aware.Quote:
Originally Posted by Vardant
What really baffles me is that I don't know why are they really doing this kind of work, since both AMD and nVIDIA hardware are capable of full-speed FP16 texture filtering. Though I don't know why I am "surprised" that they are at it again after missing shaders and objects in games like Crysis and Far Cry 2...
So you obviously have proof, if you're so adamant about this, even though NV admitted putting it in and said it's off by default and can't be turned on through CP, right?