AMD Radeon 7000-Series 28nm (Southern Islands) | 7990 7970 7870 7770 | Discussion

[TaroT]

minimum is the only thing that matters imo... average not anywhere near as important... don't even bother with max

no

having all three the same is what matters

get 72 fps even and have it sit there regardless of what is happennign then you ahve smooth gameplay(60fps for those that believe that is the threshold)

[flu!d]

Just loading the first image difference up there is a massive difference! whats wrong with your eyes Fluid????

The foliage on the ground shows some difference in the shadows in the first shot that isn't present in any of the other comparisons, therefore it can logically be assumed that the shadows are due to a slight difference in the POV in the screenshot as it would be terribly difficult to get the exact same POV every time.

Sorry guys, I can't see the difference and I'm not really up for the lag that's present with HBAO disabled - So off it goes!

[sandeep]

Originally Posted by TaroT

no

having all three the same is what matters

get 72 fps even and have it sit there regardless of what is happennign then you ahve smooth gameplay(60fps for those that believe that is the threshold)

but its impossible to get all the 3 same. There will be scenes that might have varying degree of texture implementation e.g foliage, or for that matter any outdoor vs indoor scenes. Therefore, they will never be the same unless you impose vsync. As a result, minimum and average give a better indication as to how well the graphics card run the game.

[Mau1wurf1977]

Is there a thread for Nvidia 28nm products? Fudzilla has an update and this might be of interest.

http://www.fudzilla.com/graphics/ite...28nm-taped-out

[Stevros]

oh thats some good news. Really hope they both hit the market around the same time so we get some good prices. Remembering back to everyone paying top dollar for the 5000 series AMD graphics cards.

[yanman]

Just in time for bf3!! :dancing banana:

[Luke212]

Originally Posted by TaroT

no

having all three the same is what matters

get 72 fps even and have it sit there regardless of what is happennign then you ahve smooth gameplay(60fps for those that believe that is the threshold)

minimum is all that matters. its the only time framerate is important (duh).

i think the best indicator would be "average of bottom 5% of framerates" but our reviewers arent smarrt enough to do that. besides TOX, who could probably do it if he reallly wants to push the envelope.

[Barnzz]

Originally Posted by flu!d

As an NV user for many, many years I've fiddled with AO as well as HBAO quite a bit, and I can assure you the difference is no where near as dramatic as that example, no where near as dramatic at all.

See here for comparative screenshots:

http://www.yougamers.com/forum/showthread.php?t=124384

However lets not let this thread go too far off topic.

If you cant tell a large diff between ALL of those shots(not just the first shots) then I can understand why you think 2x570's is sufficient for BC2 on 3 screens

[flu!d]

If you cant tell a large diff between ALL of those shots(not just the first shots) then I can understand why you think 2x570's is sufficient for BC2 on 3 screens

Hmmm, nice attack Barnzz....

As not to push this thread further off topic, if you have any issues with my claims or Fraps results, please, lets discuss it via PM.

[Luke212]

i bet they released hbao because they need to use up gpu power or they wont sell new ones.

we wouldnt need it if they just make decent games that stress a pc

[Frontl1ne]

Originally Posted by Luke212

i bet they released hbao because they need to use up gpu power or they wont sell new ones.

we wouldnt need it if they just make decent games that stress a pc

Like ubersampling in The Witcher 2

[ALEXH-]

Why do some games do this?

BC2 @ 35fps is more than playable, rather smooth actually but metro 2033 @ 35fps is choppy as. Y U DO DIS?

Yes BC2 in multiplayer drops to 30 - 35fps sometimes but it's all good because I don't really feel it.

[bobsled]

Originally Posted by ALEXH-

Why do some games do this?

BC2 @ 35fps is more than playable, rather smooth actually but metro 2033 @ 35fps is choppy as. Y U DO DIS?

Yes BC2 in multiplayer drops to 30 - 35fps sometimes but it's all good because I don't really feel it.

Minimum FPS. On BC2 the frames probably take an even enough rendering time for it to be considered smooth, whereas Metro may take far longer on certain frames than others, screwing up the experience.

[Luke212]

Originally Posted by bobsled

Minimum FPS. On BC2 the frames probably take an even enough rendering time for it to be considered smooth, whereas Metro may take far longer on certain frames than others, screwing up the experience.

absolutely. this is why i say we should be benchmarking on an average of the bottom 5% of framerates. TOX if you read this can you do it? you will be better than the other reviewers. remember, the original brand that does something new gets a name for themselves even after the others come to the flock.

[itsmydamnation]

Originally Posted by Luke212

i bet they released hbao because they need to use up gpu power or they wont sell new ones.

we wouldnt need it if they just make decent games that stress a pc

not really, basically all games these days do some form of AO, it just happens that HBAO is just about the best, add a good AI (indirect illimnation) and it makes a massive difference.

if you have oblivion you can run OBGEv3 and try this:
http://obge.paradice-insight.us/wiki...usion_(Effects)

there is a bunch of different AO techniques there, the best part is it allows you to compare again the same assets in the same engine. HBAO is a massive perf hit but its also the best looking by quite a long way as well.