LAIR Thread - * Rules: post #469

Status
Not open for further replies.
The art is good, the tech is good, the detail is good, even the lighting is technically good... Yet when they all come together it looks it just doesn't look right. These are certainly less than the sum of their parts...

I think it's just the use of colour is a primary problem.. It's all mud

Where is the colour...?
This sums up "next gen graphics" nicely ;)
http://www.vgcats.com/comics/?strip_id=224
 
Meh... Where is "Oprah" the Giant Sea Serpent ? :p


EDIT: Ok, I decided to add more meat to my own post, although truth to be told, I really wanted to see the giant serpent close up.

To Factor 5, thanks for the colors. I remember the old shots were brown throughout. The screens look great ! Now let's see this in motion and how well it plays.
 
Last edited by a moderator:
There are a couple which are new I think. It seems to be Sony's MO with Lair at the moment, pushing out one or two new screenshots with the rest being older. There is something interesting I noticed though on the 1080p versions (they're on the PS3.EU website) where there's some kind of sawtooth filtering along near horizontal edges (like the dot crawl effect you see from viewing games/movies over composite video signals). It doesn't seem to be present on near vertical edges.

(Magnified to 200%)
http://www.ps3.eu/flash/mediapool/game_lair/lair_0.jpg

lair1.png
lair2.png


http://www.ps3.eu/flash/mediapool/game_lair/lair_5.jpg

lair3.png
lair4.png


I don't think it's noise introduced by jpeg compression as the screens were over 1MB in size, and it's too consistent throughout the whole image. I don't think it is the product of the capture method either, it seems an intentional effect to reduce aliasing (though I feel it may be more effective in motion than in still screens). There's also a lot of ghosting along vertical edges which makes me wonder if it's utilising a framebuffer lower than 1920 pixels wide. Although there are all manner of depth of field and post processing effects going on, which may be the cause.
 
Last edited by a moderator:
It could indeed be sarcasm. I mean the only thing that comes to mind here is that it looks like the colour samples you would produce when applying AA on an edge, but they just haven't been filtered down.
 
Umh..I think I know what it is...they're clever..oh they're so clever :)
They rendered to a 960x1080 framebuffer with 2xAA, then used the unresolved data as the final image, right? Basically they reinterpreted the 2xAA backbuffer as a No-AA 1080p image. You can see that, with the exception of edges, the pixels are in horizontal pairs as you'd expect from the samples in MSAA. The post processing sometimes hides it, but you can really see it in the dragon wings.

A tad more expensive but IMO a clearly better image than you'd get from the free horizontal "scaler" of the PS3 on a 960x1080 image.

I've actually wondered why ATI and NVidia didn't offer something like this with a "turbo mode", especially for budget cards. Render at half or quarter resolution with 2x or 4x MSAA (preferably ordered grid to avoid the sawtooth effects) and just pretend you actually rendered at full res.

I've been suggesting for a while that this could be a very useful technique for shadow map rendering on Xenos. Free 4xAA + double Z = 64 pix/clk!
 
They rendered to a 960x1080 framebuffer with 2xAA, then used the unresolved data as the final image, right? Basically they reinterpreted the 2xAA backbuffer as a No-AA 1080p image. You can see that, with the exception of edges, the pixels are in horizontal pairs as you'd expect from the samples in MSAA. The post processing sometimes hides it, but you can really see it in the dragon wings.

A tad more expensive but IMO a clearly better image than you'd get from the free horizontal "scaler" of the PS3 on a 960x1080 image.

I've actually wondered why ATI and NVidia didn't offer something like this with a "turbo mode", especially for budget cards. Render at half or quarter resolution with 2x or 4x MSAA (preferably ordered grid to avoid the sawtooth effects) and just pretend you actually rendered at full res.

I've been suggesting for a while that this could be a very useful technique for shadow map rendering on Xenos. Free 4xAA + double Z = 64 pix/clk!
Wow, that makes an awful lot of sense and explains why only the near horizontal edges look the way they do. And for completeness, here's a zoomed in (600%) pic showing the pixel pairs:
lair5.png
 
If this is the case, then is it truly 1080p? Or as long as horizontal resolution is 1080p (ignoring vertical res) we can call it 1080p?
 
If this is the case, then is it truly 1080p? Or as long as horizontal resolution is 1080p (ignoring vertical res) we can call it 1080p?

As long as nAo doesn't confirm and nobody knows, its all speculation what "true" res it runs at. That means as far as anyone can tell, its truly 1080p
 
Status
Not open for further replies.
Back
Top