Xbox360, so where the hell is the AA?

[one]

You just linked to an image, not the feature: http://www.1up.com/do/feature?cId=3145953 The image itself looks odd, like it's zoomed in. I'll have to fire up COD2 to find the same image...

Also, I'm not really sure what the point of the "feature" is: some games look like crap, therefore the Xbox 360 isn't next gen? (based on the front page headline: "Is this really the HD Era. Hmmm. Are the Xbox 360 Launch Titles truly Next Gen?") But whatever it takes to drive click-thrus, I suppose.

Thanks for the link, didn't notice its Back is javascript.go(-1).
As for the title, well, it's

Does the Xbox 360 Launch Lineup Fail to Live up to the Next Gen Hype? from 1UP.com:

so no problem here, I guess. Oh, and Peter Moore quotes in the article are hilarious as always.

 

[Sis]

Thanks for the link, didn't notice its Back is javascript.go(-1).
As for the title, well, it's

Does the Xbox 360 Launch Lineup Fail to Live up to the Next Gen Hype? from 1UP.com:

so no problem here, I guess. Oh, and Peter Moore quotes in the article are hilarious as always.

The title is one thing, the front page jump another. But even with the real title, they didn't compare the Launch Lineup, they compared the games that had multiplatform releases. It seems a bit misguided and unfair to take the crappiest looking games--sans COD2--and pretend this is all the launch lineup has to offer.

But given the attention they got with their silly Xbox 360/Dreamcast 2 "feature", they recognize that feeding the console wars drives ad revenue.

.Sis

EDIT: I played COD2 on my 62 inch 720p HD set. I can see some jaggies and it looks to my untrained eye that 2xAA is enabled. However, in order to detect most jaggies I have to sit my face about 8 inches from the TV. Sitting back a normal distance makes it really hard to see them. The image linked above to 1up is definitely zoomed in and does not look like that to my eyes in normal view.

 

[Laa-Yosh]

Not to mention that IMHO, many games are indeed looking far better on the X360 thanks to better shading, that is visible even on the thumbnails... and the difference between the two versions on a HDTV would be far, far greater. I have to say that this website is seriously biased in its comparision.

 

[Bill]

I wish somebody would do a real comparison.

How fricken hard is it to create a mouseover based on the SAME SHOT IN THE EXACT SAME SPOT IN THE GAMES?

Why compare two different scenes?

Gamespot and 1up have both failed terribly at this now.

 

[ihamoitc2005]

Cod2

The title is one thing, the front page jump another. But even with the real title, they didn't compare the Launch Lineup, they compared the games that had multiplatform releases. It seems a bit misguided and unfair to take the crappiest looking games--sans COD2--and pretend this is all the launch lineup has to offer.

But given the attention they got with their silly Xbox 360/Dreamcast 2 "feature", they recognize that feeding the console wars drives ad revenue.

.Sis

EDIT: I played COD2 on my 62 inch 720p HD set. I can see some jaggies and it looks to my untrained eye that 2xAA is enabled. However, in order to detect most jaggies I have to sit my face about 8 inches from the TV. Sitting back a normal distance makes it really hard to see them. The image linked above to 1up is definitely zoomed in and does not look like that to my eyes in normal view.

I have played COD2 demo and aliasing is present but is not distracting except for powerlines/phonelines. What is disracting is texture quality on walls.

 

[ralexand]

I have played COD2 demo and aliasing is present but is not distracting except for powerlines/phonelines. What is disracting is texture quality on walls.

I hear that the final game is different than the demo.

 

[kyleb]

Did you not see the screenshots posted above? The final game lacks both AF and AA.

 

[Branduil]

Did you not see the screenshots posted above? The final game lacks both AF and AA.

What I found strange is that the game doesn't even have trilinear filtering. Talk about lazy coding.

 

[Jawed]

Did you not see the screenshots posted above? The final game lacks both AF and AA.

Looks like it could be 2xAA to me. The JPEG noise is too great to be able to tell for sure.

Since 2xAA is asymmetric, one lines that is the same as another but reflected in the y-axis should have different "characters" of aliasing - which would be easy proof of 2xAA.

Jawed

 

[Sis]

Did you not see the screenshots posted above? The final game lacks both AF and AA.

kyleb, I'm looking at 3 pixels along an edge, with the middle pixel apparently being a blend of the outer two. I would have thought this is indication of 2xAA?

.Sis

 

[swanlee]

Yes let's continue to ignore the people in this thread that have a 360 at home and have played through COD2 like me. AA is not an issue, 2X AA is there, jaggies are not a problem. Dig up screens all you want and bicker back and fourth but until you played the retail version at home in HD you really have no room to talk. Whihc is better? Wild speculation or first hand accounts from people that own the game?

 

[kyleb]

I'm not wild speculating here swanlee, I have played the first two Russian missions and watched the third on an HD display at EB. AA was clearly lacking and the shots I linked back what I saw. There could be any number of reasons why it looks like there is x2aa on your setup though, it would be great if you could post up some pictures to give us an idea of what you are seeing.

kyleb, I'm looking at 3 pixels along an edge, with the middle pixel apparently being a blend of the outer two. I would have thought this is indication of 2xAA?

.Sis

You can find some examples of what x2aa does here. Unless specify which shot and exactly where you are looking can't say for sure what you are seeing, but it is likely a combination of both resampling and compression.

 

[scificube]

I have CoD2. It has jaggies though it's not horrible. Best X360 game I've played thus far. Excellent.

 

[dubert]

The thing with PGR3 is, that it uses 16x supersampling for the REAL motion blur, not AA. It's a trade off, but what can you do...

 

[overclocked]

Yes let's continue to ignore the people in this thread that have a 360 at home and have played through COD2 like me. AA is not an issue, 2X AA is there, jaggies are not a problem. Dig up screens all you want and bicker back and fourth but until you played the retail version at home in HD you really have no room to talk. Whihc is better? Wild speculation or first hand accounts from people that own the game?

Ignoring the AA and how good the game looked or "behaved" the point kyleb as others around the net laid forward as the most distracting and it certainly was that for me, is the apparent lack of trilinear/anisotropic filtering and very aggresive LOD.

 

[dubert]

I have a 360 and an HDTV running PD0 right now... And there are jaggies, but it doesnt look that bad in HD, as it would in a TV resolution. It is a launch title though. But the DOF and the vector blur effects does help alleviate it a bit for the objects in the distance etc... So I guess the team doing PD0 were rushed to get the game out or something, since I'm seeing a physics bug in it too. Don't let this get you down, I like PD0 best out of my 360 games currently.

I don't recall seeing jaggies in Kameo though, and it is by far the best looking game I have here.
And there seems to be nice AA in Condemned too.

 

[one]

I have a 360 and an HDTV running PD0 right now... And there are jaggies, but it doesnt look that bad in HD, as it would in a TV resolution.

I think fixed pixel displays tend to highlight jaggies when connected via digital in/out which Xbox 360 lacks fortunately or unfortunately.

 

[nAo]

The thing with PGR3 is, that it uses 16x supersampling for the REAL motion blur, not AA. It's a trade off, but what can you do...

supersampling? no way, they're just re-using samples from the same image..

 

[dubert]

supersampling? no way, they're just re-using samples from the same image..

I looked at it carefully, and it seems to be supersampling.
Like so... http://www.daionet.gr.jp/~masa/rthdribl/Image/mb_01.jpg
There's even evidence, when the crowd on the side of the track takes photos, and you see the flash. The flash is only done for the exact time it happens, and is not motion blurred, since it flashes that fast.

 

[nAo]

I looked at it carefully, and it seems to be supersampling.
Like so... http://www.daionet.gr.jp/~masa/rthdribl/Image/mb_01.jpg
There's even evidence, when the crowd on the side of the track takes photos, and you see the flash. The flash is only done for the exact time it happens, and is not motion blurred, since it flashes that fast.

It's not, for a very simple reason, it would be too slow, c'mon..16x supersampling!
PGR3 does a classic image space motion blur, they render a frame and a velocity buffer which is used in a post process pass to know in which direction a pixel in the frame buffer has to be blurred, that's all.
You can take a variable number of samples, even much more than 16.
Flash fx can be applied after the motion blur pass or even before if it has a zero velocity in the velocity buffer.