Half-Life 2 XB360 won't be using tiling to achieve A.A.

New Architecture

I think what Acert93 says is is traditional rendering technique is not having acceptable performance with unified architecture. He says many ALU is not used.

What i want to understand is why developers cannot use extra ALUs with traditional rendering technique like Oblivion.

 

More ALU math vs. texture lookups in pixel shaders perhaps?

Can't shade a gajillion triangles after setup and intense vertex processing?

Cross platform development prohibits it? PC GPUs don't have ALU counts to match Xenos tit for tat except at the high end of the spectrum. (this would be true in the case of Oblivion as well)

Developers need time to learn how to best exploit Xenos's raw power and feature set?

These seems plausible reasons enough to me.

 

ALU rate

So 1900 and Xenos have less ALU use % than 850 and 7900 for Oblivion, they have more ALU idle/frame?

Also, does PGR3 (not cross platform) have very much ALU idle?

 

This was kind of answered, but just to be clear it's not the bandwidth that is lacking to do 720p FSAA without tiling. The amount of memory is too small.

 

i believe when you have unified architecture with unused features like memexport and an API which is inbetween dx9 and dx10 still in its infancy compared to an already established GPU related API of Nvidia, it would be easier for devs to show with Nvidia thier strengths eaarly and the ones with ATI Xenos, theier strengths later

i believe you are wrong in this one and hl2 screenshots of 360 will show that. just like you guys developed a system of not wasting system resources and got a form of AA with faked hdr at the same time keeping the visuals intact, Valve, who are clearly more superior developers than you guys, no pun intended, can achieve the same effect with AA on 360.

 

thats why i believe they will use memexport, you cant use tiling when using memexport and you cant use memexport when using Tiling so I believe when taking into the concept and features of using memexport, Valve could use this implementation to solve the AA problem without using AA.

 

Actually that's sounds about right. Check 1900 performance with Oblivion. I'm sure the game not being crafted to take advantage of oodles more shader power in the 1900 has something to do with why performance doesn't scale to the levels some were expecting initially. It's atypical that high end hardware is pushed to the limit int the first place in the PC space as I'm sure you're aware not to mention a jump like the 1900 or Nvidia's SLI on a single card offering.

The case is not as strong for Xenos not being leveraged but nevertheless Bethesda only had so much time before the game was launched and then Bethesda is not overly infatuated with maximizing performance if you look at their history.

 

STFU

I've been lurking for six months because I've had nothing worth adding to the discussion here. Until now.

 

What is "this implementation"? Can you, or anybody, answer Vysez's question to you before making such an abstract comment?
http://www.beyond3d.com/forum/showpost.php?p=795608&postcount=45

 

That made absolutely no sense.

And that comment to nAo... It's been noted and reported.

 

And yet you wasted it on that, GG.

Didn't ERP say something back when the board had good threads still going about tiling needing the developer to design the way their rendering engine worked a certain way? A game designed for the PC first would almost by definition not have that.

Someone mentioned something about RSX vs. 360 GPU as far as getting the most out of it, the difference is that the RSX is very similar to a part and/or technology that developers are familiar with while the 360 is unlike anything they've used before. To take advantage of it means they have to do something new while with the RSX they can do things as they've always done them. This is a very similar situation with regards to Cell vs. a conventional processor - developers who expect they can just port their code over from PC-land are going to be pretty disappointed with the performance. So that's an advantage for the RSX right now and we'll have to wait and see if it's always an advantage or if the ATI GPU has some legs.

 

Not wasted. If I had one thing worth adding in six months, it was that.

Now I'm going back to lurking. Much like those who have nothing substantive to add to the mostly excellent disscussion on Beyond3D should consider doing.

Have a nice day. =)

 

Wow, if ever there was a thread that needed closing...

 

when you dont use tiling (i dont know if they meant predicated tiling) then the performance of a 360 is just a system not getting free AA. There is an option for MSAA and FSAA as well. When you dont have the advantage of free AA, shader power is reduced trying to implement AA "inefficiently" but using memexport this can be countered by writing to memory and vfetch for reading from memory.

Valve has already stated the superior tools available for 360 and this shows the flexibility MS is providing for developing for 360 that even though tiling is prefered, its not the only solution.

 

Bandwidth

Memexport cannot help AA (ROPs inside EDRAM unit no?) but, if it can have help for AA (it cannot) there is big problem for bandwidth (22 GB/s CPU+GPU) no? So only high bandwidth connection is EDRAM (32GB/s). I think sometimes AA is no big deal. If they can make real 720P w/HDR and have no AA graphics can be very nice on 720P TV. I like more good frame-rate.

 

Ok, I will bite: What does unified shaders and the general strengths of the approach have to do with tiling hurdles or developers still locked in on SM2.0 (and below) shading models or the low amount of math to texturing typical of current game design?

Having more advanced featureset and better shader utilization or fillrate doesn't help you one bit if you are texture limited, are deploying unrefined middleware, fighting the CPU and multithreading, or just plain porting a game that is ill-conceived for the target platform given the time allocated for the conversion.

To be so blunt, if we swap what you are saying across the board we could say that Cell's supposedly processing edge should show its strength with less effort... so I believe it is the other way around. (Not that I do of course!)

But that doesn't really say much for the multitude of difficulties developers are facing in very finite time budgets. Of course you could probably lecture me on these points much more eloquently

On Xenos, if you are not shader limited in your software all the shading power in the world won't save you and give you more performance. I don't believe there is a very good arguement that RSX has a richer featureset or that RSX could not benefit in many ways by using eDRAM like Xenos (yet duly noting the reservations and limitations of the eDRAM in Xenos, but there is no denying the bandwidth it provides for its designated tasks).

What an absurd rebutal!

Ok, tongue in cheek comments out of the way, Megadrive would have been best probably giving some reasons why... although to his benefit he has stated them before and they are well known. Reading his mind I believe he probably had the following in view...

Xenos has a much broader and usable featureset, significantly faster dynamic branching in pixel shaders, much higher sustainable fillrate, easy to access features (like FP10) that give an immediate performance benefit over other hardware methods, and so forth.

Just looking at ATI's X1000 series and Xenos and where D3D10 is headed, and how RSX was implimented into the PS3, I can see a number of areas where RSX is going to show limitations in some future 3D game engines. RSX may be an SM3.0 GPU, but is has very poor dynamic branching in pixel shaders and its vertex texturing is very slow. IMO, It is an SM2.0 GPU with SM3.0 features stapled on. It really doesn't do any core SM3.0 feature exceedingly well. Further, RSX has a split memory pool, is facing texturing penalties from XDR, has half the fillrate a typical GPU with its texturing and shading power possesses, has a significant performance penalty for FP16 blending and filtering, and while it has a paper flops advantage it has less effecient ALU utilization and also ties up shader ALUs when texturing.

Not that Xenos is perfect--not by a long shot. Tiling has been an issue (too small for 720p 4xMSAA, requires a renderer rewrite if there is not an early Z pass, predicated tiling support was not really resolved until Spring 2006, memexport requires a 3x load on the CPU and was not compatible with tiling, etc) and MS did not resolve it very quickly, MS did not get SM3.0 capable GPUs into developer hands until nearly the consoles launch, the dev kit situation was fubared and illconceived, Xenos is looking 2 years down the road and has a significant amount of transistor budget allocated in a way that doesn't help a lot of current game engines, has less peak texel fillrate, etc.

There are hurdles on both sides. But I agree with Megadrive: When developers begin developing SM3.0-based engines and begin exploring basic D3D10 features there is going to be decided overlap with the core focus of Xenos.

Of course I can see the arguement now: But the PS3 can offload work to Cell. Which is true (of both consoles), but the topic is the GPUs, not the systems.

And as I mentioned earlier RSX is a much better understood -- and developed for -- architexture. The seminal aspects of RSX were introduced in the middle of 2004 via NV40. G70 is a mere refinement of the NV40 base (some bulking of the shaders, refinement of the workflow and features, and of course faster) and RSX is a direct derivative of G7x.

Sony developers have had NV40 SLI since early 2005. That is an obvious advantage to getting Xenos in July/Aug 2005 and going gold in October 2005. And MS dropped the ball on a number of issues, e.g. predicated tiling received a number of fixes in the API in March/April 2006. Likewise we have seen the problems associated with trying to shoehorn PC game engines into Xenos, which is not really unexpected as it is not a direct descendant of a PC GPU like RSX and NV2A were. Heck, you cannot even begin using tiling unless your engine does an early Z pass. Xenos has a lot of power, but you have to tailor your game to it. It is very much like the current CPU situation: They have a lot of power, but you MUST use multiple cores if you wish to extract that power. Tossing a single threaded PC game engine on Cell/Xenon is NOT going to get the results you want. Ditto Xenos. RSX, to Sony's credit, is in a MUCH better situation to port existing 3D libraries and to hit the road running. It may be the least progressive part in all the consoles, but it is also the part most well understood and ready to be exploited--today.

Of course many of us stand on different sides of the fense on these issues, even developers do. Some developers have more resources and essentially have first party status and have the luxury of figuring this stuff out, others have to make do with their purchased 3rd party middleware and get the best out of it in 18 months and move on.

Oh well, we have all ping ponged these things back and forth a million times.

Some people feel the Xenon-PC approach to multi-core, some feel Cell is. Some feel Xenos is too future looking and has too many limitations on the eDRAM, others feel RSX is an outdated model that will show its age and Xenos is struggling due to system and last gen porting "first one out" issues like the PS2.

To degree most will be right and wrong.

Ps- Boy this thread went from ugly... to nasty! This is why I avoid posting here! eeeeek!

 

I just wish I knew what Valve was talking about (remember the claim about tiling and AA posted 8 pages ago?).

My guess is that any DOF, bloom and motion blur they use will be "enough AA." Though I do wonder if the eDRAM's enormous fillrate and powerful blending features would enable some more ingenious method of AA.

 

Well one could hope, but the eDRAM in Xenos seems pretty restricted in regards to its intended purposes (Z, Alpha, etc) and is not as eloquent of a design as say, GS or Flipper, but then again it was not intended to be within the system design. We will see... dare I say PR speak?

My guess is HL2 will ship either at a lower resolution and upscales ala PGR3 or, more likely, they will take a few pages out of the Gears of War book and use DOF and motion blur and try to minimizing contrast to get a cleaner look. Call me a skeptic.

 

memexport and tiling are completely unrelated and you most certainly can use memexport when tiling.

And people, didn't you see nAo's . He was joking, at least partially. A unified architecture should be more efficient and easier on the developer. It's just that Xenos is more than a unified architecture. It has nifty features like edram, tessellation, and memexport that require effort to use well.

 

To my knowledge, at one time this was not true but I believe this has changed with newer API updates.

Dev's are not allowed to joke No, but he brings up good points and they are worth discussing. e.g. He was one of the first to comment months ago about reservations about eDRAM and that he felt relatively it was less important this generation than last generation and many of his points are very valid and definately noteworthy in the broader context of discussion.

Alas, these sort of discussions tend to implode due to the makeup of mixed forums and the fact only a fraction of a percentage of a posters really have any firm idea what is going on, a percentage have a solid understanding of the concepts, and the rest of us troll/get trolled left and right when trying to catch up

Anyhow, nAo has a lot to contribute as a [nearly] first party and obviously has a great mind (see NOA32) and has shown to be pretty creative in ways to really address the shortcomings of a platform. This is what good devs do, even if they half jokingly blow up the B3D forums... or not jokingly? But to steal a page from Mintmaster: I think the PS3 has better developers overall. Not that MS and Nintendo do not have good devs, but Sony has shown over the years to have a slew of really, really great talent and I will continue to believe that the better developers will create the best games. We can comment on hardware all we want, but what really matters is what developers can get out of hardware and Sony's devs have a long history of showing on average they get the most out of what they have.