SPU usage in games

Don't worry, I won't :)

I do, however, find it odd that you would even think PSSG was using EDGE, though.

Where from this statement: Well, I can confirm right now that PSSG doesn't use EDGE at all.

...did you get that I was thinking PSSG might be using Edge?

Tell you what, why don't you find me an article or interview or something that discusses this mythical 'PSSG,' and we'll take it from there. ;)

(no help from the outside folks)
 
Where from this statement: Well, I can confirm right now that PSSG doesn't use EDGE at all.

...did you get that I was thinking PSSG might be using Edge?

Tell you what, why don't you find me an article or interview or something that discusses this mythical 'PSSG,' and we'll take it from there. ;)

(no help from the outside folks)

Rick Roll

Also, why would you bother confirming that PSSG doesn't use EDGE at all? I mean, it's just personally amusing to me that you would even announce this "discovery" of yours. But, then perhaps you really need to understand the creation and evolution of PSSG to understand why I find it so irrelevant.
 
Word to the wise, read subject titles more carefully otherwise you might think there's "spooge in games". Carry on...
 
Rick Roll

Also, why would you bother confirming that PSSG doesn't use EDGE at all? I mean, it's just personally amusing to me that you would even announce this "discovery" of yours. But, then perhaps you really need to understand the creation and evolution of PSSG to understand why I find it so irrelevant.

That's a good article you linked to; I consider it the best there is on the subject. In fact, I had a feeling it would be the one you returned with. Do you know who wrote it? ;)

Here's what I see in you Xenon: arrogance. If you had read back even two pages in this thread (#169), you wouldn't be in the embarrassing position here of posting an article at me that I myself wrote. But, then again you feel yourself too good for reading more than the page in front of you I guess... Hell, if you'd even read and comprehended the post you yourself initially responded to you wouldn't be here entwined in this discussion, premised on your thinking I said PSSG was based on EDGE when I clearly said the opposite.

And, now you see why my comment to that effect isn't one of 'discovery' either - simply of clarification, with me myself as a source that can be trusted on the issue, through my own discussions with the director of the project.

I've read through you post history Xenon, and you come off like an a**hole; move the attitude up towards 'amicable' a couple of notches going forward from here. And read before you post.
 
Last edited by a moderator:
That's a good article you linked to; I consider it the best there is on the subject. In fact, I had a feeling it would be the one you returned with. Do you know who wrote it? ;)

Here's what I see in you Xenon: arrogance. If you had read back even a two pages in this thread (#169), you wouldn't be in the embarrassing position here of posting an article at me that I myself wrote. But, then again you feel yourself too good for reading more than the page in front of you I guess... Hell, if you'd even read and comprehended the post you yourself initially responded to you wouldn't be here entwined in this discussion, premised on your thinking I said PSSG was based on EDGE when I clearly said the opposite.

And, now you see why my comment to that effect isn't one of 'discovery' either - simply of clarification, with me myself as a source that can be trusted on the issue, through my own discussions with the director of the project.

I've read through you post history Xenon, and you come off like an a**hole; move the attitude up towards 'amicable' a couple of notches going forward from here. And read before you post.

Thanks for calling me names, but trust me, the feeling is mutual. As for writing the article - is that what they call "writing" nowadays? Because it seems it was more of a transcription service than an actual article. I don't mean to be rude, but there's an obvious reason why you went unnoticed in that piece.
 
Well, that was a fun ban. A quick tip: if you're going to try to flame a mod based on an interview that the mod wrote, you should make sure your post history is overwhelmingly helpful and positive. Otherwise, well, I lurk the CF for a reason :D
 
Excuse me? Just because their name is in the credits does not automatically mean they contributed to the creation of the engine in Resistance or the application of any EDGE technologies to the game. Mark Cerny, for example, is a consultant with Insomniac which is probably why he was thanked in the credits.

Ok, lets put it straight: I know 100% they did contribute.
 
As you know graphic engines are always evolving and the graphic engine used for Resistance pales in comparison to the current iteration of the engine being used in R&C and Resistance 2. As an example of this evolution, the physics processing power of the engine has improved 400%.

So Resistance 2 is already in the pipeline? When you say it has improved 400% you mean it takes 4 times less cpu time? Still an impressive number, where did you get that number from? Its probably accurate, Havok 4.5.x where optmized on/for PS3 and I heard a similar number there.

Its off topic but does anyone know if there will be something similiar XNA Game studio for PS3?
 
As for writing the article - is that what they call "writing" nowadays? Because it seems it was more of a transcription service than an actual article. I don't mean to be rude, but there's an obvious reason why you went unnoticed in that piece.

I think a lot of folk would be surprised by the time that goes into an interview of this nature. Questions are asked and responded to one or two at a time via email, and time zone and work considerations can lead to some big response delays. The idea behind an email interview being that the response might trigger a more logical path for the next question to be asked, and in turn lead to an interesting branch of coverage that would have gone otherwise unnoticed. Not only that, but it's easy to ask for - and receive - further clarification on something I feel has remained vague, and allows for perfect quotation. The tandem (and very active) side conversation that takes place between me and the interviewed is something you don't see in the final product, yet the product itself very much reflects it.

The interview with Jason took roughly twenty days to conduct once the approval from Sony came, for example. The goal of course is to produce a comprehensive overview of the subject matter described via the sources own words, and certainly as far as PSSG goes, that piece with Jason is still the source I myself would turn to read if I needed a PSSG reference. Let's put it like this - if the interviewee is asked and answers all the questions I myself as a reader would want to have asked and answered, then I've done my job - I don't go for fluff. That I go 'unnoticed' is the strangest thing to hear... why would I ever expect to be noticed for that kind of article? It's obviously Jason Doig and PSSG that should be remembered; hell, I don't even use my full name in the questions.

Anyway it was the first public interview allowed on the topic by Sony - I think that alone says all that needs be said.
 
Last edited by a moderator:
Thanks for calling me names, but trust me, the feeling is mutual. As for writing the article - is that what they call "writing" nowadays? Because it seems it was more of a transcription service than an actual article. I don't mean to be rude, but there's an obvious reason why you went unnoticed in that piece.
:oops:

Wow! Using a person's own article to prove how misguided they are...and then following that up with a put-down instead of a back-down?! Why do these forums allow the word 'oops' and the :oops: smiley if no-one's going to use them?

This little discussion is worthy of the hall of fame IMO!
 
What a graceless human being. The ps2dev.org forums have a 'Hall of Shame' where they move threads and posts that are a great example of what should not happen in the forums. This exchange right here would fit there nicely.

However, this thread is otherwise quite good. So I'd like to return to it. :p

It was a great interview Carl, no question about it.
 
I missed it too.

I do want to point out that real-time surround mixing up to now has always been considered a rather hefty task, and I think Lair (and a few other games) supports 7.1 surround. Now for doing real-time surround stuff, you need to track the location of the sound origin in 3D space and mix all sorts of samples (from say all the different dragons, fireballs, and anything else that makes sounds) into a 7.1 surround sound uncompressed audiostream. While I'm sure that one day this may be reduced to a fraction of an SPU, I imagine it would be easy for the amount of data involved as well as the number crunching involved to take up enough for a development team to dedicate an SPU to it.

Of course, I really don't know anything about this, so I'm also hoping someone with actual experience on this subject chimes in.
Dredging up an earlier topic here but more info has come to light. HS sound effects will apparently amount to 10GB of storage.
The Numbers:
10GB of sound FX, approximately three and a half hours of music, 4,500 lines of dialogue

The audio on Uncharted also sounds inspiring, dynamic and a discrete 7.1 presentation???
The Audio Director speaks: "If I could have a metric for how well the game is doing sound-wise, it would be how far I can get through this game with my eyes closed."
Hopefully we'll get more info closer to the release date.
 
ProFX

Any thoughts on Allegorithmic's use of procedural textures? The idea of environments & characters updating in real-time is appealing. The technology does sound demanding though:
SD: Procedural techniques are demanding in terms of power, so the more power you can get, the better it is in that respect. Going multi-threaded is quite easy generally speaking for image-based work (you can divide the image in several pieces and work in parallel), and so the more cores you can get, again, the better it will be.

That said, programming the PS3's eight Synergistic Processing Units (SPUs) can be quite challenging from time to time, but there is definitely a lot to gain from the architecture, and ProFX on PS3 should run impressively fast. (We should get confirmation on that by the end of the year or so ;)

A very cool side effect of the PS3's architecture is that ProFX should be able to continuously stream textures -- using one or two SPUs to compute the textures to be given to the GPU to be displayed, that's a huge bonus for adding rich content to games without having to stream that content from the BluRay or overloading the sole GPU of the machine.
 
Procedural textures are an ideal fit for something like SPUs; unfortunately, they present enormous art production difficulties.
 
Procedural textures are an ideal fit for something like SPUs; unfortunately, they present enormous art production difficulties.
In some approaches that may not be an issue. Procedural dirt seems an ideal solution. GTHD Prologue looks fantastic....but still has repeating ground textures! Creating, or adjusting, grass and dirt textures on the fly ought to be a good use of SPEs without being a hassle to integrate with the rest of the assets. With procedural dirt you can mix up vehicles say, so the one same model doesn't have 8 identical clones, but sufficient variations to look like 8 different units. These procedural techniques are, at east in theory, straightforward to implement, unlike procedurally generating 'ordinary' diffuse colour textures where design is an issue.

These ProFX demos show this well. http://www.profxengine.com/index.php?PAGE=GALLERY.DEMOS
 
Worked well enough in Roboblitz (50mb XBLA game on UE3)

Were you on the Roboblitz team?

The art production difficulties I'm talking about aren't something you can see from the finished product; it's the problem of replacing texture artists, who are used to painting freehand in Photoshop, with something like "texture programmers" who can decompose a target (painted) texture to a series of mathematical operations.
 
The art production difficulties I'm talking about aren't something you can see from the finished product; it's the problem of replacing texture artists, who are used to painting freehand in Photoshop, with something like "texture programmers" who can decompose a target (painted) texture to a series of mathematical operations.
There's different types of procedural synthesis. One is to record the artist's actions in a graphics application as a super-macro, and then repeat the same graphics functions at runtime to generate the texture. eg. Record the points of a beizer curve and fill gradient settings set by the artist, and recompose into a texture.

It's also likely some of the artists are using procedural synthesis techniques offline already, so it shouldn't be such a jump. eg. ID5 populates the scene with procedural content first of all before the artists dive in and place stuff.
 
The art production difficulties I'm talking about aren't something you can see from the finished product; it's the problem of replacing texture artists, who are used to painting freehand in Photoshop, with something like "texture programmers" who can decompose a target (painted) texture to a series of mathematical operations.

Um, AFAIK in Roboblitz the artist makes the texture, then it gets thrown in some kind of compilator thingy that generates a macro that will look very close to the originial, but generated on the fly (thus it will be different each time, but very slightly)
 
Back
Top