arstechnica commentary about the Octopiler

[Carl B]

Man the Feb event is coming any minute now - I can feel it Joe I can feel it!

(thought I'd take this opportunity to derail things a little)

 

[ban25]

Auto-parallelisation and OpenMP are hardly unique among compilers.

 

[mckmas8808]

You made the vague claim, not me. Why and how should I search for something I don't recall happening on this board? I assume you have some reccolection of a specific conversation or two.

Again, what I recall when it comes to "PS2 developers", is that they are much happier with the PS3 graphics sub-sytem relative to PS2. I don't recall hearing much at all about Cell programming relative to PS2 programming, but I do hear lots of grumbling in general about multithreaded programming in general.

I meant what the Guerrilla developers were saying on the Playstation forum. I made the thread it's called something like "Guerrilla answers forum questions" or something like that. Here are some of the quotes.

Now we're working with the Cell chip, which is a whole different ballgame. It's easier in some respect (easier to program for), but it's also a lot more complex, so a lot more cores to keep busy at the same time.

But in the end, yeah, programming a PS2's EE will give you a leg up programming for multi-core architectures, especially the Cell. Hope that answers your question! If you have any more, shoot!

And

Actually, the EE on the PS2 had vector units we *had* to use for graphics, with Cell you can write to the RSX from any part of Cell, but you don't have to. It's more of an option, so that's cool, it makes it easier. Also, cache and DMA management on the PS2 was very difficult and they've taken another look at that with Cell and made it easier (i.e. Cell does a lot for you that you had to do manually before).

 

[Joe DeFuria]

Well, in all honesty, that's one developer with what sounds like a neutral opinion to me.

 

[mckmas8808]

Well, in all honesty, that's one developer with what sounds like a neutral opinion to me.

He basically says some things are easier, while other things are harder I agree, but some people here talk as if games aren't going to be ready this year because the Cell is so hard to program for. And trust me he's not the only PS2 now to be PS3 dev that feels like this. The Cell processor is not all around in every way harder than the EE to program for.

 

[Griffith]

He basically says some things are easier, while other things are harder I agree, but some people here talk as if games aren't going to be ready this year because the Cell is so hard to program for. And trust me he's not the only PS2 now to be PS3 dev that feels like this. The Cell processor is not all around in every way harder than the EE to program for.

I think that you semplify the question
a game can be done even on a single PPE without spu, this happen with launch games for 360
I don't think that "games aren't going to be ready this year"
but
"good games aren't going to be ready this year"

the ps2 takes 4-5 years to use the power of the architecture, if ps3 is easy than ps2, then maybe it will take 2-3 years, I suppose
forget to see games tapping all the power of the console in the first year

the point is that the CGI videos put pressure on developers raising the quality bar required to not disappoint the public opinion, so waiting some tools that help devs to obtain this, makes sense.

 

[mckmas8808]

II don't think that "games aren't going to be ready this year"
but
"good games aren't going to be ready this year"

the ps2 takes 4-5 years to use the power of the architecture, if ps3 is easy than ps2, then maybe it will take 2-3 years, I suppose
forget to see games tapping all the power of the console in the first year

the point is that the CGI videos put pressure on developers raising the quality bar required to not disappoint the public opinion, so waiting some tools that help devs to obtain this, makes sense.

WHAT?!? 1st question. Do you consider a game like GRAW for the Xbox 360 a "good game"? If so how do you know that it's not going to be available this year for the PS3. And you don't really think that there will be 0 "good games" released for the PS3 this year do you?

Medal of Honor: Airborne so far seems to be a good game and that will be released this year for the PS3. You should do a search to see the possible launch games for the PS3 to see what possible to be good and what's not.

And I never thought that the PS3's full power would be realized within the first year of its existence. Where are you getting this from?

 

[Griffith]

WHAT?!? 1st question. Do you consider a game like GRAW for the Xbox 360 a "good game"? If so how do you know that it's not going to be available this year for the PS3. And you don't really think that there will be 0 "good games" released for the PS3 this year do you?

Medal of Honor: Airborne so far seems to be a good game and that will be released this year for the PS3. You should do a search to see the possible launch games for the PS3 to see what possible to be good and what's not.

And I never thought that the PS3's full power would be realized within the first year of its existence. Where are you getting this from?

calm down dude

you can't react as you done, everytime someone touch the ps3, so first, calm down


I think that graw, at this time is the best, but it don't scratch the surface of 360, I think that GOW, HALO3 and so on, will be in a superior category

I believe that the first generation of ps3 games will uses spu for not-graphical job, this is common sense, logic, call it as you want
maybe in the first year the console will reach graw quality, who knows?
but at this time, 360 will have better game on his side, all using tiling, hdr fp and msaa and starting using more the architecture
the tools that microsoft provide to devs are simply better, the power of 360 will be exploited before, and final kit of 360 have reached the devs 5-6 months before the sony's one

this is logic, but please, before replying, take a deep breath

 

[Sethamin]

This is all true, but it has nothing to do with when PS3 launches. Or whether PS3 Cell will be well utilised going forward. I don't expect Octopiler will ever be available for PS3 development, but I don't expect that will negatively impact the learning curve that's expected..you won't be able to take x86 code and run it well with little modification on Cell, but then that isn't strictly necessary (however desireable that might be). Even if Octopiler was available, and available for launch title development, while it'd make things friendlier, I'm not sure how well it could make lazy code use the chip anyway, for launch titles.

Well, as I already said, that depends on what the reason for the PS3 delays are (or if there even is a delay, which is still unconfirmed). If it's because Sony doesn't believe the launch titles are ready, then it certainly could have an impact. But we don't know what the issue or issues are at all right now, so it's all speculation. All I'm saying is that it's certainly possible.

Tangentially, how many different compilers for the CELL architecture do you think there are? I would imagine that IBM wrote the compiler that most devs are using for the PS3, considering that IBM has a lot of expertise with this (certainly a lot more than Sony or Toshiba). The Octopiler is probably built off the same codebase as the PS3 compiler, or at least is a branch off that project that shares a lot of code with it. I would be very surprised if this were not the case.

WHAT?!? 1st question. Do you consider a game like GRAW for the Xbox 360 a "good game"? If so how do you know that it's not going to be available this year for the PS3. And you don't really think that there will be 0 "good games" released for the PS3 this year do you?

Medal of Honor: Airborne so far seems to be a good game and that will be released this year for the PS3. You should do a search to see the possible launch games for the PS3 to see what possible to be good and what's not.

And I never thought that the PS3's full power would be realized within the first year of its existence. Where are you getting this from?

He's trolling. Ignore him.

Anyway, the point is not that games won't be ready at all. The point that games won't be as polished as they could be given that their most talented programmers will have to spend a lot of time dealing with the SPEs. I don't doubt that the PS3 is easier than the PS2 is some ways. They improved the biggest gripes devs had from the last generation. But they also added a whole new slew of difficult problems to address with the SPEs. So they giveth and they taketh away. (Sidebar: Saying it's easier than the PS2 is not actually saying all that much, considering how notoriously hard the PS2 was to program for.)

 

[Griffith]

He's trolling. Ignore him.

the old strategy of personal attack, never dies, sadly

if you agree with Sethamin, you are a good dude, if don't, you troll.
I don't know what is a troll for you, but I can suggest you to learn what means "respecting different ideas"

 

[mckmas8808]

but at this time, 360 will have better game on his side, all using tiling, hdr fp and msaa and starting using more the architecture
the tools that microsoft provide to devs are simply better, the power of 360 will be exploited before, and final kit of 360 have reached the devs 5-6 months before the sony's one

this is logic, but please, before replying, take a deep breath

*Breaths* Okay wouldn't the X360 devs get their final kits months before the PS3 devs because the PS3 is coming out months after the Xbox 360? And I just don't see what point you are trying to make. What's your overall point?

 

[mckmas8808]

Anyway, the point is not that games won't be ready at all. The point that games won't be as polished as they could be given that their most talented programmers will have to spend a lot of time dealing with the SPEs. I don't doubt that the PS3 is easier than the PS2 is some ways.

Okay, but doesn't this happen with pretty much all consoles when they are first launched? I mean really we all know that the first gen games on the Xbox 360 are only tipping the iceberg.

*Disclaimer: I'm not attacking you are your opinions. I'm just giving back needful debate.

 

[Titanio]

but at this time, 360 will have better game on his side, all using tiling, hdr fp and msaa and starting using more the architecture
the tools that microsoft provide to devs are simply better, the power of 360 will be exploited before, and final kit of 360 have reached the devs 5-6 months before the sony's one

I think people will take issue with the suggestion that "tiling, hdr fp and msaa and starting using more the architecture" = good games, or better games. Overall game quality is not solely a function of its technical merit, not by a long shot.

That said, solely in terms of technical merit, I think it's fairly reasonable to expect PS3's first generation to be more than a match for X360's second. There's been comment and rumblings to that effect, although we should wait till we see what's coming for ourselves I guess..

 

[Joe DeFuria]

That said, solely in terms of technical merit, I think it's fairly reasonable to expect PS3's first generation to be more than a match for X360's second.

Why?

I think that's fairly unreasonable. The question will be if any PS3 delay is long enough such that XBox360 starts it's 2nd gen title releases.

I expect XBox 2nd gen titles to start in '07. (I consider xmas '06 to be the last push of the first gen titles.) Assuming PS3 ships in '06, then we're essentially looking at first gen titles competing against one another.

 

[Sethamin]

Okay, but doesn't this happen with pretty much all consoles when they are first launched? I mean really we all know that the first gen games on the Xbox 360 are only tipping the iceberg.

Absolutely. It's just a question of degree. Anyway, all of this was just the long winded way of saying "Yes, the state of the compiler does matter". The degree to which it makes a difference is certainly debatable, and we probably don't have all the information to know one way or the other, but it does make some difference.

*Disclaimer: I'm not attacking you are your opinions. I'm just giving back needful debate.

No need for a disclaimer. I welcome intelligent debate. And actually, you can attack my opinions all you like and counter them; that's what debate is. It's ad hominem attacks against someone's character that are wrong.

 

[mckmas8808]

Why?

I think that's fairly unreasonable. The question will be if any PS3 delay is long enough such that XBox360 starts it's 2nd gen title releases.

I expect XBox 2nd gen titles to start in '07. (I consider xmas '06 to be the last push of the first gen titles.) Assuming PS3 ships in '06, then we're essentially looking at first gen titles competing against one another.

Well some people would consider that launch titles vs. first gen titles. Some people will consider the PS3's real first gen titles to be released in March 2007. And if that's the case then the PS3's first gen games (i.e. March 2007 games) will be up against Xbox 360's second gen games.

 

[Titanio]

Why?

I think that's fairly unreasonable. The question will be if any PS3 delay is long enough such that XBox360 starts it's 2nd gen title releases.

I expect XBox 2nd gen titles to start in '07. (I consider xmas '06 to be the last push of the first gen titles.) Assuming PS3 ships in '06, then we're essentially looking at first gen titles competing against one another.

OK, well we'll need to define our generations then. When I said "second generation", I was talking about late 2006 releases. If we define it as 2007, then obviously that's something we can't really comment on yet.

Leaving aside ambiguous "generation" classifications, I guess I mean that I expect that of late 2006 PS3 titles compared to late 2006 X360 titles.

 

[Carl B]

I think that's an important and valid point raised; exactly what would one consider the generational breakdowns to be as far as games are concerned? I've always kind of though of Gears of War and the like as representing the second gen of Xox 360 titles, but maybe this isn't the best way to look at it. Perhaps the first crop of games to feature engines built around the tiling would be a better yardstick?

So that instead we have 'launch titles' and 'first-generation' games, which I see as kind of a half-step away from one another. So it'll be 360 'first-generation' games vs Sony 'launch titles,' for whatever that's worth.

 

[Sethamin]

the old strategy of personal attack, never dies, sadly

if you agree with Sethamin, you are a good dude, if don't, you troll.
I don't know what is a troll for you, but I can suggest you to learn what means "respecting different ideas"

I didn't say you were a troll, I said you were trolling. As they say: "attack the behavior, not the person". Anyway, I though this in particular was kind of a troll-like comment:

good games aren't going to be ready this year

It just seemed to me like a ridiculous statement deliberately trying to inflame opinion. But, on second read, I suppose it's fine if you actually believe that. It seems pretty out there, but if that's really your view, then I apologize.

 

[Bobbler]

I think that's an important and valid point raised; exactly what would one consider the generational breakdowns to be as far as games are concerned? I've always kind of though of Gears of War and the like as representing the second gen of Xox 360 titles, but maybe this isn't the best way to look at it. Perhaps the first crop of games to feature engines built around the tiling would be a better yardstick?

So that instead we have 'launch titles' and 'first-generation' games, which I see as kind of a half-step away from one another. So it'll be 360 'first-generation' games vs Sony 'launch titles,' for whatever that's worth.

I'm not a huge fan of the 1st gen, 2nd gen, 3rd gen, 4th gen, etc, monikers. For me it seems generations should either be the number of titles for a given developer (like their first title is first gen, their second is a second gen, etc), or talking about actual devkit development (1st gen games are generally started on alpha/beta kits, 2nd gen and beyond could be started on final kits -- but that is really only two generations then). Otherwise it just seems so arbitrary to do it by years or something else.

It never made much sense to me for a first time dev releasing a game late in a consoles life to be called a 4-5th generation game, when in reality it could have just as likely been released much earlier had the dev team been so inclined. It just seems too arbitrary. The real improvements that I've seen are when developers make their 2nd, 3rd, and 4th games for a given system... the average for first time devs doesn't go up much just because they started their game 4 years into the consoles life vs 1 year (for the most part, at least).

For example: GoW is still a 1st gen game, if you ask me... it's the first game by a team at Epic for the X360.

I dunno though, the term gets overused a bit I think.