"Taming The Dragon: Next-Generation Asset Creation for PS3" (Some Lair info)

[Qroach]

ihamoitc2005 is trying to make up excuses to support his theory, and they are the most ridiculous assumptions I've ever heard of.

ihamoitc2005 look, the game had noticeable slowdowns, period. End of discussion. There's no denying this because you possibly didn't see it, or think you had less slowdowns. The game had the same slowdowns in the sameplaces for each & eveyr person that owned the game.

Can we drop this part of the thread now?

 

[ihamoitc2005]

Unneccessary and pointless attack

ihamoitc2005 is trying to make up excuses to support his theory, and they are the most ridiculous assumptions I've ever heard of.

ihamoitc2005 look, the game had noticeable slowdowns, period. End of discussion. There's no denying this because you possibly didn't see it, or think you had less slowdowns. The game had the same slowdowns in the sameplaces for each & eveyr person that owned the game.

Can we drop this part of the thread now?

I am sorry you feel that way but I have no reason to make up anything. You only have to read reviews of many games and you will find some say slowdown others say perfect frame-rate. If you do not understand or agree why manufacturing variation occurs and how it affects performance of complex systems made of many inter-dependant components it is ok, but no point making personal attack post.

 

[Shifty Geezer]

I dont know what you mean by "official" reports but look at reviews for certain games will give you enough information for you. You dont have to believe me. I am sure you have seen reviews where one review will say certain game has slowdown and and another review will say same game has no slow-down.

That is a really bad way to look at things. Human beings are subjective. eg. Do a RottenTomatoes search on a movie. I did this for Minority Report before buying the DVD for a present. Lots of people rated it highly, but some said it was rubbish. By your argument I'd assume that ws because they were watching a different film. Some said the acting was great, others said it stank. By your reasoning those that complain about the acting were watching the film using different actors.

Have you got anywhere measurements of the tolerences used in consoles? How much variation is there in the clocks? Have you any figures to show how these work together to cause large variations in system performance; a difference between smooth gameplay and frequent slowdown? Or ihave you just heard different people's observations and decided it must be due to hardware variation as all people are equal?

 

[Shifty Geezer]

Actually, I have seen performance variation in games between different gamecubes, so it does happen.

Feel free to disbelieve me too, though.

I know system performance varies and don't deny it. But in my experience more often then not it's the optical drive that makes the difference. ihamoitc2005 is talking about whole pieces of chip being disabled and clock differences sufficient to make a difference. It's his reasoning as to why he's seena flawless GC game and others have seen the same game a jittery that I'm having issues with . If he put it down to a dirty lens on the reviewer's GC rather then a complex interaction of miniscule differences of the sort that can dasiable a SPE I wouldn't have had any cause to mention it.

 

[Qroach]

You only have to read reviews of many games and you will find some say slowdown others say perfect frame-rate. If you do not understand or agree why manufacturing variation occurs and how it affects performance of complex systems made of many inter-dependant components it is ok, but no point making personal attack post.

I said your assumptions are ridiculous , i didn't say "you" were ridiculous. There was nothing personal about that. You're trying to blame something like framerate perception on hardware, when it's in fact already known to be a user perception issue. How many times have you seen it discussesd how some people can't tell the difference between 30fps and 60fps? Many times no doubt...

Basically it irks me that you're even trying to come up with reasons like this, instead of realizing perhaps "you simply didn't notice the issue at hand". Jeez, I mean what do you take us for? yeah there's small differences in hardware and there's a chance that someone has some broken and or bad hardware, but this was generally seen by many that have played the game. some people just don't notice it as much as others.

 

[Qroach]

Shifty, having a dirty lens would only have a noticeable effect in games that stream all their data off the disk. From what I recall, the games we are discussion would load individual levels in to ram. Or at least they didn't have seemless or zero level loading...

 

[Bohdy]

Well I don't exactly share his argument, and would first chalk up a reported difference to the subjectivity of the viewer (many people just don't notice framedrops), but I'm just saying that I have personally seen variation to the extent of a portion of a game that runs at 60FPS consistently on one GC, dropping to 30fps sporradically on another GC, so what he is saying may not be complete bunk.

Anyway, personally I though that the framerate of Rebel Strike was overall an improvement over that of Rogue Leader, and generally quite good even if it was not always consistently 60fps. There were parts of it that did run at 60fps, with nearly every screen pixel bump-mapped and stacks of geometry, though, which is the important part in this argument, and not the average framerate of the whole game, right?

 

[ihamoitc2005]

Slowdown different from taste

That is a really bad way to look at things. Human beings are subjective. eg. Do a RottenTomatoes search on a movie. I did this for Minority Report before buying the DVD for a present. Lots of people rated it highly, but some said it was rubbish. By your argument I'd assume that ws because they were watching a different film. Some said the acting was great, others said it stank. By your reasoning those that complain about the acting were watching the film using different actors.

Movie/acting review is based on cultivated taste, like preferance for food, but slow-down is slow-down, like difference between running and walking, different from taste. If you question other people's ability, especially experienced reviewers, to notice slow-down without aid of frame-rate counter perhaps you must wonder if youre experience with slow-down was not actually slowdown.

Have you got anywhere measurements of the tolerences used in consoles? How much variation is there in the clocks? Have you any figures to show how these work together to cause large variations in system performance; a difference between smooth gameplay and frequent slowdown? Or ihave you just heard different people's observations and decided it must be due to hardware variation as all people are equal?

What would you do with such data? You are the one suggesting you have knowledge of clock-speed variation of specific components. You said it in your previous posts if you look.

As for how how it can cause variation in performance at limit, you must first realize that even what you call "frequent" slowdown is tiny fraction of time during which game runs how much of that is due to excess load and how much is due to natural performance variation is a good subject for you to do for phD. In the mean time, try to prove that mathematically it is possible given system of many complex parts operating at different speeds where the actual speed can vary by, according to you, +/- 1% to maintain perfect and predictable performance scaling to simultaneous limit of many of those interdependant parts. That is good subject for second phD and will no doubt get nobel prize because no one has been able to make this happen.

Back to topic, given how F5 star wars games = best overall performance of previous gen, I am very interested in how well they will push CELL and balance SPE for graphics vs other tasks. EA is using 3 SPEs for graphics and 4 for everything else.

 

[Shifty Geezer]

What would you do with such data? You are the one suggesting you have knowledge of clock-speed variation of specific components. You said it in your previous posts if you look.

blah blah...PhD...blah blah

These arguments are so very annoying. You say a game is written to run smooth, and that when a reviewer sees glitches it's because of hardware differences. You offer no real-world evidence to support this idea such as measured chipset speeds. And then when someone suggests you're wrong, rather than come up with the evidence you use the old 'prove your point first' manoeuvre. Why do people do that? Why don't they just argue their point, instead of always complaining their opponent failing to support their claims with data? I've questioned you POV. Can you back it up with facts and figures and anything substantial? If not, it's what you believe, which is fine, just so long as you appreciate what you believe to be happening might be wrong. As for me, I believe you are wrong, but said as much. I've no evidence to support the contrary other than little bits of knowledge on electronics

As for my +/- 1% figure, that's based on what little I've done measuring clockspeeds between PCs. Ishould have qualified it with some uncertainity as to that actual figure. If you measure a series of 500 MHz Pentium II's you find a varience in the clock, but it's always small. That's based on tolerances on the manufacturing plant. Anything too far from the clock speed won't be released. Same on consoles. They ensure if there's severe defects that console won't be released. Okay, that's me making wild unsubstantiated claims that these companies have quality control departments. I've never seen these for myself, nor read about them specifically, so for all I know Nintendo release GC's with CPU's clocked at 300 MHz because they can't be arsed to test them, and so yes the hardware is substanitally different and that causes the different experiences reviewers have had.

However, TTBOMK the varience between consoles is very slight, and there's no real evidence to support the idea that these small variances add up to a game running smooth on one platform, jerky on another. Until someone can present such evidence I'll stick to believing that the hardware companies' manufacturing capabilities are actualy pretty good and variation beween people's description of their experience of a game is more likely due other factors.

 

[London Geezer]

Movie/acting review is based on cultivated taste, like preferance for food, but slow-down is slow-down, like difference between running and walking, different from taste.

Nonsense. Slowdown is mathematical. When running a game at a certain speed, slowdown is what happens when the frame rate SLOWS DOWN. Everyone sees it, only some people don't give a damn.

As for how how it can cause variation in performance at limit, you must first realize that even what you call "frequent" slowdown is tiny fraction of time during which game runs how much of that is due to excess load and how much is due to natural performance variation is a good subject for you to do for phD. In the mean time, try to prove that mathematically it is possible given system of many complex parts operating at different speeds where the actual speed can vary by, according to you, +/- 1% to maintain perfect and predictable performance scaling to simultaneous limit of many of those interdependant parts. That is good subject for second phD and will no doubt get nobel prize because no one has been able to make this happen.

I'm sure the Nobel people have much more interesting things to take care of. Like world peace or cancer treatment drugs. And giving prizes according to the political mood in the world.

 

[_phil_]

I'm sure the Nobel people have much more interesting things to take care of. Like world peace or cancer treatment drugs. And giving prizes according to the political mood in the world.

cough..cough *remembering when kissinger had peace nobel price* cough..!
sorry for the disruption.

 

[pixelbox]

i thought noticable slowdown was cause the hertz of a TV...

 

[scooby_dooby]

Nonsense. Slowdown is mathematical. When running a game at a certain speed, slowdown is what happens when the frame rate SLOWS DOWN. Everyone sees it, only some people don't give a damn.

so you're agreing with him then? He's saying framerate is NOT subjective, to refute Shifty's point of how movie reviews vary. He's saying reviews are completely subjective, framerate is not, so you've just agree with the point he was trying to make.

 

[Qroach]

This thread hertz...

 

[ihamoitc2005]

Yes.

so you're agreing with him then? He's saying framerate is NOT subjective, to refute Shifty's point of how movie reviews vary. He's saying reviews are completely subjective, framerate is not, so you've just agree with the point he was trying to make.

Yes thank you for explaining. Sometimes I think it is hard for some to understand my style of writing and so misunderstanding occurs. I am trying to improve this.

 

[overclocked]

Yes thank you for explaining. Sometimes I think it is hard for some to understand my style of writing and so misunderstanding occurs. I am trying to improve this.

You spell good but you change words in "yoda" type.. I think its charming.

 

[Mefisutoferesu]

He spells WELL not good. ^_^

 

[LunchBox]

For a tech forum, there seems to be a lot of English Teachers around.

just kidding

 

[TrungGap]

Actually manufacturing flaws or "variation" can make a big difference. Why do you think so x number of pipes in GPU are "disabled" or PS3 CELL has only 7 functional SPEs? Because although units are identical in design, manufacturing process is such that it is not possible to have perfect part duplication so natural variation occurs. This is not just true of microchips. Also true in automobiles, or anything else that is manufactured. Variation in quality of final product is unavoidable problem unfortunately so margin of error must be accounted for in design to limit performance variation. But more complex the system is the more difficult and costly it is to limit performance variation in finished product.

Let's consider a car. Can you imagine the variation in the parts that go into a car? Now, imagine the fault tolerance of the car in order to pass QA. The variation between different car of exact model is consider huge. So when a car's gas milage is rated for the model, you'll going to get a wider variation and then you have to factor in the driver. Eh, you're not going to get a car rated 30MPG that will get you only 15MPG or a 15MPG suddenly getting 30MPG for another driver.

Yes, you are correct in your assumption that there variation. But it's not going to be to a point where it matters. Because if that was the case, developers would have factored that in when they design the game/product.

 

[mckmas8808]

Good response TrungGap. You analogy seems pretty good to me. But I don't make games so...