Why does Sony create such wacky architectures?

marconelly!:
Btw, Dead or Alive 2 on DC does look pretty dated by today's standards.
Dated compared to which PS2 fighting game? Would that be the awfully interlaced VF4 with its restrictions to single-tiered environs and only two characters on screen simultaenously, coming out many moons after DoA2 DC released? You said that today's PS2 standards made DoA2 DC look pretty dated, so maybe you were talking about the rather unremarkable fighter Tekken 4 with its two fighter on-screen limit and single-tiered environs, also coming out many moons after DoA2 DC?

Bowie:
Actually, this isn't really true. If the game is running at 30 fps, then it is outputed at full resolution. At 60 fps, if flicker-filtered, then it is similar to 640x240 with 2xAA.
Correct, but I meant in the sense of the actual pixel area output. It's a visible improvement, nonetheless.
 
All i do know is that Soul Calibur 2 looks much better than Soul Calibur 1. Yes! With all the texture variety and clean image intact. ;)
 
Dated compared to which PS2 fighting game?
I wasn't thinking about fighting games specifically, but even there, TTT and T4 are definitely more accomplished technically. Not to mention the upcoming VF4:Evo and SC2 (that both support progressive scan). DOA2 (PS2 or DC version, doesn't matter) just doesn't look that hot anymore, regardless of the genre.
 
Progressive/Interlaced isn't really a 'wacky architecture'..

All the DC/PS2 discussions seem to have moved away from the topic.
VGA support is more a marketing decision - Sega decided that VGA was important, as their architecture was aimed at home and Arcade. Sony's was aimed squarely at the home, so VGA output wasn't a priority.

If Sony would have offered the VGA output as an option at launch a lot of games would have implemented it straight away - There's no technical issues, just the perceived market for the product. Your arguments remind me of the PC / console comparisions where PC owners would deride the 'mere' 640x480 compared to their 1280x1024, and the 'mere' 60Hz refresh... If you aim for the TV set market then you support the limitations of that market.

F355 is a very good looking game - and was very comparable to the early shots and demos of GT2000, but the final GT3 is better. In the same way GT3 is still comparable to the best on the Xbox or cube.

You always go on about how competitive a N2 based machine would be against PS2, but you fail to realise that Sega lost money making the DC. If they were unable to be cost effective with the DC why would they be effective with a much more expensive machine.

Wacky architectures - they're all around .... programmable pixel combiners... 16 stage TEVs , split L2 caches..... ;)
 
Burnout 2 is on the cube...

but it doesn't seem to have been targetted as much as expected...
BOSS going down didnt help much though...
 
"I'm sure someone going in to see Star Wars Episode I in the theatre on a nice screen and then seeing Star Wars Episode II on a slightly less impressive projection screen could easily have the more fulfilling visual experience from Star Wars Episode I, depsite the fact that special FX in the sequel were beyond that of its predecessor. The projection of the picture isn't by any means insignificant, especially when we're talking about differences as small as two consoles of the same generation."

I think the toy story comparison is much better. When moving from EPI to EPII, the humans remain with the same details, and so do many of the enviroments, I think. Sure, you might have more SFX in epii compared to epi , but it's like taking two games that are equal in gphx(aka real life.), and making one of them flashier.
 
Mortal Kombat looks hot on PS2 too! :oops: Lili or whatshername has more bouncy tits than any DOA fighter. 8)
 
Heck, it drives me nuts even when I play beautiful Xbox games like PDO and Halo at my friend's on his TV as I struggle to make out the little details in the graphics that my mind subconciously expects should be there (from the visual training its recieved during the times I played those games on a HDTV/monitor.)


ENJOY THE FRICKING GAMES... I play them through the VCR still I can appreciate them...

And I am sorry, I'll take modern PS2 games with AA ( they have it ) and much higher polygon count and lighting than the DC could ever dream of at such a fluid refresh-rate... and I will enjoy them for what they are...

I can and do play my DC games, I have a compact but very nice library IMHO ( Sonic Adventure 1, Shenmue I [I have Shenmue II Xbox], Sega Rally 2, Crazy Taxi 1, Sega Smash Pack I [emulation is so-so, but there are 12 games] ... I have played Soul Calibur and RE: Code Veronica and others but never bought them ) and I MANAGE to enjoy it on this regular TV ( connection through the VCR )...

And while I ENJOY those games I have no problem conceeding that a game like SH3 or Z.O.E. 2 look better than what Sega DC could have pushed out... pro-scan or not...
 
marconelly!:
TTT and T4 are definitely more accomplished technically.
TTT PS2 definitely more accomplished than DoA2 DC technically? How do you figure?

You know, I think some of you just believe a PS2 game is more advanced out of reflex. Somewhere in your minds, you've probably convinced yourself it's pushing substantially more geometry regardless of how you actually judge the visuals on screen.

Someone here already mentioned that TTT was pushing 3 million polygons per second. Well, Tecmo released a spec sheet for their Dead or Alive 2 engine on Dreamcast when they announced the home versions of the game. Among the features listed like the specifics on lighting and framerate was that the game was pushing over 3 million polygons per second too. And if you've seen some of the absolutely massive environments, like the three-tiered castle, or the fact that you can tag opponents in for more than two characters on-screen at once, you'd see that the geometry levels are quite high. Despite the fact that any reasonable person could tell you TTT wasn't high poly compared to DoA2 upon sight, now you have confirmation that both games are similar in polygon counts.

Textures? Dead or Alive 2, no contest whatsoever.

Image quality? Well, both are quite nice (the US version of TTT anyway), with TTT having that smooth Namco look. But, DoA2 really stands in a class by itself, above even almost any other Dreamcast game in IQ. Problems that plague other games like texture shimmering, alaising, or blending issues are like non-existent in this game. It's one of the sharpest, clearest, and cleanest things I've ever seen barring FSAA, and it runs in progressive scan at 60fps through VGA on Dreamcast too. TTT... interlaced.

Load times are completely eliminated in DoA2 after you select your character. One match flows right into the next, with new fighters and environments, with no loading pause whatsoever. If that's not an astonishing technical programming achievement for a fighter looking that nice, I don't know what is. TTT... load times.

But somehow, you've convinced yourself that TTT PS2 is "definitely" more technically accomplished, so I'm going to have guess you were blown away by the lighting or something.
 
You are truly hopeless. I don´t know where you get that DOA2 has better looking textures than TTT. If anything, they´re pretty much equal. Are you sure you are not confusing texture quality with art direction? Because if that were the case, I´d agree with you when saying that DOA2´s textures looked better, but only from that viewpoint.

I also would love to see that spec sheet, seeing that DC wasn´t capable of more than 3 mpps (that is coming from Sega´s own mouth), and I very much doubt that a near to launch game would max out what the system had to offer. I am also the one that brought up that 3 mpps number, and I would have to remind you that SC1 used 500k pps. Seeing that many people couldn´t figure out which fighting game looked better, I do not think DOA2 used a substancially higher number of polygons, especially considering the timeframe.

I also did not have a problem with TTT´s 2 or 3 second loadings before fights. DOA2 "suffered" from the same flaw, although it is evident which developer knew how to hide them in a better way.

I do not see the problem with the interlaced output, you make it out to be a huge advantage over PS2. Details remain the same in whichever output, and as someone else pointed out, DC´s graphical flaws are only maximized when displaying it in VGA form. It is undeniable how substancially better a game can look when applying reflections, a substancially higher polygon count, lightning and so on and so forth.
 
t is undeniable how substancially better a game can look when applying reflections, a substancially higher polygon count, lightning and so on and so forth.

You know it, Marconelly knows it, Vince knows it, I know it and most people know it...
 
There are 50-60 million PS2 owner in the World today. How many of those play games through VGA or HDTV? 1000?

I appreciate interesting tech discussions, but all this talk about DC's resolution in VGA mode vs. PS2 on a TV screen, or various prog. scan comparisons is, in my humble opinion, an utter waste of space. I am sorry Lazy8 - but you have argued your point, and important as it may be to you, I assure you: It really makes ZERO difference to the world around you.

Who gives a **** about VGA/progressive scan this gen?

Perhaps it would be a good idea to stick to how games actually look on TV screens - not in some obscure mode used by noone...
 
How do you figure?

tekkentag_screen006.jpg


vs

doa2_screen035.jpg


and

tekkentag_screen011.jpg


vs

doa2_screen029.jpg



That's how.
 
Almasy, you are incorrect in stating the 3 million pps number. The Dreamcast can do more than that, take this as an official confirmation. Go play Test Drive Le Mans on the Dreamcast and you'll be able to find the proof you need. The DC is more than capable of pushing 4 million pps.

The SH4 was rated at 5 million pps itself, but 6 million has been gotten. If you do a little research for yourself you can see it.

Textures in DOA2 for the Dreamcast are better than TTT on PS2. But Tekken 4 blows DOA2 on DC out of the water.
 
I was aware of that, but to ge to those levels quite a bit of digging and optimizing would be necessary, not to mention experience with the hardware, wouldn´t you agree? I was just stating what Sega originally stated, and I can´t picture a developer getting 3 mpps right away. That was all I wanted to say.:p

Although I can´t honestly see in what way textures are better in DOA2 are, although I could be wrong. Maybe they´re, maybe they are not, but I think we can see which one turned out to be the better looking game, thanks to the screenshots marconelly posted.
 
DOA2 is pretty cool, but the models show their age, and definitely dont compare to the TTT models. ( Personally I think it was technically one of the best DC games ever, much more impressive on that level than Soul Caliber ) I actually prefer playing it ( on DC or PS2 ) to TTT, as I was never a fan of the infinite plane floor.
TTT was a launch title though, and a conversion of a PSX based arcade title as well, so the comparision is a bit stilted.
In a way I always had high expectations from Tecmo, as their Saturn and PSX DOA implementations were both excellent.
A fairer comparision would be TTT to VF3tb, or if you really want to be cruel compare T4 to Fighting Vipers 2....
 
Almasy:
I don´t know where you get that DOA2 has better looking textures than TTT.
The rocks and vines on Tengu's stage. The whole castle. The opera house. The stain-glass window. The waterfall stage. All high-res, no mip banding, no bad shimmering, all looking very solid and defined. The environments are in a class by themselves. TTT's aren't bad by an means, though.
I also would love to see that spec sheet, seeing that DC wasn´t capable of more than 3 mpps (that is coming from Sega´s own mouth),
That didn't come from SEGA's own mouth. What SEGA said of Dreamcast when they first announced it was "can push more than 3 million polygons per second." They weren't releasing a performance spec sheet, they were simply conveying a realistic figure that people could expect in-game. I know that's almost inconceivable in this day and age of hyping theoretical numbers and all, but you'll notice Nintendo also conveyed a realistic figure for GameCube instead of the theoretical numbers when they claimed GameCube at 6-12 mpps.

Anyway, before the DC ever even launched in Japan, president Irimajiri (sp?) of SEGA said that Suzuki's team was already exceeding 3.5 million polygons per second in their initial projects. Another dev that gave an indication of performance numbers was Melbourne House. They revealed their Test Drive Le Mans for the Dreamcast was pushing around 5 million polys per second when they discussed it at the DC Technical Pages message board. The developer also said they were considering use of a field-rendering technique for any future Dreamcast projects to allow them more free memory for storing polygon display lists. They said this would allow them to push even more geometry per second, above the approx 5 mpps they were already getting.

Anyway, if you want to see documentation about performance specifics for DC from some SEGA engineers, look at the following provided link. There's a pretty fascinating analysis covered over two pages, and here's the reference for a more theoretical limit on total system performance:

http://www.computer.org/micro/articles/dreamcast_2.htm

"Recent tests by Sega, using the company's graphics libraries proved that Dreamcast uses 100% of the CPU and rendering engine's ability and delivers 6 million textured and lit polygons per second."
I also did not have a problem with TTT´s 2 or 3 second loadings before fights.
That's nice. DoA2 was perfectly seamless though. Perfectly seamless wins every time.
DOA2 "suffered" from the same flaw, although it is evident which developer knew how to hide them in a better way.
There is no flaw. DoA2 didn't pause for a loading screen. You were simply taken from one gorgeous stage to the next, with a new challenger. The fact that seamless games like Jak and Daxter and DoA2 can retrieve new information (of course they load, it's not like they're pulling info out of thin air) without pausing everything for a loading break is what makes the accomplishment impressive.
I do not see the problem with the interlaced output, you make it out to be a huge advantage over PS2.
Progressive output - twice the pixel output every update. Two fold. Makes a difference to the home theatre enthusiasts of the world.
Details remain the same in whichever output, and as someone else pointed out, DC´s graphical flaws are only maximized when displaying it in VGA form.
Not totally. The analogy is like me punching you in the stomach if you were complaining about your headache. Sure, you'll stop worrying about your head pain as you crumple over clutching your stomach in agony, but it's just using one flaw to cover up another flaw. The games have a lot of graphics defects and artifacts, but TV's blurriness and slower fade just apply another defect on top making you notice the first problem less.
It is undeniable how substancially better a game can look when applying reflections, a substancially higher polygon count, lightning and so on and so forth.
And TTT, an early game reprogrammed specifically for the PS2, has these over DoA2, an early game programmed sepcifically for the DC? Reflections, substantially higher polygon count, lighting, etc.?

DoA2 actually has the best reflections... remember the opera stage where the glossy floor is reflecting the ceiling? Both games are pushing similar geometry counts. The lighting was quite nice in DoA2, and TTT was roughly similar too. Anything else I'm not aware of to offset the load time issue, the progressive issue, etc?
 
Oh no, it's the age old "let's post screenshots" :rolleyes:

Better yet, let's post screenshots of the DOA2 stages that (intentionally) don't use lots of textures. :LOL:

Dude have you even played some of the texture heavy stages of DOA2 through a VGA connection?

I also think if Namco had continually evolved the SC engine until today for DC, we would see a very nice upgrade from SC to SCII on DC. Anybody know how many polys SC II is pushing?

Thanks for that link Lazy 8)
 
Back
Top