PSP: Playable on show floor, impressions inside.

[Lazy8s]

You cited that there should be equal terms. The DC has a better implementation of things like z-buffer precision and mip-mapping than PS2, so it would only be fair.

 

[Phil]

Of course it would be fair - I wasn't excluding it. I just named examples of games that have up-to-par if not superior image quality, so the comparasment should be no problem.

 

[Crazyace]

Phil:

Since this was about DC vs PS2, we should actually compare the two under same conditions.

If you truly want them under the same conditions, the PS2 is going to have to pick up its slack in IQ as well.

It is always going to be difficult to compare 2 different systems - DC has VGA support, whereas PS2 is mainly aimed at standard TV , why not argue about somethings else thats pointless like DC not having optical out, so it's sound is inferior.

archie4oz:

If you want any CPU left over to do anything they are...

Most PS2 games were tested to be between 2 and 5 mpps. The CPU resources weren't lacking for AI and game logic even in a comparably complex DC game like Test Drive Le Mans.

I am sure that complex graphical DC games suffer from a drain on CPU resources by the T&L process.. Some PS1 driving games showed good AI and game logic after all..

Fafalada:

Except that it wasn't just that.

It was a native effect, so the associated set-up costs had been considered when it was chosen for hardware support. Bump mapping is an effect you'd only want to use selectively anyway, and it's not like the DC had zero polygons to spare. For 640x480 console games, there should be acceptable balances that could be achieved from the trade-off of bump mapping.

Once you start thinking realistically about the use of effects, the costs balance out - the DC will have the dot3 as a cheap second texture pass, but the setup cost takes time, the PS2 render cost may be more expensive in terms of passes, but the setup cost is not as much of a problem.

 

[Lazy8s]

Which PS2 games demonstrate superior image quality? Remember that the games which used FSAA via supersampling on PS2 traded off color depth, full height buffers and the ability for proscan support.

 

[Lazy8s]

Crazyace:

It is always going to be difficult to compare 2 different systems - DC has VGA support, whereas PS2 is mainly aimed at standard TV , why not argue about somethings else thats pointless like DC not having optical out, so it's sound is inferior.

I didn't specify VGA. And like optical audio out (which gives better sound output but not "sound" in general - DC's sound chip was more robust for audio generation), there are some incomparable console specific features that will still make a difference and be appreciated by its users.

 

[Fafalada]

Remember that the games which used FSAA via supersampling on PS2 traded off color depth, full height buffers and the ability for proscan support.

Considering that 99% of DCs games traded off color depth, you shouldn't even begin to consider that as an argument.

 

[Fox5]

Crazyace:

It is always going to be difficult to compare 2 different systems - DC has VGA support, whereas PS2 is mainly aimed at standard TV , why not argue about somethings else thats pointless like DC not having optical out, so it's sound is inferior.

I didn't specify VGA. And like optical audio out (which gives better sound output but not "sound" in general - DC's sound chip was more robust for audio generation), there are some incomparable console specific features that will still make a difference and be appreciated by its users.

Wasn't the DC's sound chip just a genesis cpu? Anyhow, the PS2 has a lot of raw cpu power it could use for sound, the n64 did it, and I actually think some n64 games had better quality sound than dc.(shenmue and sonic adventure 2 had very cracky sound for some reason, plus not a single dreamcast game supported even pro logic, or at the very least dc sound quality was about equal to psx and n64)

 

[DeanoC]

IIRC DC sound chip had an ARM in it.... Would have remember a 68000 (where I learnt to program many moons ago...).

As for which sound system was more powerful, remember a PS2 has most of the CPU power of a 36Mhz R3000 for its own devious needs.

 

[Crazyace]

I didn't specify VGA. And like optical audio out (which gives better sound output but not "sound" in general - DC's sound chip was more robust for audio generation), there are some incomparable console specific features that will still make a difference and be appreciated by its users.

Not really, as optical out supports multichannel discrete systems such as DTS
or Dolby digital. - and I think you might need to really reconsider the 'robust' comment.

But in terms of image quality I have DC games with terrible IQ (Seventh cross for example ), most noticably showing extremely visible mipmap lines and terrible Z fighting ( Sonic was guilty of that on Eggmans carrier. )

 

[max-pain]

Wasn't the DC's sound chip just a genesis cpu?

That was the Saturn (68000).

 

[Lazy8s]

Fox5:

.(shenmue and sonic adventure 2 had very cracky sound for some reason,

Shenmue didn't stream music off of the disc since it was too busy streaming data for the game. It was the sound chip which did all of the mixing on a multitude of sources that included lots of ambient layers and even two songs simultaneously (background music plus the music being played by the shops on the streets of Dobuita). The voices were heavily compressed since the game had to fit dozens of hours of spoken dialogue (not an exaggeration) - it was what caused Shenmue to be spread across multiple GDs in the first place.

Never heard anything wrong in Sonic.

 

[archie4oz]

Most PS2 games were tested to be between 2 and 5 mpps. The CPU resources weren't lacking for AI and game logic even in a comparably complex DC game like Test Drive Le Mans.

Does most include the plethora of 2D titles (which the PS2 easily dwarfs the DC library, even if you include the 'fan-service' content still being produced for it)? If you're gonna use a DC showcase vs. a PS2 average, why not use a PS2 showcase (perhaps like GPC by the *same* studio!)? Also, what are you looking for? Actual renderable polygons? (i.e. discounting rejected/culled polys, zero area polys, multi-pass generated polys)

You cited that there should be equal terms. The DC has a better implementation of things like z-buffer precision and mip-mapping than PS2, so it would only be fair.

You mean the DC secretely supports doubles or something for it's Z-buffer? What's next!? Are we going to find out about a pixel shader that Video Logic somehow forgot to document?

EDIT: Or are you referring to w-buffer support?

IIRC DC sound chip had an ARM in it.... Would have remember a 68000 (where I learnt to program many moons ago...).

45MHz ARM7 vs. 37MHz MIPS3K I'd consider the two comparable (I got Tremor to decode vorbis streams on both (eating up pretty much all the resources on both) so I'd say that's good enough for me)...

 

[Lazy8s]

archie4oz:

If you're gonna use a DC showcase vs. a PS2 average, why not use a PS2 showcase (perhaps like GPC by the *same* studio!)?

You figured DC was out of PS2's league for geometry, so I referenced a DC example that showed overlap into PS2's typical range. Maybe you have a different definition for 'league', but I would expect no overlap with the majority range at all as a prerequisite for such a term... at the very least.

You mean the DC secretely supports doubles or something for it's Z-buffer?

No, I meant that it sustains 32-bit precision all of the time.

 

[Squeak]

IIRC DC sound chip had an ARM in it.... Would have remember a 68000 (where I learnt to program many moons ago...).

45MHz ARM7 vs. 37MHz MIPS3K I'd consider the two comparable

Doesn’t the M3k have IO duties to look after that the ARM7 doesn't?

(I got Tremor to decode vorbis streams on both (eating up pretty much all the resources on both) so I'd say that's good enough for me)...

How much power does it take to real time encode DTS or 5.1?

 

[archie4oz]

You figured DC was out of PS2's league for geometry, so I referenced a DC example that showed overlap into PS2's typical range. Maybe you have a different definition for 'league', but I would expect no overlap with the majority range at all as a prerequisite for such a term... at the very least.

Because it isn't... Nor would I consider the PSOne in the same class geometry throughput wise to the DC even though we managed to squeeze enough out of it for Tobal 2 to push it into the realm of quite a few low-end DC titles... My point was rather simple in that it's pretty easy to push geometry two to thee times of what would peg the CPU on the DC, on the PS2 with relatively low CPU utilization...

No, I meant that it sustains 32-bit precision all of the time.

I don't see how locking the z-buffer precision makes it a "superior" implementation... I'd say on-chip z-sorting makes it a better implementation...

Doesn’t the M3k have IO duties to look after that the ARM7 doesn't?

The IOP does, however that also includes more hardware than just the R3K core... In any case they're close enough computationally wise and feature-set wise that I consider them fairly even...

How much power does it take to real time encode DTS or 5.1?

2-6% CPU depending on the # of channels you're encoding. Dunno about VU0 utilization since I never profiled it or anything, but it didn't appear to burn up too much of it... However since you're essentially DMA'ing audio in chunks of 512 samples all over the system with each call to the DTS lib, and thread priority for it is pretty high and they didn't like you interrupting it to use VU0 (caused all sorts of problems)... However they may have improved/fixed it since and I imagine since it's a pretty constant routine it'd be easy to schedule around it assuming it still doesn't want to hog VU0's state... However I never did anything extensive with it, and that was a long time ago that I played with an evaluation of the SDK... (like when it came out, and we were pretty tight with EA)

 

[PC-Engine]

FWIW, I've compared DC SC to GCN SCII in progressive scan on my VGA monitor and DC has better texture detail and overall image quality. GCN textures are pretty blurry and overall sharpness of the output is not as sharp as DC. Don't know how SCII on PS2 compares as I don't have a PS2. I picked SC because it's a good sample for comparison.

 

[Ty]

Look I have never gone gagga over the aesthetic design of any system and I'm not about to start now. Its not something I care to much about, never have. So please allow that to sink into your head and accept it

Fair enough but you certainly have commented favorably about how the GCN looks so obviously it matters somewhat to you. How important a handheld looks to you is really a personal question for sure though I think it matters more to most people than yourself (witness Ipod and how "cool" it is).

I agree, and this is one of the ways that the DS looks better in reality then those shots we saw (when I say reality I mean in someone's hand in a video... a lot closer to reality then a pic). Its suprising how different it looks to those images infact, much smaller and a lot less plastic-like.

I'm not sure I understand how a big closeup picture is less realistic than an amateur video? If anything video is generally less accurate because it has less resolution in addition to be farther from the source. Therefore the video is very likely LESS indicative of the DS than the screenshots Nintendo provided. Or would you expect Nintendo to provide less than stellar screen shots to the press?

I just posted the PSP pic to make a point. That you shouldn't assume that something can't look a lot better in real life then it does in a pic.

But you also need to consider the fact that the DS images are from Nintendo themselves. If THEY can't make them look good, who can? I mean, c'mon, those are some of the most expensive shots (in marketing dollars) they're likely to spend for sometime on the DS.

I'm not giving the DS the benefit of the doubt because I haven't seen it in person. I am saying that I have seen the DS in video's in use and it looks much better then it does in the images we saw.

I'm wondering if there is an analogy of TVs outputting "better" images because of their "built-in" AA with regard to my point about most video cameras having less accuracy (both in color and resolution) than most digital still cameras. Have a handy link to the video you're speaking of btw?

Were as nobody is telling me that they have seen the PSP keyboard addon in video's or in person and that it looks much better in reality then in those pics. Some are just saying that they think the keyboard looks sexy from seeing the same pics I've seen. Of course that keyboard addon could possibly look better in realife, I have no problem with that. But, just like with DS until I actually saw it in video's, I'm not going to assume it looks good.

Ok, I have and it's WAY better than those lame pics that still blow the DS out of the water.

P.S. Charles prefering Camilla to Diana is a mystery to anyone appart from Charles himself, nothing to do with being British.

Heh, I just had to get that off of my chest. The dude MUST be blind.

P.P.S aesthetically PSP looks better then DS from what I've seen.

Yay! Shall I place a preorder in your name?

 

[Farid]

As for never knowing Sony to make flimsy expensive stuff. Why have so many people had to buy replacement PS2's? Why did so many people have to do the same with PS1?

Sorry to interrupt you guys while you're practicing your Quote-Fu.

But about this "playstations are not solid" discussion, i have to say that the attach rate of each consoles shows that playstations are not specially doing worse than the others (actually they're first on the attach rate figs, Nintendo shown the attach rate of the consoles at E3 this year).
Except if Playstation owners buy really more games than Xbox/GC/N64/DC/Sat owner, this "playstations are not solid" discussion sounds like an "urban legend" to me.

There's an explanation possible for this discussion, since there's more PS sold, there's more of them that going to after-sales services, but proportionally it shouldn't be THAT different from the competition.

 

[PC-Engine]

I'm not Teasy, but I just wanted to mention that IMO, video footage gives a better indication of how the DS really looks because you can see it from different angles dynamically. Still shots bring out the minute details, but it doesn't give a good indication of overall dimensions unless you have something placed right next to (not behind) it for reference. With video you have distance perception.

Also IMO iPOD looks fugly. It looks like a soap bar with a bigass dial. Looks like some funky Star Trek communication device.

 

[cthellis42]

Hey! No cheap shots at the iPod or I'll CUTCHA! <brandishes a pair of wire strippers>