Support NeoGAF

[Ishida]

Pretty much everything you've written here is complete nonsense.

Yeah... Same thing I noticed. WTF.

 

[makingmusic476]

N64 games always looked smoother to me. Like they have less aliasing or run at a higher rez or something. Everything looks smoother and less pixelated. The shots I've seen in here back that up.

I prefer that smoothness even if some of the textures in N64 games have less detail or other minor things are missing.

 

[qualitydisc]

Pretty much everything you've written here is complete nonsense.

Can't tell if it's ESL or nonsense

 

[HTupolev]

Can't tell if it's ESL or nonsense

It's a bizarre attempt (by someone who likely doesn't understand the particulars) to articulate the distinction between PSX's non-perspective-correct nearest-neighbor and N64's mipmapped trilinear filtering, combined with a general preference of N64 performance over PSX performance.

 

[RedSwirl]

Btw, talking about Turok, Shadow of Oblivion was a looker. First game I played with lip-sync in real-time cutscenes.
http://www.youtube.com/watch?v=bZ_4NKSraaQ&feature=player_detailpage#t=308s
http://www.youtube.com/watch?v=A-UvK3zxfNE&feature=player_detailpage#t=122s

Damn I remember this. More proof of what companies like Rare and Acclaim could pull off, especially in the N64's later times.

I was surprised when I downloaded this for my Vita over the weekend when it said was only a 90 something mb file. Thought something had gone wrong somewhere lol.

Vagrant Story was one of the few PS1 RPGs with only one FMV at the beginning and no voice acting.

 

[Beerman462]

Not a chance. I was using high-end PC equipment in 99 and there wasn't a single game that came remotely close to matching Daytona or anything Model 3.

I had a Voodoo 3 in 1999. I bought NFS Porsche unleashed in March 2000.

Daytona

NFS

Pretty decent

 

[Ric Flair]

NFS Porsche looks like an early ps2 gen game, really impressive

Also the Mission Impossible comparison is interesting, if I remember correctly Duke Nukem 64 had reflections for the mirrors, I wonder if it was just something the dev team didnt have time to implement or whether it was simply not worth the effort

 

[rjc571]

NFS Porsche looks like an early ps2 gen game, really impressive

That looks nothing like an early PS2 game whatsoever. It looks like a PS1 game ported to PC with minimal improvements outside of resolution. Which is exactly what it was.

I'd probably take Quake 3 running on 1999 PC hardware at 1024x768 and 80 fps over Daytona or anything else on Model 3, though.

 

[Beerman462]

Pretty much everything you've written here is complete nonsense.

I'll translate. Shadow Man ran like shit on PSX and with less effects. PSX was not powerful enough to run Turok 2. N64 used texture maps that would be wrapped around a model. On PSX each polygon had to be textured individually.

I'm not backing his claims, but that's what I took the post to mean.

 

[Sipheren]

These are the 2 games I played a lot back in the day. It's crazy looking back at them not just how bad they were...

World Driver Championship - N64

Gran Turismo - PSX

 

[LifeIsLife]

The big thing the N64 struggled with (compared with the PS1) is the number of polygons, I believe. I don't think I ever read about the N64 having trouble with particles, especially from game experience.




The PS1 was incapable of providing perspective correction for the 3D graphics it showed on screen. So whenever the camera moved, you could see all the polygons warp and change form.

Not sure what the original asker was actually going for but the issue's with textures and/or polgons on the PS1 was due to the fact it could not do accurate floating point calculations quickly so everything had to be "estimated" and stuff would not line up properly or would twitch due to the "estimation" constantly changing

 

[HTupolev]

That looks nothing like an early PS2 game whatsoever. It looks like a PS1 game ported to PC with minimal improvements outside of resolution. Which is exactly what it was.

I'd probably take Quake 3 running on 1999 PC hardware at 1024x768 and 80 fps over Daytona or anything else on Model 3, though.

That would be one hell of a 1999 PC.

Not sure what the original asker was actually going for but the issue's with textures and/or polgons on the PS1 was due to the fact it could not do accurate floating point calculations quickly so everything had to be "estimated" and stuff would not line up properly or would twitch due to the "estimation" constantly changing

There are two problems.

The aliasing and graininess with the textures was caused by the use of plain nearest-neighbor "interpolation"; there was no mipmapping or linear interpolation between texels.

The wobbliness was caused by the lack of perspective correction. It wasn't so much that the estimation was constantly changing, it's that the "estimation" practically didn't exist. Texture coordinates were simply linearly interpolated in screen-space across the surface of a polygon, which is horribly inaccurate.

 

[Ric Flair]

That looks nothing like an early PS2 game whatsoever. It looks like a PS1 game ported to PC with minimal improvements outside of resolution. Which is exactly what it was.

I'd probably take Quake 3 running on 1999 PC hardware at 1024x768 and 80 fps over Daytona or anything else on Model 3, though.

I realize it's a different genre, but not every launch PS2 title looked like Ridge Racer V

 

[A Black Falcon]

Oh please.

It's not "lazy developers" why the games look worse. The N64 was also a bitch to program for and contained several bottlenecks (shitty latency, limited texture cache, tiny storage capacity). The micro code or whatever that made N64 games run better was suppose to be off limits to developers, hence why only Rare, Factor 5, Nintendo took advantage of it.

If Nintendo designed the N64 to be more third party friendly *cough* *cough* I think we could have seen better results all around.

Edit:Correction about the microcode. I meant Turbo3d was off limits, developers used the poorer fast3d mode instead (with some writing custom to that).

This is ridiculous! Turbo3d is worse than Fast3d in every possible way. Yes, it has a significantly higher polygon-count, but with no perspective correction or Z-buffering, everything would look PS1-inaccurate but with more polygons, and that is NOT what anyone should want from their 3d; it's a good thing when polygons stay where they're supposed to be and you aren't seeing popping seams all the time! Nintendo did the right thing to not allow anyone to use Fast3d.

Well, the reason for using many small triangles as opposed to fewer, larger ones aren't issues with seams... It's issues with affine texture mapping that will result in textures on entire polygons looking warped in perspective space (and even worse when clipping triangles against near/far planes).

Subdividing geometry gives a more correct image as the area of affine texture mapping gets smaller before being corrected by a projected vertex.

Ah, yes... I'm sure you're right, of course. But I know that developers did use overlapping polygons on the PS1 in order to cover up the popping polygon seams, and I think you see some of that in those polygon meshes. So that's two things that forced PS1 polygon counts to be significantly inflated, hiding seams and reducing texture warping.

Compared with Arioco's post something is amiss about this comparison. Given Many of the games it looks like the PS pushed more polys per second and despite the trilinear filtering on the N64, most textures seem better.

Given how powerful the N64's processor was I imagine this was an example of smarter design winning out over brute force. In addition to the cartridge format, I am guessing the unified Rdram of the N64 probably is the cause of alot of the issues because the devs had to deal with latency. Despite the core of the PS main chip being slower, the graphics and mdec engine had their own separate performance rating and the separate GPU and SPU also had their own discrete pool of ram. Chances are the sum of all parts is what made alot of games simply look and perform better on the PS. Granted this is also due to developer ability but man, both systems had some good looking games which makes it a bit difficult to compare art design vs technical achievement without some info on the bts performance. So again, spec wise the N64 should have danced circles around the PS1 but the real world results (sans mem expansion) made it seem as if there were some bottlenecks in N64'd design that prevented it from being the unquestionable king of the hill.

examples:
http://www.lensoftruth.com/retro-head2head-resident-evil-2/

I'll never understand how PS1 fanboys like you clearly are manage to convince themselves that perspective correction, Z-buffering, anti-aliasing, and all those things don't matter. Because what you're missing is that the N64 uses a LOT of hardware power to keep polygons where they should be, while the PS1 uses none because it can't do that. If Fast3d -- that is, the N64 microcode with none of those features -- had been allowed, N64 games would look as inaccurate as PS1 games, with texture warping and polygon popping galore, but polygon counts would be far above where they are on the PS1. Nintendo chose to require better-quality graphics instead, and it was one of the better moves they did with the system, I would say. Games look and perform better on the N64 than the PS1 because the 3d actually looks like 3d graphics, and not like a warping popping mess as it does on PS1. Just look at how many polygons developers had to waste to try to cover up the worst of the effects of those problems! And that couldn't deal with it all.

As for the RAM, RDRAM has higher latency, but very fast access during each read. It's got just as many plusses as it does minuses, and was VERY fast for the time. This is why the first revi8sions of Pentium 4 motherboards in the early 2000s require RDRAM, and why the Playstation 2 uses RDRAM for its RAM as well. Eventually DRAM got better and exceeded RDRAM, but the idea that RDRAM's slower access times actually was an overall drawback is false; in fact, it's very good RAM for the time, some of the best. It just works slightly differently.

As for textures, the N64's limited texture cache is its one real design flaw. A few of the best developers figured out how to work around this and produced amazing-looking things, but most studios couldn't match Nintendo, Rare, or Factor 5... (Some people would also complain about the lack of a dedicated sound chip in the N64, but I think N64 audio was good enough.) The idea that texture quality matters more than overall image quality, as PS1 fans always seem to insist, is wrong on any objective level. It matters, yes... but PS1 image quality matters more.

Nintendo didn't use custom microcode internally they used and optimized the same thing they and SGI had developed and made available to 3rd parties. The RCP was also fully open to 3rd parties which is why you saw more capable developers like Rareware, Boss Game Studios, Leftfield Productions or Factor 5 doing it for their games. Nothing was off limits.

Yeah, this is misleading. While it is true that a few developers were given the ability to do their own microcode, Nintendo had to really be pushed to allow it, and very, very few third-party studios were ever given the information necessary to do their own microcode. Looking at Boss Games and Factor 5's games shows what you can do with your own microcode, but not many others managed to convince Nintendo to let them try.

That looks nothing like an early PS2 game whatsoever. It looks like a PS1 game ported to PC with minimal improvements outside of resolution. Which is exactly what it was.

I'd probably take Quake 3 running on 1999 PC hardware at 1024x768 and 80 fps over Daytona or anything else on Model 3, though.

You've never played NFS Porche Unleashed on PC, have you. No, it's not a PS1 port, it's a PC-exclusive game. The PS1 "NFS Porsche Unleashed" game is from a different developer and is a different, much simpler game; the PC game is more complex and has more sim elements than most NFS titles. All track layouts are different between the two games, the PC game has more cars, different handling, etc etc. PC NFSPU is still an arcade racing game, but it's got some depth, and great graphics too just like the PC version of NFS High Stakes. It's a great game, like all of the late '90s PC NFS games. The PC versions of NFSHS and NFSPU look fantastic for the time!

 

[Easy_D]

I thought the PC game was called Porsche 2000 and that it came out in 2000.
Edit: I'm a very confused fellow

 

[HTupolev]

The idea that texture quality matters more than overall image quality, as PS1 fans always seem to insist, is wrong on any objective level.

It's not really debatable that N64 games tend to be more visually stable than PSX games (at least in terms of how crap is sampled). It's also not really debatable that PSX games can store (far) more texture data.

But the extent to which it matters is a question of how much a person values it, which is intrinsically subjective.

 

[Uncle Rupee]

Is there even a single character model on the N64 that has as many polygons as this dude from FF7? Please post example pics if you can find them.

Look at the musculature, individual fingers, individual teeth, modeled chain-links, and other details on these models. Nothing on the N64 comes close to that.
As we've seen in the wireframe comparisons of the kart racers, the PS1 seems to push way more polygons than the N64.

That thing looks like it's not using any textures (it's all vertex colors).

Exactly. This means that certain details can't be faked, and actually require MORE polygons, such as the pupils and wrinkles for the eyes, nostril, and rows of teeth.
But given the N64's dismal texture memory, textures wouldn't really improve these models. I'm talking about raw polygon processing.

 

[Branduil]

Compared with Arioco's post something is amiss about this comparison. Given Many of the games it looks like the PS pushed more polys per second and despite the trilinear filtering on the N64, most textures seem better.

Given how powerful the N64's processor was I imagine this was an example of smarter design winning out over brute force. In addition to the cartridge format, I am guessing the unified Rdram of the N64 probably is the cause of alot of the issues because the devs had to deal with latency. Despite the core of the PS main chip being slower, the graphics and mdec engine had their own separate performance rating and the separate GPU and SPU also had their own discrete pool of ram. Chances are the sum of all parts is what made alot of games simply look and perform better on the PS. Granted this is also due to developer ability but man, both systems had some good looking games which makes it a bit difficult to compare art design vs technical achievement without some info on the bts performance. So again, spec wise the N64 should have danced circles around the PS1 but the real world results (sans mem expansion) made it seem as if there were some bottlenecks in N64'd design that prevented it from being the unquestionable king of the hill.

examples:
http://www.lensoftruth.com/retro-head2head-resident-evil-2/

The main problem with N64's textures was the tiny texture cache. If they'd just given it a decent cache the system would blow away PS1 visually.

 

[JordanN]

Is there even a single character model on the N64 that has as many polygons as this dude from FF7? Please post example pics if you can find them.

As we've seen in the wireframe comparisons of the kart racers, the PS1 seems to push way more polygons than the N64.

That thing looks like it's not using any textures (it's all vertex colors).

 

[OpinionatedCyborg]

Pretty much everything you've written here is complete nonsense.

So that means this is the year 2014, I am reading a videogame messageboard, and Project 87, the plan to converge an aberrant universe with ours, was a success. Everything's normal - thanks for the confirmation.

 

[A Black Falcon]

It's not really debatable that N64 games tend to be more visually stable than PSX games (at least in terms of how crap is sampled). It's also not really debatable that PSX games can store (far) more texture data.

But the extent to which it matters is a question of how much a person values it, which is intrinsically subjective.

Objectively the N64 is significantly more powerful in every way other than texture quality. Saying that one thing matters more than everything else -- RAM, system power, polygon-pushing potential, polygon accuracy (warping, broken edges, etc.), and everything else is nonsense. Winning in one category and losing every other in-game graphical category is not a tie! That's not how it works.

Of course subjective opinion is different, but objective fact is that the N64 is the better, more feature-rich, and more powerful system.

 

[Uncle Rupee]

The main problem with N64's textures was the tiny texture cache. If they'd just given it a decent cache the system would blow away PS1 visually.

Except the cartridge memory would also limit the size/detail of the textures, so the N64's texture cache isn't the only bottleneck.

 

[encephalon]

I will never grow tired of this rivalry.

 

[dark10x]

I had a Voodoo 3 in 1999. I bought NFS Porsche unleashed in March 2000.

Daytona

NFS

Pretty decent

I had NFS as well. It's not even close. The amount of geometry (quads, in this case) being pushed in Daytona 2 far exceeds NFS on the PC. It also features superior textures and runs at a higher frame-rate. Those older NFS games don't run at a proper 60 fps (engine design limitation, I suppose).

In those days the PC wasn't particularly capable of throwing around large numbers of triangles. It simply couldn't handle the more geometrically dense games being released in the arcade or on consoles especially not at a constant 60 fps as we saw with Model 3 games. There was eventually a PC port of Sega Rally 2, of course, but it was cut down compared to the arcade release (much shorter view distance, less detailed cars, less detailed shading, missing lighting, etc). It looks fine at a glance but it's not comparable to the proper Model 3 version. Nor was the crappy Windows CE Dreamcast release.

 

[Branduil]

That would be one hell of a 1999 PC.


There are two problems.

The aliasing and graininess with the textures was caused by the use of plain nearest-neighbor "interpolation"; there was no mipmapping or linear interpolation between texels.

The wobbliness was caused by the lack of perspective correction. It wasn't so much that the estimation was constantly changing, it's that the "estimation" practically didn't exist. Texture coordinates were simply linearly interpolated in screen-space across the surface of a polygon, which is horribly inaccurate.

There was also a third problem which doesn't directly involve textures but which does indirectly affect them, which is that PS1 uses fixed point precision for vertices instead of floating point. This means that polygon vertices "snap" into pixels instead of moving smoothly. Combined with the lack of texture filtering or perspective correction, this creates the jello-like shaking effect common in PS1 games.

 

[nOoblet16]

I haven't seen one PS1 game that looks as good as Mario 64 and that's a lunch title

I thought Sheep, Dog n Wolf was the best looking PS1 game, it looked just like looney tunes !
https://www.youtube.com/watch?v=PLjz1XJuAN4

On a side note, why was this game never popular? It was hilarious, unique and had great music and gameplay.

 

[Zen]

40 Winks (ps1) looks on par to that.

http://youtu.be/6KykbYzzszY

 

[Bam Bam Baklava]

For the love of god stop calling it PSX.

 

[pixelbox]

http://cdn.akamai.steamstatic.com/s...53fcebebd544f4cdf6.1920x1080.jpg?t=1354222545

 

[Uncle Rupee]

I haven't seen one PS1 game that looks as good as Mario 64 and that's a lunch title

 

[foxuzamaki]

This thread's soul still burns

 

[baphomet]

For the love of god stop calling it PSX.

Considering it was called the psx for like 10 years before that was released, I think its going to stick.

 

[Branduil]

Is there even a single character model on the N64 that has as many polygons as this dude from FF7? Please post example pics if you can find them.

Look at the musculature, individual fingers, individual teeth, modeled chain-links, and other details on these models. Nothing on the N64 comes close to that.
As we've seen in the wireframe comparisons of the kart racers, the PS1 seems to push way more polygons than the N64.

Exactly. This means that certain details can't be faked, and actually require MORE polygons, such as the pupils and wrinkles for the eyes, nostril, and rows of teeth.
But given the N64's dismal texture memory, textures wouldn't really improve these models. I'm talking about raw polygon processing.

Well, no one should dispute that PS1 is better at raw polygon processing power. In certain instances, such as in closed rooms with a small amount of characters onscreen, this power could be put to very good use, and Square was masterful at doing just that.

The N64 was much stronger by comparison when it comes to wide-open spaces, where it can make fuller use of its filtering capabilities, and the PS1's advantages are mitigated by the need to waste tons of polygons just to keep the textures from swimming everywhere. So if Final Fantasy boss fights tend to showcase the best of the PS1, the best of the N64 would be stuff like Zelda's overworlds, the landscapes of Battle for Naboo, or Rareware games.

 

[Sipheren]

For the love of god stop calling it PSX.

It was the PSX first, they later used the name for that thing. If Sony didn't want it called the PSX they shouldn't have made the PS1 (PSOne) later on.

By the end of 1992, Sony and Nintendo reached a deal whereby the "Play Station" would still have a port for SNES games, but Nintendo would own the rights and receive the bulk of the profits from the games, and the SNES would continue to use the Sony-designed audio chip. However, Sony decided in early 1993 to begin reworking the "Play Station" concept to target a new generation of hardware and software. As part of this process the SNES cartridge port was dropped and the space between the names "Play Station" was removed becoming "PlayStation", thereby ending Nintendo's involvement with the project.[citation needed] The PlayStation was also known as the PS-X[13] during development. This led to the console sometimes being referred to as PSX upon its release, although this was not endorsed by Sony as the PSX name was intended only as a provisional codename. In 2003, an entirely different console with the name PSX was released in Japan.

Source

 

[HTupolev]

Objectively the N64 is significantly more powerful in every way other than texture quality.

Scale of contents in general, not just textures.

But you're acknowledged that there are some respects in which the N64 can't keep up. Respects which obviously make it impossible for the N64 to do certain things the PSX does. Which, if these things are extremely highly valued by someone, makes the PSX a more capable system.

Of course subjective opinion is different, but objective fact is that the N64 is the better, more feature-rich, and more powerful system.

The problem here isn't that there's an objective answer and a subjective answer. It's that the question being asked doesn't have a strict objective answer.

You've laid out a list of attributes, where the N64 wins more boxes, to argue that the N64 is inarguably superior. But the act of coming up with those checkboxes places subjective value on the particular system attributes. You haven't come up with an objective answer to the question, you've come up with an objective answer to a qualified variation on the question.

 

[dark10x]

can you show us the back of your setup? i wanna see the forest of cables

Sure, why not.

Setup and behind the setup (at least behind the retro systems). Kind of hard to see but it's not too messy, I suppose. They're all hooked up and active (though I have to switch the RGB cable between the SFC and SNES).

Interesting... the major advantage N64 has over the PSX is that it can display vast pieces of terrain using only a couple of large polygons, while the PSX has to subdivide the same terrain into a much larger number of polygons. Because of this, the PSX's polygon count advantage effectively goes to waste.

EDIT: Looks like CTR gets around this to a degree by using quad rendering on the terrain. Very interesting!

As someone else noted this was necessary to avoid more severe texture warping. I'd love to see a wireframe of Quake 2 PSX as that particularly game managed to skirt the issue rather impressively.

The N64 approach was interesting, however, and one that worked well when artists designed around it. Mario 64 remains one of the best examples of stretching low resolution textures across large triangles successfully. They typically designed simple patterns that focused more on colors than fine details and allowed the texture filtering to create smooth surfaces from massive pixels. Finer details, such as bricks, require higher resolution textures in order to not look completely awful. When developers attempted to squeeze too many unique textures into an area and focused on more realistic surfaces you ended up with muddy, unrecognizable designs. The system worked best with more abstract surfaces. There was a real art to texture design and management on that system though the same can be said of the PSX in a very different way. Due to the issue with affine texture mapping this approach would not have worked and you'd end up with massive, warping terrain.

 

[nOoblet16]

It was the PSX first, they later used the name for that thing. If Sony didn't want it called the PSX they shouldn't have made the PS1 (PSOne) later on.



Source

It was the codename, you don't hear people calling the PS4 PS Orbis do you?

 

[Zen]

What game is this?

I think in the end, even though the N64 is more powerful, it had its own drawbacks that were just so bleh. In a lot of ways the higher resolution and hardware AA just helped to highlight some games (and the systems limitations, where as the less powerful PS1 produced a dirtier look that oddly helped to obstruficate that.

Can you imagine how bad some PS1 games would have looked if they ran at the resolution of most N64 games?

Do not forget that during this era, 3D was still pretty bad all around, so the importance of good looking FMVs was kind of a thing, and the N64 had such limited cartridge capacity that... yeah

http://youtu.be/bmViSPAxLpM

 

[nded]

It was the codename, you don't hear people calling the PS4 PS Orbis do you?

No, but you do hear people calling the PS1 PSX.

 

[RedSwirl]

I think we also need to observe that far fewer developers actually worked with the N64 compared to the PS1. Not as many people took cracks at what the system was capable of. You're talking about a piece of hardware that had just under 300 games released for it in North America.

 

[baphomet]

It was the codename, you don't hear people calling the PS4 PS Orbis do you?

Sony referred to the PlayStation as the psx. It wasn't PS and it obviously wasn't PS1.

 

[Uncle Rupee]

What game is this?

Brave Fencer Musashi. That particular shot compares well with Mario 64's opening shot (at least initially - the camera doesn't actually move around), and it has some really detailed textures, but the character and enemy models tend to look pretty bad. It's a cute little game, though it's no Zelda killer.

 

[MrCunningham]

Woah, hello necro thread from 2012...

I had NFS as well. It's not even close. The amount of geometry (quads, in this case) being pushed in Daytona 2 far exceeds NFS on the PC. It also features superior textures and runs at a higher frame-rate. Those older NFS games don't run at a proper 60 fps (engine design limitation, I suppose).

In those days the PC wasn't particularly capable of throwing around large numbers of triangles. It simply couldn't handle the more geometrically dense games being released in the arcade or on consoles especially not at a constant 60 fps as we saw with Model 3 games. There was eventually a PC port of Sega Rally 2, of course, but it was cut down compared to the arcade release (much shorter view distance, less detailed cars, less detailed shading, missing lighting, etc). It looks fine at a glance but it's not comparable to the proper Model 3 version. Nor was the crappy Windows CE Dreamcast release.

I remember Superbike 2000 being fairly impressive back in the day: https://www.youtube.com/watch?v=qkgbSDK98sQ

...But yeah, I don't thing there was anything comparable to a Model 3 game on PC. PC games back then were still being scaled to run on lower end PC's with no hardware acceleration at all. So you didn't see a lot of polygons being pushed around the screen.

Sega did port some of their Saturn games to the PC around 1996/ 1997-ish with games like Sega Rally Championship 1995, Daytona USA and Daytona USA CCE. I owned all three back in the day, and I don't think that Sega Rally Championship even had hardware acceleration on PC. It ran like garbage on my 1999 era Intel Celeron 500 MHz/ Geforce 256/ 256MB's of RAM PC. It was completely unplayable, even though it was priced as a budgetware tittle. Daytona USA and CCE on PC did have early DirectX implementation, which fared better. Though I think that was a patch that came out much later. But both of those games were still locked at 30FPS and 640x480 resolution max.

 

[A Black Falcon]

Scale of contents in general, not just textures.

I think you exaggerate this. N64 carts are large enough for plenty of texture variety, and with compression plenty of good-quality textures. You'd see plenty of N64 games with a solid variety of good-quality textures if the system had a larger texture cache, but as it is you see that with the best-looking games on the system, which show what it can do when pushed. I know I've read that N64 cartridges were more than large enough for plenty of higher-res textures, had it been easier to get them into the game.

But you're acknowledged that there are some respects in which the N64 can't keep up. Respects which obviously make it impossible for the N64 to do certain things the PSX does. Which, if these things are extremely highly valued by someone, makes the PSX a more capable system.

Absolutely not, no. This is wrong. Saying that just because a system is more powerful in one way it is more powerful overall even though it's less powerful in every other way almost never actually holds up to any scrutiny. I'm sure I've said it before, but by those standards teh Virtual Boy is more powerful than the Wii U since it can do 3d graphics and the Wii U can't. Obviously that's absurd. Saying that the PS1 is more powerful than the N64 just because most PS1 games have better-quality textures than most N64 games isn't much better than that.

And once again, go look at Factor 5 and Rare's later games for examples of N64 games with great texture work. You can get good textures out of the N64 with great programming.

Graphically the only other 'advantage' the PS1 has is an average higher polygon count, but as most of that is chewed up by stuff like those wireframes show -- trying to make the graphics look less awful by using up lots of otherwise-unnecessary polygons -- that doesn't matter very much. And the N64 can match the PS1 in polygon count with custom microcode. Boss Games' World Driver Championship is known to push as many polygons as much of anything on the PS1, for example.

oThe problem here isn't that there's an objective answer and a subjective answer. It's that the question being asked doesn't have a strict objective answer.

There IS an objective answer. Every single graphical category except for textures in most games, versus textures and absolutely nothing else. Add up the score on each side, it's lots of points of the N64 and part of one point for the PS1 (part of a point because gaems like Conker, Indy, and Battle for Naboo show how nice N64 textures can get with developers who knew how to get the most out of the system). And there's your objective answer.

You've laid out a list of attributes to argue that the N64 is inarguably superior. But the act of coming up with those checkboxes places subjective value on the particular system attributes.

Wrong. You're the one putting subjective value on one category above all other ones. Saying that that is something objectively defensible is completely ridiculous.

You haven't come up with an objective answer to the question, you've come up with an objective answer to a restricted variation on the question.

Nope. You're just trying to warp things so much that the PS1 ties a competition it objectively lost. The N64 is newer and better hardware than the PS1. It has a 3x faster CPU, more RAM, more than double the potential polygon count, equal polygon count with perspective correction, Z-buffering, anti-aliasing, and more, etc, etc. Saying that there's any kind of equivalence here is being a Playstation fanboy. Sure, I'm an N64 fan, but that's not why this argument annoys me so much. That it's so obviously wrong and biased against Nintendo is why. People love to concern-troll N64 hardware to such an absurd degree! It's really, really annoying, and inaccurate.

As someone else noted this was necessary to avoid more severe texture warping. I'd love to see a wireframe of Quake 2 PSX as that particularly game managed to skirt the issue rather impressively.

The N64 approach was interesting, however, and one that worked well when artists designed around it. Mario 64 remains one of the best examples of stretching low resolution textures across large triangles successfully. They typically designed simple patterns that focused more on colors than fine details and allowed the texture filtering to create smooth surfaces from massive pixels. Finer details, such as bricks, require higher resolution textures in order to not look completely awful. When developers attempted to squeeze too many unique textures into an area and focused on more realistic surfaces you ended up with muddy, unrecognizable designs. The system worked best with more abstract surfaces. There was a real art to texture design and management on that system though the same can be said of the PSX in a very different way. Due to the issue with affine texture mapping this approach would not have worked and you'd end up with massive, warping terrain.

One other trick Mario 64 used is that for some things, including most of Mario's character model, the game uses shaded polygons, instead of textured ones. It's been said that Nintendo thought that games would do that, and that's one reason why the texture cache was too small. Some PS1 games do this as well, such as Crash Bandicoot. By not long after those games, though, it became clear that people wanted textures on pretty much everything, and the N64 suffered as a result.

I had NFS as well. It's not even close. The amount of geometry (quads, in this case) being pushed in Daytona 2 far exceeds NFS on the PC. It also features superior textures and runs at a higher frame-rate. Those older NFS games don't run at a proper 60 fps (engine design limitation, I suppose).

In those days the PC wasn't particularly capable of throwing around large numbers of triangles. It simply couldn't handle the more geometrically dense games being released in the arcade or on consoles especially not at a constant 60 fps as we saw with Model 3 games. There was eventually a PC port of Sega Rally 2, of course, but it was cut down compared to the arcade release (much shorter view distance, less detailed cars, less detailed shading, missing lighting, etc). It looks fine at a glance but it's not comparable to the proper Model 3 version. Nor was the crappy Windows CE Dreamcast release.

Huh? Late '90s PC hardware with a Voodoo card could beat the graphics in anything on any console of the era, certainly!

 

[dark10x]

One other trick Mario 64 used is that for some things, including most of Mario's character model, the game uses shaded polygons, instead of textured ones. It's been said that Nintendo thought that games would do that, and that's one reason why the texture cache was too small. Some PS1 games do this as well, such as Crash Bandicoot.

Yes, this is an important thing to note. Using flat or gouraud shaded polygons eliminated problems associated with affine texture mapping. It definitely produced cleaner looking models that hold up better today. Tobol 2 uses this method as well in addition to running at a higher resolution and 60 fps allowing it to stand out as one of the best looking games on the system.

This method meshed really well with N64 games, though. If you used shaded polygons in combination with low resolution soft textures ala Mario 64 the results could appear rather cohesive and nice. Too many later N64 titles tried to FORCE texture detail into their games when it clearly wasn't the right approach.

Huh? Late '90s PC hardware with a Voodoo card could beat the graphics in anything on any console of the era, certainly!

Late 90s PC hardware could easily best N64 or PSX, sure, but not Dreamcast, Model 3 and certainly not PS2. PC hardware of 2001 wasn't capable of throwing around triangles and effects like a PS2 could. Not to say PC games of that era didn't have their own advantages, but most of them were limited to very simplistic models and geometric setups. PC developers relied more on higher resolution textures and higher resolutions in general to achieve attractive results. You weren't going to see something like MGS2 running on PCs from that era. The PC port came quite a bit later and had its own set of compromises (like the Xbox version).

 

[ZombiePlatypus]

My god this thread is glorious. Most endearing console war ever. Wash all over me. Take back to the 90s. Subbed.

 

[Seik]

Sure, why not.

Setup and behind the setup (at least behind the retro systems). Kind of hard to see but it's not too messy, I suppose. They're all hooked up and active (though I have to switch the RGB cable between the SFC and SNES).




As someone else noted this was necessary to avoid more severe texture warping. I'd love to see a wireframe of Quake 2 PSX as that particularly game managed to skirt the issue rather impressively.

The N64 approach was interesting, however, and one that worked well when artists designed around it. Mario 64 remains one of the best examples of stretching low resolution textures across large triangles successfully. They typically designed simple patterns that focused more on colors than fine details and allowed the texture filtering to create smooth surfaces from massive pixels. Finer details, such as bricks, require higher resolution textures in order to not look completely awful. When developers attempted to squeeze too many unique textures into an area and focused on more realistic surfaces you ended up with muddy, unrecognizable designs. The system worked best with more abstract surfaces. There was a real art to texture design and management on that system though the same can be said of the PSX in a very different way. Due to the issue with affine texture mapping this approach would not have worked and you'd end up with massive, warping terrain.

What a lovely setup.

I need to buy better furniture. I'm still stuck with my mother's old furniture she gave me when I moved, so I have something like 5 visible consoles while the rest is all in boxes in a closet.

 

[HTupolev]

I think you exaggerate this. N64 carts are large enough for plenty of texture variety, and with compression plenty of good-quality textures. You'd see plenty of N64 games with a solid variety of good-quality textures if the system had a larger texture cache, but as it is you see that with the best-looking games on the system, which show what it can do when pushed. I know I've read that N64 cartridges were more than large enough for plenty of higher-res textures, had it been easier to get them into the game.

Yet the PSX can handle far more with far more variety across larger games. This isn't an exaggeration at all; the basic storage unit for a PSX game is literally ten times larger than the largest carts that were ever made for N64, and you could boost that a lot farther by using multiple discs (and even when you did that, manufacturing could still be cheaper than N64 games).

Absolutely not, no. This is wrong. Saying that just because a system is more powerful in one way it is more powerful overall even though it's less powerful in every other way almost never actually holds up to any scrutiny.

I never said that. Obviously it would be self-contradictory, since you'd be saying that the N64 is objectively more powerful than the PSX and the PSX is objectively more powerful than the N64.

What I said was that the N64 is not objectively more capable than the PSX. It does not follow from that that the PSX is objectively more capable that the N64, unless there exists a strict objective inequality, which I've argued does not exist.

but by those standards teh Virtual Boy is more powerful than the Wii U since it can do 3d graphics and the Wii U can't.

Err, the WiiU can do stereoscopic 3D. But even if it couldn't, all that could be said is that the Virtual Boy would be more capable in that it could do 3D. It wouldn't make it objectively more capable. Though certainly someone with an insanely strong preference toward 3D imagery could consider the Virtual Boy more capable in their own estimation.

Wrong. You're the one putting subjective value on one category above all other ones. Saying that that is something objectively defensible is completely ridiculous.

I haven't put anything anywhere. I haven't laid out any value claims. I haven't said the PSX is objectively more capable than the N64. I haven't claim it tied. I haven't claimed that anything is objectively more capable than anything else, I've simply criticized your claim that that's a claim that we can even make in these cases.

 

[Bam Bam Baklava]

Sony referred to the PlayStation as the psx. It wasn't PS and it obviously wasn't PS1.

although this was not endorsed by Sony as the PSX name was intended only as a provisional codename. In 2003, an entirely different console with the name PSX was released in Japan.

What part of that is ambiguous? Even if at one point they did call it that, the name has been used on an entirely different product and is no longer valid.

 

[A Black Falcon]

Yes, this is an important thing to note. Using flat or gouraud shaded polygons eliminated problems associated with affine texture mapping. It definitely produced cleaner looking models that hold up better today. Tobol 2 uses this method as well in addition to running at a higher resolution and 60 fps allowing it to stand out as one of the best looking games on the system.

This method meshed really well with N64 games, though. If you used shaded polygons in combination with low resolution soft textures ala Mario 64 the results could appear rather cohesive and nice. Too many later N64 titles tried to FORCE texture detail into their games when it clearly wasn't the right approach.

The better teams got great-looking fully-textured graphics out of the N64, so the system CAN do it -- it was just really hard to do.

But sure, more textured/shaded-mixes would have been good for all the consoles that generation, really. But the market came out strongly against such things, and developers followed. I agree that the flat-shaded look has aged pretty well, though, sure; it looks a lot better now than people thought it did in 1996, for sure.

Late 90s PC hardware could easily best N64 or PSX, sure, but not Dreamcast, Model 3 and certainly not PS2. PC hardware of 2001 wasn't capable of throwing around triangles and effects like a PS2 could. Not to say PC games of that era didn't have their own advantages, but most of them were limited to very simplistic models and geometric setups. PC developers relied more on higher resolution textures and higher resolutions in general to achieve attractive results. You weren't going to see something like MGS2 running on PCs from that era. The PC port came quite a bit later and had its own set of compromises (like the Xbox version).

Well, I guess it is true that consoles have often exceeded the PC for a short time after their release, because they can focus all of their efforts on gaming power. PCs always exceed them within a few years, though. For instance I got a new computer in late 2001, and it was probably more powerful than any of that gen's consoles... and high-end PCs by that point were better than any console for sure. You see something similar earlier too, with the SNES era (more powerful than PC games in some ways for a while, but PCs eventually passed it of course.), the N64 (more powerful than consumer PCs for the few months until the Voodoo1 card released), etc.

From what I've heard though, the one and only thing the PS2 really was good at versus other systems of the time is particles and effects and such; it can put a LOT of effects on screen, even more than the Xbox or Gamecube. But does that mean that the PS2 is more powerful than the GC and Xbox (and PC!) when otherwise it's quite a lot weaker? I think not. You have to look at the overall picture.

What part of that is ambiguous? Even if at one point they did call it that, the name has been used on an entirely different product and is no longer valid.

Hah, yeah right. Sony probably only called that thing the "PSX" in order to stop people from people calling the PS1 the PSX, but after a full decade of both Sony and everyone else often calling the PS1 the PSX it didn't catch on, and it shouldn't have. The PSX is the PS1, period.

What are you going to say next, that we should all say Sony PlayStation, with the S always capitalized, because that's how Sony writes it? That we must use GCN instead of GC as the Gamecube's short form, and always spell it GameCube instead of Gamecube because that's how Nintendo wants it written? Come on, you don't need to obey corporate marketing type requirements when they don't make much sense!