Console that fared best vs PC's available at launch

Well, N64 "emulation" is more N64 approximation right now. The majority of the library doesn't run correctly. There are maybe 10 games that work really well, and the rest range from not working or working with noticeable bugs. So you have to try the games out on 3 different emulators and try different plugins to see if you can get things right. Surreal64 on modded Xbox actually includes Project64, 1964, and UltraHLE to choose from and it still won't run anywhere close to all of the games.

I haven't used Nintendo's Virtual Console on Wii yet, but it seems to do the job well. Apparently it just uses the ROMs from the carts with various tweaks. I'm sure it's HLE. I'm just not that interested in buying the games again. ;)

PJ64 runs Mario 64, both Zelda's, Mario Kart 64 and F-Zero X.

That'll do for me ;)
 
Well, N64 "emulation" is more N64 approximation right now. The majority of the library doesn't run correctly. There are maybe 10 games that work really well, and the rest range from not working or working with noticeable bugs. So you have to try the games out on 3 different emulators and try different plugins to see if you can get things right. Surreal64 on modded Xbox actually includes Project64, 1964, and UltraHLE to choose from and it still won't run anywhere close to all of the games.

I haven't used Nintendo's Virtual Console on Wii yet, but it seems to do the job well. Apparently it just uses the ROMs from the carts with various tweaks. I'm sure it's HLE. I'm just not that interested in buying the games again. ;)

True. The only point of the comparison is just a rough apples to apples comparison. I mean, one system emulating the other in real time to some reasonable precision would clearly establish a lower bound of some sort in terms of speed.
 
Im somewhat confused on the basis of the comparison. Perhaps a redundant question, but I wanted some clarification (sorry as I wouldnt doubt this has been addressed).

Are we comparing the overall potential rendering/processing power of a console vs. an up to date pc at the time of console release or a games comparison between those available on a newly released console vs. games available on an up to date pc?

The difference could result in completely different estimates/opinions.
 
Last edited by a moderator:
Proper emulation forms a lower bound on what the performance of a PC has to be to match the console. A poorly coded emulator would bring a Penryn to its knees, so in this case, a Penryn would be established as some kind of coarse lower bound for the equivalence hardware. The question here is what the best coded emulator would require. Without any special interpolation, for the N64, this lower bound would be a pentium II w/ a voodoo. But there is significant overhead and I'd say that UltraHLE proves that it's quite a bit less than the PC of the N64's heyday.

Maybe you misunderstood me, or I'm misunderstanding you.

I was saying that you could make a slow as hell emulator and a fast one. The thing I left out was a better description of what I meant. The fast one would most likely not be very accurate, and could have many issues not caught by most people due to one thing or another(for example, zsnes' sound core is very bitchy. a minor change can completely break a game, and it gets a lot of games wrong. Even the programmers have admitted this. It can also runs code that would never work on the real console due to this last of accuracy.). The slow emulator could be a very accurate one that gets everything as correct as reasonably possible(bsnes, the most accurate emulator to date, runs fullspeed on a core 2 Duo 2.0 ghz or higher. Any CPU less than that will most likely not reach full speed or higher. Myself, I only get 140 fps on the most demanding scenes, and I have a 3.0 ghz Pentium Dual-Core 2160).

Am I making myself clear? Sorry, I just don't communicate well.
 
Im somewhat confused on the basis of the comparison. Perhaps a redundant question, but I wanted some clarification (sorry as I wouldnt doubt this has been addressed).

Are we comparing the overall potential rendering/processing power of a console vs. an up to date pc at the time of console release or a games comparison between those available on a newly released console vs. games available on an up to date pc?

The difference could result in completely different estimates/opinions.

An excellent question and one that should definatly be addressed when making a comparison. For the record all the posts I have made were on the basis of hardware capability (since that interests me more than game graphics and is less subjective). However that may not have been the OP's intent at all.

If I were to post on those terms then I would probably go with the N64 as Mario64 just obliterated anything on PC's at the time. I'll caveat that statement though by saying I only got properly into gaming at around the time of xbox launch so at least graphics wise, anything before that i'm a bit uncertain on.

Certainly I don't think xbox or xbox 360 were significantly ahead when they launched. Marginally perhaps but nothing that would put them in another league like the N64 was.
 
An excellent question and one that should definatly be addressed when making a comparison. For the record all the posts I have made were on the basis of hardware capability (since that interests me more than game graphics and is less subjective). However that may not have been the OP's intent at all.

If I were to post on those terms then I would probably go with the N64 as Mario64 just obliterated anything on PC's at the time. I'll caveat that statement though by saying I only got properly into gaming at around the time of xbox launch so at least graphics wise, anything before that i'm a bit uncertain on.

Certainly I don't think xbox or xbox 360 were significantly ahead when they launched. Marginally perhaps but nothing that would put them in another league like the N64 was.

If we go by just in-game graphics, the PS1 beats the n64. It launched at the end of 1994, and Ridge Racer seems to be best looking game released on launch day. PCs were even further away from that than they were compared to the n64 in 1996.
 
Maybe you misunderstood me, or I'm misunderstanding you.

I was saying that you could make a slow as hell emulator and a fast one. The thing I left out was a better description of what I meant. The fast one would most likely not be very accurate, and could have many issues not caught by most people due to one thing or another(for example, zsnes' sound core is very bitchy. a minor change can completely break a game, and it gets a lot of games wrong. Even the programmers have admitted this. It can also runs code that would never work on the real console due to this last of accuracy.). The slow emulator could be a very accurate one that gets everything as correct as reasonably possible(bsnes, the most accurate emulator to date, runs fullspeed on a core 2 Duo 2.0 ghz or higher. Any CPU less than that will most likely not reach full speed or higher. Myself, I only get 140 fps on the most demanding scenes, and I have a 3.0 ghz Pentium Dual-Core 2160).

Am I making myself clear? Sorry, I just don't communicate well.

I understand what you're saying. You can still make a reasonably good comparison about how capable hardware is via emulation over a direct hardware spec comparison across different architectures even though it'll be partially qualitative.

If one system can emulate another in some way that's 90% accurate in real time (probably true if you can fool a human or run most of its software library) which even the fast emulators do, I'd say it's safe to say that the emulating system eclipses the one being emulated. It takes a significantly faster computer to emulate the original to taht accuracy in anything close to real-time. I'd say fast emulators are perfectly relevant to our comparison because they toss aside irrelevant computations, like digital rights management, and various interrupts not involved in directly running the software, and it's presumably the game software itself which we care about. (I do agree w/ the other comment above though that Mario 64 blew away most PC graphics at the time, but that's due to good programming and design rather than hardware prowess.)

My point was you can then estimate the overhead and compare the emulating system, in this case a P2 using x86, more directly with the older x86 systems which we were interested in comparing to the N64. I wasn't entirely sure UltraHLE would run on a p5 MMX. It's not completely accurate no, but I think the overshoot here is in favor of the x86 system despite the optimizations.
 
I'd say fast emulators are perfectly relevant to our comparison because they toss aside irrelevant computations, like digital rights management, and various interrupts not involved in directly running the software, and it's presumably the game software itself which we care about. (I do agree w/ the other comment above though that Mario 64 blew away most PC graphics at the time, but that's due to good programming and design rather than hardware prowess.)

My point was you can then estimate the overhead and compare the emulating system, in this case a P2 using x86, more directly with the older x86 systems which we were interested in comparing to the N64. I wasn't entirely sure UltraHLE would run on a p5 MMX. It's not completely accurate no, but I think the overshoot here is in favor of the x86 system despite the optimizations.

I dunno, I'd say mario 64 blew away pc graphics because of the hardware. Higher polygon counts, smoother framerate, and later n64 games easily eclipsed mario 64 in detail. (say, banjo kazooie) In fact, n64 did fairly faithful representations of voodoo 1 era games (though probably lower res and almost certainly lower framerate).

And as far as UltraHLE...I don't htink it'd run on a p5 mmx. I remember needing at least a k6 or pentium pro class processor for it to run anything at decent speed, and goldeneye wasn't at playable speeds until a 500mhz athlon.
 
I dunno, I'd say mario 64 blew away pc graphics because of the hardware. Higher polygon counts, smoother framerate, and later n64 games easily eclipsed mario 64 in detail. (say, banjo kazooie) In fact, n64 did fairly faithful representations of voodoo 1 era games (though probably lower res and almost certainly lower framerate).

And as far as UltraHLE...I don't htink it'd run on a p5 mmx. I remember needing at least a k6 or pentium pro class processor for it to run anything at decent speed, and goldeneye wasn't at playable speeds until a 500mhz athlon.

Goldeneye on UltraHLE was screwed though. It ran at about 3fps when I had Mario64 (a much better looking game) running smoothly.

Goldeneye was IMO one of (and possible THE) most overrated games of all time. Quake 2 easily exceeded it graphically but it was hailed as some sort of second coming of the FPS.
 
I ran N64 Golden Eye at 640*480 with a TNT2 and a Celeron 500MHz. The framerate was around 10-30fps with 15 as average. Dunno what emulator I used though since it was such a long time ago. Though the game game ran horrible on the N64 aswell.

Im sure the emulators made with glide in mind would run far better on a Voodoo card and weak CPU than a powerful CPU + non glide card.
 
Goldeneye was IMO one of (and possible THE) most overrated games of all time. Quake 2 easily exceeded it graphically but it was hailed as some sort of second coming of the FPS.

If I recall correctly, the graphics weren't the big deal with Goldeneye. However, in 1997, enough people didn't have 3D cards that non-pixellated 3D was still a pretty big deal. Anyway, what made Goldeneye a big deal at the time was the AI, the mission structure, the interactivity, and the multiplayer options. It was the first to do a lot of things that are pretty standard now. Quake II didn't even begin to compare on that front. At its heart, it was just a prettier Doom. It didn't really do a lot to advance the FPS from a design standpoint.

You want overrated, try Donkey Kong 64. Now there's an absolute dog of a game.
 
Last edited by a moderator:
Back
Top