State of SNES Emulation - 2010

[Sessh]

A couple? There's options besides SSF?
Though I've not had trouble with SSF in my (admittedly limited) usage.

Yeah, there's one called Satourne and a handful of others, but I can't even figure out how to load a game in any of them. Not user friendly, no load game option or anything. I suppose I could try burning the game ISO to a CD and trying it that way, but seems like a pain to do with every game I wanted to give a look at.


Just a side note: If you go to reply to a post and it makes you log in first, the site should take you straight to the "Post Reply" page or, at the very least, to the page you were at last instead of sending you all the way back to the Board index. Quite annoying!

[Gil_Hamilton]

A couple? There's options besides SSF?
Though I've not had trouble with SSF in my (admittedly limited) usage.

Yeah, there's one called Satourne and a handful of others, but I can't even figure out how to load a game in any of them. Not user friendly, no load game option or anything. I suppose I could try burning the game ISO to a CD and trying it that way, but seems like a pain to do with every game I wanted to give a look at.

Holy cow, when did Satourne start seeing work again? I was SURE that one was dead and buried.

I use Daemon Tools to load my disk images, personally.

[Sessh]

Holy cow, when did Satourne start seeing work again? I was SURE that one was dead and buried.

There seems to be four Saturn emulators that are still in development which are Saturnin, Satourne, SSF and Yabause. Discontinued ones are A-Saturn, GiriGiri, Hyperion, PC-Saturn and Titan. Don't know how accurate that is, though.

I use Daemon Tools to load my disk images, personally.

Ahh, maybe I'll have to try that. Never used Daemon Tools before. I'll have to try and mess with it.

[grinvader]

Saturnin

the duck wins

[IglooBob]

It isn't like the SNES hardware is all of the sudden going to stop working. I imagine in 100 years you will still be able to play the SNES system without issues. The SNES only draws 17 Watts, far less than any computer.

A new "my SNES completely stopped working, what are my options" topic shows up on GameFAQs practically every day. D-- (who I've seen on the bsnes boards but I'm sure shows up at other places) talks periodically about how SNES cartridges can't be expected to last forever. One quote from a topic there in particular captures that:

Anyone who says mask ROMs will last until we have a nuclear war is either ignorant or expecting nuclear war in the next 50 years.

With that being said, seems to me bsnes is absolutely essential in archiving a system that isn't likely to last forever, made of parts that are likely to be increasingly difficult to find, which receives no support or repair from it's creator. Without at least one emulator with completely accuracy as the goal, you can only hope you worked out the Speed Gonzalez or A.S.P. issues in time. Once the last SNES keels over, you're stuck with what you've got as far as knowledge goes.

Edit: one last thought

The "I still don't see the point of dot based rendering" seems like it's a product of right now. If every budget computer could run the bsnes accurate core, no one would feel that way. People would look at everything else and say "I don't see the point in doing things incorrectly."

[badinsults]

I still maintain it would be far more useful to create hardware specs so that accurate chips can be produced. There is a market for that, if you see the amount of clones out there. It would be far more important than trying to create a perfect emulator on another platform, which realistically will never happen.

[IglooBob]

I still maintain it would be far more useful to create hardware specs so that accurate chips can be produced. There is a market for that, if you see the amount of clones out there. It would be far more important than trying to create a perfect emulator on another platform, which realistically will never happen.

Why will it realistically never happen?

Also, if there's a market for accurate console clones, why is it that existing console clones are for the most part pretty shoddy? I look at the emulator scene vs the console clone scene and find that across the board emulators are in a better place. Perfect reproductions of old consoles would be swell. I assume based on the results that there's probably a good reason that isn't happening while emulators like nestopia and bsnes are doing pretty well.

[Gil_Hamilton]

I still maintain it would be far more useful to create hardware specs so that accurate chips can be produced. There is a market for that, if you see the amount of clones out there. It would be far more important than trying to create a perfect emulator on another platform, which realistically will never happen.

I'm not really convinced that market is as large as you think.

Look at the Famiclones. They actually used to be more accurate than they are now.
The first out the door with a one-chip solution took over the market, even though it was a piece of shit relative to the multi-chip solutions. It doesn't even work right with Legend of Zelda, which is about as obvious a test cart you can think of.

And no one will fix it because that costs more money than NOT fixing it and using the known-defective 1-chip solution.
Not even relatively high-profile people with relatively large marketing budgets that promised to fix it, but instead released the same chip as everyone else, only wired up wrong to be even less compatible, somehow. (Hello, Messiah! No, I'm not holding any grudges! What gives you that idea?)

For the clones, doing it CHEAP matters more than doing it RIGHT.

Also, if there's a market for accurate console clones, why is it that existing console clones are for the most part pretty shoddy?

It DID happen with the Atari Flashback 2. Which was designed by an enthusiast, and designed to be easily moddable(so you could add a cart slot, even though it wasn't in the company's design specs), compatible with vintage accessories, and... a complete and utter commercial failure(to be fair, it DID follow the famiclone-based Flashback 1, which we shall not speak of again after pointing out the name was tainted..).

[badinsults]

Why will it realistically not happen? Because the computing power needed to achieve it makes it impractical.

And the reason why console clones are shoddy (though that might be a debatable statement) is that no one has bothered to draw up the specs needed to created hardware clones. Creating a design to perfectly clone hardware is in demand, as evident by the success of NES On a Chip (which itself was not all that accurate). The NES On A Chip was used in a variety of clone system.

The patents on SNES hardware are starting to run out, so there is no legal reason why hardware could not be developed. Creating a low Wattage SNES clone would be far more practical than an emulator that requires a high end supercomputer to run.

[IglooBob]

Why will it realistically not happen? Because the computing power needed to achieve it makes it impractical.

I would submit that this means "it isn't practical right now" not "it will realistically never happen."

And the reason why console clones are shoddy (though that might be a debatable statement) is that no one has bothered to draw up the specs needed to created hardware clones. Creating a design to perfectly clone hardware is in demand, as evident by the success of NES On a Chip (which itself was not all that accurate). The NES On A Chip was used in a variety of clone system.

Which leads me to re-asking my question, I guess. If we could draw up the specs to create a perfect NES clone, but we haven't done it, why not? I have a hard time believing that it's just an issue of "no one has bothered to do it." People are choosing to put their time into writing emulators instead, yeah? Gil_Hamilton's comment about "doing it CHEAP matters more than doing it RIGHT" seems like it's probably true. Seems pretty clear IMO that the market for console clones is not one that would make accuracy a top priority. Certainly from the information we already have, it looks that way.

[grinvader]

seems to me bsnes is absolutely essential in archiving a system

hahahahahahaha

hahahaha


hahaahahahahahahahaha


good one

[IglooBob]

hahahahahahaha

hahahaha


hahaahahahahahahahaha


good one

thanks for the useful reply

[Gil_Hamilton]

Why will it realistically not happen? Because the computing power needed to achieve it makes it impractical.

And the reason why console clones are shoddy (though that might be a debatable statement) is that no one has bothered to draw up the specs needed to created hardware clones. Creating a design to perfectly clone hardware is in demand, as evident by the success of NES On a Chip (which itself was not all that accurate). The NES On A Chip was used in a variety of clone system.

Remember what I said about the older multi-chip NES clones being MORE accurate than the NOAC? Famiclones actually went downhill after the NOAC came out.
Now granted, at least some of the old clones were likely made with parts stolen from Nintendo's own production line, but still.

The one-chip solution was the CHEAPEST way to do it, which is why it displaced all the multi-chip solutions, and why no one's designing a new one.
Maybe if they could advertise that they're the best clone and all others are inferior, it'd matter. But if they start naming games, Nintendo will drop on them and throw a hissyfit. And the retailers will probably just buy the cheaper, easier to move clones anyways.


Right now, the clone market is driven far more by cost than quality. It's a sad state of affairs.
Oh, and confusing gullible consumers by putting it in a knockoff of a Wii case. Driven by cost and deception more than quality.

[grinvader]

hahahahahahaha

hahahaha


hahaahahahahahahahaha


good one

thanks for the useful reply

more useful than your fanboi rant by a couple magnitudes

[badinsults]

Why will it realistically not happen? Because the computing power needed to achieve it makes it impractical.

I would submit that this means "it isn't practical right now" not "it will realistically never happen."

And the reason why console clones are shoddy (though that might be a debatable statement) is that no one has bothered to draw up the specs needed to created hardware clones. Creating a design to perfectly clone hardware is in demand, as evident by the success of NES On a Chip (which itself was not all that accurate). The NES On A Chip was used in a variety of clone system.

Which leads me to re-asking my question, I guess. If we could draw up the specs to create a perfect NES clone, but we haven't done it, why not? I have a hard time believing that it's just an issue of "no one has bothered to do it." People are choosing to put their time into writing emulators instead, yeah? Gil_Hamilton's comment about "doing it CHEAP matters more than doing it RIGHT" seems like it's probably true. Seems pretty clear IMO that the market for console clones is not one that would make accuracy a top priority. Certainly from the information we already have, it looks that way.

Your arguments about economics have nothing to do with the argument of emulation versus hardware.

As I stated in an earlier post, if someone were to draw up the specifics to create a perfect reproduction of SNES hardware chips, there surely will be some hobbyist engineer who will do it. In fact, if you look at some of the things that people like d4s has done (custom SNES BIOS), I can guarantee it will happen. Also, look at the number of flash carts that have come out in the past couple of years. The fact is, people want to play games on real hardware. An emulator is fine for programming, hacking, screenshots or doing certain other things that a real system can't do (like TASs), but if you go up to regular gamers, most will say they prefer playing a game on a real console on a TV. And I am one of those people. Why do you think that those clones sell?

[IglooBob]

You have brought up what there is a market for, what people are buying, and what you think that means. If you don't want to talk about economics, stop making arguments about economics.

You stated that you don't see the point of dot based rendering, and that writing a perfect emulator will never happen. Are you willing to actually address those points? I am genuinely interested, if so. If not, please let me know and I'll stop wasting our time.

[badinsults]

What is the point of dot based rendering if it requires a supercomputer to accomplish it? Because even after that, you cannot simulate the exact manner of randomness that a silicon chip provides. The end result will prove to be something that no one will use to play games, because no one is going to buy a supercomputer to play Super Nintendo. Hell, I wouldn't even justify purchasing a new computer to play the "performance" mode at fullspeed.

The Super Nintendo is a machine that is meant to play games. I think some people have lost sight of this. Squeezing out that extra .01% accuracy is basically wasted effort. For all reasonable purposes, the program bsnes is complete.

I've been very closely following the classic video game scene for many years. The SNES is largely ignored by the enthusiast crowd, despite most people highly regarding the quality of games on the platform. You don't see the careful documentation like Atari Age. You don't see the insane collect-ability of the NES, which is found on Nintendo Age. You don't see the dedicated gamers like you do on Sega-16. I am even witnessing the rise of dedicated communities of Nintendo 64, Playstation 1 and Dreamcast fans. Where is the SNES scene? At one point in time, the emulation scene was dominated by the SNES, but I think interest in emulation is dying out as people are getting older. Back before 2004, the ZSNES and SNES9x board were filled with SNES fans, people who loved the system and its games. I remember the release of the Star Fox 2 ROM, and the flurry of excitement it generated. The point of my State of Emulation was to show that there are great emulators out there, but emulation is not going to drive the scene. Creating a super accurate emulator has not increased interest in this system that we all love. And byuu has effectively alienated himself from every fragment of the SNES scene that does exist. I want the SNES scene to grow and for it to have the same amounts of enthusiasts as contemporary consoles.

I hope this elaborates upon what my point is. My original intent was to be diplomatic about it, but ultimately I felt my point was not getting through. You know, I do like bsnes, it has some nice features and it does play games other emulators don't. But nothing feels more right than sticking a cart in my SNES consoles (I have both NTSC and PAL consoles, side by side), and playing them on a real TV. And if you look at the aforementioned websites, that is the preference of most classic gamers.

[Gil_Hamilton]

As I stated in an earlier post, if someone were to draw up the specifics to create a perfect reproduction of SNES hardware chips, there surely will be some hobbyist engineer who will do it. In fact, if you look at some of the things that people like d4s has done (custom SNES BIOS), I can guarantee it will happen. Also, look at the number of flash carts that have come out in the past couple of years. The fact is, people want to play games on real hardware. An emulator is fine for programming, hacking, screenshots or doing certain other things that a real system can't do (like TASs), but if you go up to regular gamers, most will say they prefer playing a game on a real console on a TV. And I am one of those people. Why do you think that those clones sell?

Well, they don't sell because people want to play their games ACCURATELY, that's for sure.
I'd rank the Famiclone well below an emulator.


Sure, someone will reproduce a full SNES chipset with as close to perfect accuracy as possible. IF it's cheaper than the clone chipset that's out there now, it will get picked up.
Eventually, someone will integrate an entire SNES into one chip, making it as cheap as possible. And they WILL take over the market, virtually overnight. We can only hope that someone knows what they're doing, because once the 1-chip SNES hits, it's over.


I'm not arguing emulation VS hardware. I'm arguing ideal VS reality.
Ideally, accuracy will win out over cost. In reality, it won't.

My point isn't that emulators are better because China can't fuck them up.
My point is that I see no reason to believe SNES clones are fundamentally different than NES clones.

We've ALREADY seen cheaper hardware won out over more accurate hardware. No one MAKES any Famiclone hardware but the 1-chip solution, even though better chipsets exist and were used in older Famiclones.
And no one wants to re-invent the wheel and make a chip that's cheap AND accurate, because it would cost more money than pretending everything's OK and ignoring the elephant in the room.

The hobbyists have their refurbished NESes and Famicoms, with a ready supply of replacement parts from "broken" systems on eBay. The masses have their flawed clones. And never the two shall meet. That's the state of the NES now.



Now you tell me why, if someone creates a perfect reproduction of the SNES, they will not be supplanted by someone that designs an SNES-ish that fits on a single dirt-cheap chip. What makes the NES and SNES so different?

[paulguy]

Hearing about the state of the NES and SNES hardware, like cheap inaccurate clones with seemingly no accurate alternative or the fact that there will be some time in the future, potentially in my life time, that the systems and games themselves will just be completely unrecoverable apart from maybe a few systems and carts that are holding out is just depressing in a way...

I'm all for highly accurate (As far as specification goes. There's always emulating the whole electrical system from an analog point of view but that's typically more useful when simulating specific system failures and stuff, not intended behavior, though I guess some earlier systems had quite a bit of analogue circuitry.) emulators that probably will be playable in the likely not too distant future. Just 5 years ago running the current bsnes version would just be laughable on pretty much any system available (Most supercomputers are made for heavy parallelization, where as far as I know bsnes really is not.) but will run pretty reasonably, given the performance core, on most new systems (Aside from netbooks and handhelds and whatnot.), so I imagine it wouldn't be too long until highly accurate emulators for all of the older systems will be able to run, and maybe low level emulators for N64 with approximately zsnes's level of accuracy (Still not 100% but pretty decent, and a good start towards something more accurate than HLE can do.), eventually.

Sure, perfect hardware clones would be great, and I'm sure some dedicated people will make some small runs of them for personal use, and maybe publish open schematics and stuff, but as has been said, unless it's cheaper to produce, it won't take off, and likely the more accurate circuits will probably be more complex and expensive, especially if the clones are just lacking features entirely (See Genesis clones that don't work with 32X or sonic and knuckles or virtua racing or have the expansion port required for Sega CD.). Hell, I've seen clones that just use an ARM chip of some sort and embedded emulator, which produces considerably worse results than the clones, but those chips can probably be gotten for much cheaper than a custom chip, even if it's the NOAC, but they suffer from much worse accuracy and lower frame rates and often even slowdown due to trying to fit some kind of emulator on to a low spec general purpose CPU.

Basically my point is that I'm going on about whatever for way too long like I always do and there is no point, so I guess just take whatever from that you want as my opinion on certain things.

[Gil_Hamilton]

Hearing about the state of the NES and SNES hardware, like cheap inaccurate clones with seemingly no accurate alternative or the fact that there will be some time in the future, potentially in my life time, that the systems and games themselves will just be completely unrecoverable apart from maybe a few systems and carts that are holding out is just depressing in a way...

The NES is the one in real immediate danger. A lot of people throw them in the trash when the damned ZIF connector in the cart slot gets flaky.
Add the inability to get anything but NTSC/PAL RF and composite video out of it, and it's days are seriously numbered outside the collector's market.

I think most of the Atari-era stuff is already IN collectors' hands(or lost in basements and attics). It all vanished from the standard thrift store and garage sale outlets several years ago.

[IglooBob]

Thanks for making your point clear. I can't really comment much on the health of the SNES scene. I do remember looking for a Sega-16 equivalent and being disappointed that I couldn't quite find one. My initial reaction is probably something like "well hey, they're still translating/hacking SNES games and people are still playing the translations/hacks so we're doing okay" but I can't be sure. SNES seems like it has one of more vibrant communities in that regard. Compared with say the Sega Genesis, where basically one guy's translating stuff and a few people are making Sonic hacks.

[magitek369]

Pardon my ignorance, but what causes complete failure in a system as old as the SNES? Mine's damn near 20 years old and still ticking. Maybe it won't last forever, but when compared to modern systems, the classics were built like tanks.

[Gil_Hamilton]

Pardon my ignorance, but what causes complete failure in a system as old as the SNES? Mine's damn near 20 years old and still ticking. Maybe it won't last forever, but when compared to modern systems, the classics were built like tanks.

Well, there's a few things that could cause it.

Corrosion, bad solder joints failing, worn connectors, static discharge, blown fuses, just plain parts failure... the list goes on. And what may seem like dead to one person can beeasily recoverable to another(See: NES cart slot, 5200 controllers)

As far as built like tanks... for every VCS and SNES I can point to a NES or 5200.
Hardware durability has a lot more to do with individual system design than it really does era or company.

[magitek369]

As far as built like tanks... for every VCS and SNES I can point to a NES or 5200.
Hardware durability has a lot more to do with individual system design than it really does era or company.

Not to be a smartass, but I feel like the following applies here.

[adventure_of_link]

As far as built like tanks... for every VCS and SNES I can point to a NES or 5200.
Hardware durability has a lot more to do with individual system design than it really does era or company.

Not to be a smartass, but I feel like the following applies here.

And to be a further smartass, you seem to forget to include a picture of the NES game being displaying 100% correctly on the TV as it should be.