N64 Anyone?

[God Ginrai]

For the sake of being thorough (what a strange looking word), how much of the original of any existing emulator would be left after porting?
Very little. Strmnnrmn's Daedalus is heavily optimized for the PSP. He spent a huge amount of time on it, it's about as optimized as it can get, it's still only marginally playable with two or three games, and that's without sound.

Considering the vast amount of software that can easily be ported to the Pandora, the thousands of open source games and applications that will run flawlessly within weeks of launch, it's a little silly to dwell so much on one emulator that may or may not ever happen to a useful extent, and certainly won't happen soon.

Chip...
[awesome vader pic]
I find your lack of faith disturbing...

Chip, I have to agree with keaft here. You're taking the dev standpoint from the GP2X. This is not the GP2X. This is the Pandora. I know you of all people should not have to be reminded of that, but you need to remember that the Pandora DOES have the power to emulate the N64. I can't make any promises about fullspeed, but we all know that it should be at least better than the N64 emulation on the PSP. This emu will happen, we just don't know when.

-God Ginrai

 

[PoisonedV]

I find your lack of doubt disturbing.

 

[.Gogeta§§J4BR.]

For the sake of being thorough (what a strange looking word), how much of the original of any existing emulator would be left after porting?
Very little. Strmnnrmn's Daedalus is heavily optimized for the PSP. He spent a huge amount of time on it, it's about as optimized as it can get, it's still only marginally playable with two or three games, and that's without sound.

Considering the vast amount of software that can easily be ported to the Pandora, the thousands of open source games and applications that will run flawlessly within weeks of launch, it's a little silly to dwell so much on one emulator that may or may not ever happen to a useful extent, and certainly won't happen soon.

Chip...
[awesome vader pic]
I find your lack of faith disturbing...

Chip, I have to agree with keaft here. You're taking the dev standpoint from the GP2X. This is not the GP2X. This is the Pandora. I know you of all people should not have to be reminded of that, but you need to remember that the Pandora DOES have the power to emulate the N64. I can't make any promises about fullspeed, but we all know that it should be at least better than the N64 emulation on the PSP. This emu will happen, we just don't know when.

-God Ginrai

you simply took the words out of my mouth!
chip: this is not the gp2x and you are not davec.
you dont need to act like a fanboy and be overprotective about a piece of hardware.
the thing can definitely play most n64 roms if you port daedalus and make some smart optimizations. end of story.

 

[Mithrildor]

Let's be captain obvious for a sec:

We are all having a huge discussion about if the Pandora can handle it. We all know it can. The only problem is: Its the best to start from scratch and optimize it for the Pandora, but it requires alot of work. Lets make a new topic so we can see easily who can develope it.

 

[Chip]

the thing can definitely play most n64 roms if you port daedalus and make some smart optimizations. end of story.
See, this is exactly why I'm constantly trying to downplay N64 / NDS / DC / etc. on the Pandora.

You're not qualified to make a statement like that. But any random person wandering into the forum searching for "daedalus" is going to find that post, assume you do know what you're talking about, and take it as gospel. This causes problems for the project in general and me in particular.

Even if that statement is accurate (and it probably is), it is still misleading. It makes playable N64 emulation seem inevitable and immanent, and it is neither.

Metaphor: It is technically possible for humans to visit Mars. The technology necessary to make it happen already exists. However, until the leader of some country with both the money and resources needed for the project gets in front of a TV camera Kennedy-style and says "We're going to Mars. Here's the plan...", nobody is going anywhere. No amount of standing around shouting "We can do it!" is going to put a person on the red planet.

Similarly, it doesn't matter how many armchair programmers say "N64 emulation is totally doable." Until somebody with the knowledge and resources says "I'm confident I can make it happen, and I'm working on it", nothing is getting done.

Like I said in my previous post, there are so many things the Pandora can and will do right off the bat. It is a waste to spend all your time pining over one emulator that might not happen. It's actively detrimental when you state that it will happen when that is not necessarily the case. Misinformation like that is why bad articles get written and it's why people come here with unreasonable expectations and get disappointed when they aren't met.

 

[Happy]

I would agree with Chip here. The Pandora has the potential for N64 emulation, but it is very far from reaching it. I would be very pleased with the device for its UMPC nature, regardless of what games you could play on it. I wanted to make sure I wasn't being short-sighted in completely dismissing a port, if anybody had a qualified statement as to how much time and work could be saved by porting.

I also think it doesn't get us anywhere to discuss potential, but would much rather get on with implementation. Let there be no doubt, I have the knowledge and a plan, but resources, such as time and motivation (maybe a development environment too) may be harder to obtain. Please do not construe this to be any sort of promise.

 

[kin]

Same, faith is one thing. As a non-technical guy I would say it's capable of producing that kind of 3D graphics like PS1. But how to implement it with the hardware/software etc... is a guessing game.

And you know... guessing games + false misunderstood hope = bad community karma ;-)

 

[quartercast]

The Pandora has the potential for N64 emulation, but it is very far from reaching it.

How do you know a developer isn't working on an emulator as we speak? Not all devs are vocal about what they are working on. All we Armchairfolk can do is wish, speculate, and blabber. We are good at that, and I don't think we cause any harm in doing so.

If someone can't tell the difference between speculation and fact, well, good luck to them. They obviously aren't fit to be Armchairfolk.

 

[Chip]

If someone can't tell the difference between speculation and fact, well, good luck to them. They obviously aren't fit to be Armchairfolk.
You may want to go back and read some of Happy's other posts in this thread.

Happy's statement wasn't speculation on whether some mysterious stranger may or may not be working on the issue, but an educated opinion based on the condition of currently available emu sources. And unlike us "armchairfolk", he does appear to be qualified to make that judgment. He also appears interested in giving it the old college try, so perhaps you should cut him a little slack.

 

[Happy]

Only the silent developer would know.

I don't think the speculation causes any harm either, but I would rather an accurate picture to avoid disappointments. I like to be pleasantly surprised when the unexpected arrives out of nowhere, instead of slowly let down when the eagerly anticipated fails to make an appearance.

 

[scottm]

Chip, by the community keeping this thread active I don't believe we are giving any false hope, wasting time etc. All we are doing is keeping the issue current, highlighting the fact that we would like to see a N64 emulator one day, and are willing to setup up a dev fund to possibly entice a developer.
You can't make us talk about what you want to hear (what software the pandora does currently run etc), this is a public forum, and we should be free to express our opinions (where they don't impinge on others). I don't see any harm in that at all - and I feel you are being a bit draconian.

Happy, your posts sound promising - and I hope you decide to investigate this further (porting or writing from scratch) And I hear what you say about "no promises"

 

[GoodLawd]

This is oddly reminiscent of a thread on PSX emu on the GP2X from a couple of years back (I think). I think the general opinion was that it won't do very well. It runs now, right?

 

[Exophase]

For the sake of being thorough (what a strange looking word), how much of the original of any existing emulator would be left after porting?

My thoughts are that no dynarec exists for R4300 or RSP to ARM/THUMB/NEON or C64x DSP, so a lot of intimate knowledge of instruction sets and processors will be needed to support a dynarec. Then graphics are not written for OpenGL/ES. I also think the GUI and controls would also need considerable work. What I really want to know, though, is if any of the remaining bits (emulator engine, HLE audio, ...) have, first, any potential for porting with minimal effort, and second, do they any exceptional features such that it would be best to copy them?

Again, I'm not expressing an opinion of disdain for any of the emulators, as I think they all represent serious works of programming. I just want to hear if anyone else thinks any of them are anywhere close to ready for working on the Pandora (or are closer than any of the others).

If someone knows what they're doing (and jumping into understanding Daedalus isn't a 1 day job) then a port probably wouldn't be as big of a change as you think.

Daedalus is written both for x86 and MIPS32, and because of this a lot of the recompiler framework is modular enough to support both. A lot of the work is really in the frontend portion that manages recompiled blocks, determines when new ones should be compiled, links them, etc. ARM/NEON (forget Thumb, forget C64x for now) can do most of what MIPS32 can in the same number of instructions or less, it just doesn't have as many registers. Fortunately, the register caching model on Daedalus is dynamic and adjustable for target (again, since it supports x86 and MIPS32). The definitions for ARM have to be made, but the number of registers for caching is not hard-coded into the system.

OpenGL to OpenGL ES might not be an overwhelming change because Daedalus may not be relying a lot on OpenGL to do geometry and lighting to begin with - I don't know if the HLE is quite that high level. But I haven't looked at it indepth so it's just a guess. Either way, this isn't going to be the only program facing the transition and I'm sure that tools will emerge to help (some translation layers are already available). At the very least this is something that can probably be done on the PC end, it could probably be done right now.

The Pandora's interface is largely a superset of PSP's so making that transition will probably consist of a minimal amount of work. The GUI can be extended more for Pandora's features/environment later.

If I haven't made this clear, the reason why I think Daedalus is a good choice is because it has already proven itself in having a recompiler framework that was written with portability in mind. It'd probably be easier than trying to adapt another framework that might not be as flexible. But really, someone well in the know should sit down with both it and Mupen (the next viable option) and evaluate what would be involved for transitioning either.

 

[Happy]

Thank you very much for your evaluation Exophase. I was hoping for this kind of response. You also suggested to me some of the terminology I was grasping to find, although unfortunately not quite in the same sense in which you used it.
As we are talking about compilation, it makes perfect sense to refer to a front-end and back-end and an assembler, I just hadn't stumbled upon thinking of it in that way.
I know the assembler will need to be provided for the target platform. Optimizations and analysis can likely be split into system independent and system dependent portions. I would like to hear if anyone else has an evaluation of the sophistication of emulator dynarec front-ends and the system independent optimizations and analysis that they can provide.

I also hadn't considered that Daedalus used OpenGL for graphics, which would provide an easier route to OpenGL ES, but it will remain to be seen how efficient it is for performance (hopefully it's good).

 

[Exophase]

The ARM assembler can be adapted from somewhere else, like the one used in gpSP/psx4all (I'm going to be replacing this one with my own eventually..)

The emitters will have to be rewritten for ARM, but hopefully what's being used for MIPS will translate pretty easily a lot of the time. Some instructions like SLT will probably have to be rethought slightly, but everything would still be in a pretty self contained little chunk.

 

[HackModford]

QUOTE
Like I said in my previous post, there are so many things the Pandora can and will do right off the bat. It is a waste to spend all your time pining over one emulator that might not happen. It's actively detrimental when you state that it will happen when that is not necessarily the case. Misinformation like that is why bad articles get written and it's why people come here with unreasonable expectations and get disappointed when they aren't met.



With all due respect Chip that was exactly my intentions :rolleyes: I was fed up with my N64 post being turned down so with a slight rewording (and months of waiting) I came back and have achieved my goal. I'm trying to tell developers "IF YOU WANT FAME IN THE PANDORA COMMUNITY... HERE'S WHAT WE REALLY WANT!" Yeah all the other stuff is great and we appreciate it but this is what we want.

 

[Eniko]

Yeah all the other stuff is great and we appreciate it but this is what we want.
This is completely misleading. What "we" are you referring to? Do you mean everyone wants N64 emulation the most? Personally there are a lot of things I'd rather see than a working N64 emulator so I can play the two or three N64 titles I enjoyed. Also eesh, way to demotivate any and all developers who are doing other stuff. "Well that's nice dear but we want N64 emulation".

That's probably not how you mean it, but at least for me that's how it does come across regardless.

 

[God Ginrai]

Yeah all the other stuff is great and we appreciate it but this is what we want.
This is completely misleading. What "we" are you referring to? Do you mean everyone wants N64 emulation the most? Personally there are a lot of things I'd rather see than a working N64 emulator so I can play the two or three N64 titles I enjoyed. Also eesh, way to demotivate any and all developers who are doing other stuff. "Well that's nice dear but we want N64 emulation".

That's probably not how you mean it, but at least for me that's how it does come across regardless.


I know where both of you guys are coming from. I know that I don't want the fact that I want N64 emulation to discourage developers from doing other things. However, I must admit that I am one of a sizable amount of Pandora users that really want to see an N64 emulator on the Pandora.

If I could have your MMO, lots of new homebrew, and Ocarina of Time on the Pandora, the system would be Gold in my eyes. I've wanted a handheld solution for Ocarina of Time for so long that anyone who could give that to me would definitely be on my list of heroes.

-God Ginrai

 

[mali]

Guys? :huh:

Edit:
I see, Gruso Something no dictionary is able to reproduce, unfortunately.

 

[gruso]

It's a non gender specific term in some contexts, in some some parts of the word, colloquially, sometimes. I use it that way myself. Eniko and her fine sisters ought not feel alienated by it.