What could we hope to emulate?


To answer the original question, if the correct chipset is chosen (not the OMAP 5) then Gamecube is absolutely within the realm of possibly. Its near playable on some android phones today. give it another year and I expect it to be full speed on the latest hardware.
I don't know exactly what the difference between almost playable and full speed is since the former is so subjective, but I expect that means it needs to be at least 2x faster. Expecting 2x single threaded performance improvement in mobile CPUs within one year from now is naive.
 
1. Pandora has a small user base, we don't need to meet every single person's request. Unless ED's selling Pyras for 1000 bucks, in that case I would request every pyras be made of titanium and come with free chocolate and a frog.

2. The more I've to wait for Pyra, the more likely I would spend all of my desposible income on other silly stuffs. (New cameras, smart phones, laptops, pc upgrades, tablets, new games, a 3d printer, a few dogs, and doughnuts, and maybe even a PS4)

3. Even if the pyra can emulate the gamecube, that thing has not enough games. It would be like a having a barber buying up all the advertising space in a cancer ward. Emulating ps2 would be a sounder goal since everyone except those casual wii owners know that the PS2 was and still is the king of video games. (smooth gameplay with 1 or 2 frameskips would be sufficient)

4. The earlier we push the pyra out, the more time developers will have. Pandora's strong point is never cutting edge hardware.

5. As steam is going to linux, more and more linux/debian based devices will be produced by all kinds of communities. We need to get to the party ASAP before all the cakes are gone. (cake joke, lol)

We can always have a Pyra 2ghz edition after pushing the first gen out. At this stage, there's only so many socs we can get our hands on, I'm pretty sold already, the earlier I can get my hands on one, the better to me it is.
 
Last edited by a moderator:
Expecting 2x single threaded performance improvement in mobile CPUs within one year from now is naive.
Hey, it's EssoAir who's talking here so it's not like we don't know how reliable his claims are. :)
Have you seen some of the videos of current hardware running the games?

www.youtube.com/watch?v=7UzBCKnWy0I
 
Last edited by a moderator:
do some reading on forums.dolphin-emu.org/forum-android
Or you can just summarize:
Dolphin used to require full OpenGL, but was updated a few months ago to also support OpenGL ES3.0 for mobile platforms.


Some other details in there, but that's the gist of it: to run Dolphin we need ES3.0 support at least.
Its been almost a year now, actually. iirc it was February 2013 when Sonic made his first build.
 
Expecting 2x single threaded performance improvement in mobile CPUs within one year from now is naive.
Hey, it's EssoAir who's talking here so it's not like we don't know how reliable his claims are. :)
Have you seen some of the videos of current hardware running the games?

www.youtube.com/watch?v=7UzBCKnWy0I
It's not that playable in that video.
Im aware its not, but its imo really close. Sonic has said on IRC that the only thing that prevents it from making significant strides in speed is the time he has. he's quite busy with his job and so cant dedicate much time to development work.
 
Im aware its not, but its imo really close.
It's not really close. When a few karts start appearing on-screen it grinds to a halt. Those performance videos showing time trial runs with no other racers are very misleading. In other words, yes, you can find games that run okay under low CPU pressure situations. Maybe there are some games that are always very CPU light, there are always games like that. But that shouldn't warrant blanket statements about things being fullspeed.

I'm going to stand by what I said, a 2x performance improvement over the high end now in only one year seems unrealistic. And I think that it's going to need more than 2x to keep high CPU spikes from ruining the experience in a huge number of games.

Sonic has said on IRC that the only thing that prevents it from making significant strides in speed is the time he has. he's quite busy with his job and so cant dedicate much time to development work.
What does significant strides mean exactly, do you know? It could be 10% performance. If I increased average recompiler performance by 10% in DraStic I'd consider that very significant. I can't count the number of times I've done optimizations which I thought would produce incredible performance changes only to find that they were pretty marginal. I remember when drk was saying nullDCe would be fullspeed on a PSP, sometimes even brilliant emulator coders can make poor estimations :p I even once told someone that DS emulation might be viable on the old MMSP3 with 300MHz ARM9, that was incredibly poorly thought out and informed.
 
Last edited by a moderator:
4. The earlier we push the pyra out, the more time developers will have. Pandora's strong point is never cutting edge hardware.
Wrong. The reason the Pandora has a OMAP chip instead of a weaker Freescale was because we chose cutting edge.

-God Ginrai
 
id like to remind people of a thread from a while back where this among other things was said

So ARM is not supported?
Well if the source code is available and ARM based processors make a huge improvement we may have dolphin running 2 or 3 frames a second within the next 5 years or so if we're lucky.
See how far we've come in only a year? If trashy and others thought Gamecube on ARM would be that slow 5 years down the line, Sonic has already far and wide shown it to be far better than that. its amazing imo and im really happy with how things have progressed, so thats why I dont understand the lack of optimism. if such strides have been done in recent times, whos to say equal or even greater strides are not going to be made in the year to come? I really hope ED holds off on choosing the CPU until the next gen comes out. if they are able to run dolphin well (as I predict they will) it might be worth it. And if not, then I bet the OMAP 5 will be cheaper by then so waiting is a win-win situation. Either save money, or have support for a whole new game system. :D



also this.


sorry to pick on you Trashy

What about a 2ghz ARM device?
Clock speed isn't everything, you can't compare GHZ.. Hell There were Pentium 4's that ran over 3Ghz that can't match the performance of a 1.2Ghz Core processor.. Also the GPU is a big obstacle, regular PCs use huge video cards drawing hundreds of watts of power to accomplish heavy graphic processing.. Mobile GPUs will need to see a huge leap in technology that my guess won't be coming for at least 10 years..
 
Last edited by a moderator:
If trashy and others thought Gamecube on ARM would be that slow 5 years down the line


Like I'm an expert on anything.


Edit, I still stand by my second quote in general. 
 
Last edited by a moderator:
Just like how we got the rebirth and th 1ghz Pandoras. So were the 1ghz pandoras made by elves at the north pole or ED (and his German buddies) ?

May be I have overlooked something, my points are made based on a few rather unreliable assumptions. 

Wrong. The reason the Pandora has a OMAP chip instead of a weaker Freescale was because we chose cutting edge.
-God Ginrai
There's always more cutting-edge hardwares coming out. By the moore's law, the raw computational power of chips two folds every two years. 

In 2013, there really isn't any soc that is ahead of its time unlike the OMAP when it first came out. (Have a look at the xbone and PS4, they can barely catch up with a mid range gaming pc.) 

Waiting forever may not be as sound of a decision as you may think. Even if something good is coming out, is not likely we can get it. They will be exclusive to the big guys for a while. 

Again my point was not Pyra should not have cutting-edge state-of-the-art hardware. My point was, surely hardware matters, but a better industrial design, lots of functions, an active community is what keeps pandora alive for the last few years. (people are still willing to buy an 1ghz pandora because of the above reasons even though it's about as powerful as a graphics calculator at this point in time)

Surely if we can get something that's really great, why not? But if we can't then it does not matter that much in the grand scheme of things. 
 
Last edited by a moderator:
Wrong. The reason the Pandora has a OMAP chip instead of a weaker Freescale was because we chose cutting edge.

-God Ginrai
There's always more cutting-edge hardwares coming out. By the moore's law, the raw computational power of chips two folds every two years. In 2013, there really isn't any soc that is ahead of its time unlike the OMAP when it first came out. (Have a look at the xbone and PS4, they can barely catch up with a mid range gaming pc.) 

Waiting forever may not be as sound of a decision as you may think. Even if something good is coming out, is not likely we can get it. They will be exclusive to the big guys for a while. 

Again my point was not Pyra should not have cutting-edge state-of-the-art hardware. My point was, surely hardware matters, but a better industrial design, lots of functions, an active community is what keeps pandora alive for the last few years. (people are still willing to buy an 1ghz pandora because of the above reasons even though it's about as powerful as a graphics calculator at this point in time)

Surely if we can get something that's really great, why not? But if we can't then it does not matter that much in the grand scheme of things.
I believe you are mistaken. I did not make any post suggesting what strategy we should use. I merely pointed out that the Pandora is capable of what it is today precisely because we chose cutting edge.

However, in the grand scheme of things, I don't believe your comment about pushing the console out earlier. If EvilDragon keeps compatibility with the Pandora, there will already be a large library for users to take advantage of. Not to mention, even if that wasn't the case, the time after release when all of the developers are releasing new software day after day is one of the most exciting times to have a homebrew handheld. This was true for the GP2X and the Pandora. Although it wasn't as strong with the Pandora due to the complications causing it so that little Pandoras were getting to users and developers at the beginning.

-God Ginrai
 
However, in the grand scheme of things, I don't believe your comment about pushing the console out earlier. If EvilDragon keeps compatibility with the Pandora, there will already be a large library for users to take advantage of. Not to mention, even if that wasn't the case, the time after release when all of the developers are releasing new software day after day is one of the most exciting times to have a homebrew handheld. This was true for the GP2X and the Pandora. Although it wasn't as strong with the Pandora due to the complications causing it so that little Pandoras were getting to users and developers at the beginning.
-God Ginrai
That's a valid arguement you got there. I'm totally agree with you. I've overlooked backwards compatibility's effect on this subject. Thanks for pointing that out for me. 

I guess it would be a difficult decision to make between backwards compatibility and future proof. 

Let's say, the newest OMAP can run most of Pandora's commands nicely without too much tweaking, but the tegra 4, which I'm only using it as an example because it sucks, can theoretically emulate the gamecube at 25 frames per second, and is more powerful and generally better documented than OMAP. Which option would you choose? 

I'm ok with either, I don't want any of our effort on the pandora be wasted, but I also do not want the pyra be left behind by all the big boys. If we are to go with the first, then we don't have that many choices to begin with. We can push the pyra out without waiting for the next best thing. We still have to spend time on designing a better case and keyboard of cause. 
 
Just like how we got the rebirth and th 1ghz Pandoras. So were the 1ghz pandoras made by elves at the north pole or ED (and his German buddies) ?

May be I have overlooked something, my points are made based on a few rather unreliable assumptions.
 Maybe I have overlooked something, I'm not very well informed about TIs roadmap regarding the OMAP Line - are there higher clocked chips announced already ?
 
Maybe I have overlooked something, I'm not very well informed about TIs roadmap regarding the OMAP Line - are there higher clocked chips announced already ?
It does not have to be TI,  we should generally take future upgradability into consideration. Tegra may look promising, but every tegra is completely different from its predecessor. So if we are to choose Tegra 4 for the pyra, we have to accept a rather short product lifespan. ( that is Pyra will be outdated sooner than pandora.)

ED still has not giving us any solid info, so I'm not sure we are even going with TI. (I think it's hinted somewhere though) I'm ok with anything though, even a cortex a9 quad core cheapo would be good for me. 
 
Last edited by a moderator:
Back
Top