Post subject: Programmable Controllers
Limne
Any
Joined: 2/24/2010
Posts: 153
A few months ago I got to talking with an electrical engineer who told me that his final project at the end of his last year of university was to make a programmable controller to help disabled children play video games; he explained the concept in terms of being able to play a fighter game by pushing a single button to input the complicated input sequence necessary to do a move or even a combo. Thus, I got to thinking about whether it might be possible for such a device to deliver a TAS file directly to the original hardware. If such a thing could be done then there'd be no need to question the authenticity of the emulation because you'd be dealing with the actual console. Likewise, it might provide the community with some interesting publicity if physical demonstrations of the technology could be set up on actual consoles for a live audiences. There's a console mod community that might be able to do stuff like that but I wonder whether they could make the frequency of the input device synch with the CPU. It probably wouldn't be very useful for TASing in general because I imagine that runs optimized for emulators would de-synch on the original hardware, but it might help to give some indication of how reliable any given emulator actually is and provide for some interesting demonstrations if runs could be optimized for the intended hardware. At the very least I'd say it's an interesting thought experiment that might help illustration just what TASing actually is beyond the usual vague description of a "god-like player."
Senior Moderator
Joined: 8/4/2005
Posts: 5769
Location: Away
Yes, Morimoto has done that with a PS2 game once: here. He used a programmable controller called SuperComaCon (Super Command Controller), and I have no idea how he managed to battle desyncs — surely the emulator wasn't capable of doing something like that in early 2006.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Joined: 7/2/2007
Posts: 3960
He probably had to make multiple attempts, like the people who have made Guitar Hero "bots".
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Senior Moderator
Joined: 8/4/2005
Posts: 5769
Location: Away
I'm afraid "multiple" isn't really saying much, since in Guitar Hero the time between the events (strums in this case) is always the same, so as long as you keep the timing you know you will succeed if the first hit lands on time. In case with a boss fight as demanding as that a slight deviation early on will most definitely lead to a progressive desync in no time. I'm guessing that took more to the extent of hundreds of attempts.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
nesrocks
He/Him
Player (240)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
Programmable controllers have been commercially available for a long time, example PS2 as posted by moozooh and the SNES http://www.gamersgraveyard.com/repository/snes/peripherals/controllers-programmable.html SN Programpad, Top Fighter, XE-1 Maybe you can just mod one of these to put a full TAS on it.
Senior Moderator
Joined: 8/4/2005
Posts: 5769
Location: Away
It will have to be a very simple and/or lenient TAS then. There was a talk about making the SMB any% work on a real NES, here.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Joined: 2/19/2007
Posts: 424
Location: UK
This is a collection of clips for marvel vs. capcom 2 made using programmable controllers: http://www.youtube.com/watch?v=hx_xAmwENa8 It looks like the timing requirements should be quite strict here, so console based TASes might not be as infeasable after all. It would be very interesting to hear what desyncs were encountered when making this TAS.