Build Your Own Arcade Controls Forum
Main => Software Forum => Topic started by: AmericanDemon on May 31, 2004, 12:07:48 am
-
Anyone have experience with this product?
http://www.kaillera.com/
It says it will allow multiplayer over a network, basically allowing two cabinets to work together as if they were one. Kinda cool.
-
My friend used to use MAMENet for networking MAME computers.
http://members.tripod.com/~adcockm/
But it seems untouched for a few years now. If you find something current and stable, be sure to post it here!!!!
-
I've been looking into adding Kailera to No Name MAME for a while now, but I don't really understand how it works.
-
It works pretty well ... I used to play over the internet with it all the time. :)
-
From what I can gather on their message boards, the project has been abandoned, but there are still a few people there who use it even with the current version of MAME.
-
yeah ive played this several times. not only mame, but with other emus as well. i played mario kart 64 with some people a while back....it ran kinda slow, but i think that was cuz the other pcs were slow. i did play 4 player tmnt on it and it ran flawlessly. its really cool, especially if you dont have any people around to play.
-
I've been looking into adding Kailera to No Name MAME for a while now, but I don't really understand how it works.
Trust me, you don't want this :) I think it would be easier to start over and use SDL_net.
Kaillera does alot of hacks to enable things that aren;t normally in mame too.
To answer the first person, kaillera works ok, but is very laggy. It seems like most people play fighters with it, which I don't understand because it is very laggy.
-
I've never worked with network stuff at all (outside of Windows Networking, even than I use the wizards), so I don't really think I know enough to add it from scratch.
-
SDL had some pretty good tutorials and examples if you know where to look.
-
Heh... I'll tell the truth.... killera is crap. The only thing I've been able to play lag free on it is tetris on the old impact emulator.
-
good ping = good kaillera netplay.
A friend of mine who lives local has a 35-40ms ping to me (he as cable, I have adsl). I set up the kaillera server on my machine. Note, ideally the server should reside inbetween the two parties distance wise to get the best results (if possible).
At a 20 keyframes set, we where playing Street Fighter 2 Hyper Fighting in a more than acceptable state, with only the odd bit of lag. Not bad, considering we where also running Skype VoIP at the time, so we could sh!t talk each other during the rounds... The performance if more than acceptable, bordering excellent IMO!
In a LAN environment, Kaillera has worked flawlessly when ever I tested it. Set the keyframes to 60, and its is as if you are both sitting on the same machine.
While kaillera may not let me play people on the other side of the planet, I can't really complain as even PS2 online only lets me play against peepz within Australia/NZ. Keeping within those boundries (with acceptiable ping to the kaillera server being <100ms) and I feel kaillera performs well too.
Its just a shame development on it has seemly ended...
If you look on the Kaillera forums, you should find posts pointing to more recent releases of a MAME (.81 at last check) that support kaillera.
-
Its really too bad that nobody has come up with a good generic solution for mame. The Mamedev team has said they would consider including any good crossplatform solution into the main stream...
I have done a lot with streams and trying to send values fast (but dind't care about ping for work)... but I have no idea what you would send around to sync two mame versions.
-
The problem with that is networking isn't cross platform... at least on the software level. Each os has a different way of accessing the tcp-ip stream. I think one of the things that's really holding mame back is the insistance of the devs to have cross-platform support for everything.
The windows port is the main port at this point, so imho anything that is added that's windows only should simply be left to the windows.c driver. The other people in charge of porting to other oses should be responsible for finding a similar function for their respective oses. Afterall, how many people do you know that can program on linux, windows, dos and osX at the intimate hardware-interface level?