skeezix
Internal Development
Given I'm stupidly busy, I'd like to keep my time thats for this project perhaps in the growth, the code, etc, but less in the administrivia. (I'm in a call right now and typing in background, so permit me to be scatterebrained like normal
Ultimately I could move the server to EDs box, but I wouldn't want ED to take over admin of it anyway, he's got enough on his plate.. so those issues are down the road.
But right now, I'm thinking.. for adding new 'MAME games' for the c4a-mame side of compo4all, its not too hard to add them; the time sink is in the stages of testing..
- picking a game(s) to add
- testing they work in PanMAME and on Pandora (valid controls, performs well, and playing well enough to register high scores as needed)
- testing they work in the decoders for the compo4all server (I have written custom decoders, but now we also have an automated decoder from other people, but its hit or miss.)
---> once good, I can just add the conf to the server over 30s and the game goes live (and milkshake does artwork etc.)
Normally I do it all myself of course, takes a few hours to test the games, and then I release them to production environment,.
This month I've had _no time_ at all hardly, so I just said hell with it and tosses out a dozen games to see 'what sticks', let people try them out; pretty poor odds.. maybe 3 out of 10ish worked.. annoying Mostly its annoying to players.. and as you know, I like to present a pretty solid front to users; I don't mind having it rough in alpha, but even ebta shoudl be pretty solid. So throwing some out is not happy for me
Maybe next month I'll have more time, but right now its looking like June will be rough (RL madness.)
So it occurs to me..
- Given a sufficiently Linux talented person (not a coder, but someone not afraid of command line), I could show them how to do the decoder test pretty easily
- They could do the PanMAME testing (or coordinate crowd sourcing, like I do.. get our boards to do it!)
- And they could also then pick the games to add each month; again, I crowdsource it, get people to suggest, etc.
---> I would still do the server side conf work/etc; the topic here is the actual game picking and testing, which can be done under an x86 linux install easily enough (even other OSes, but I'd like testing to be done in the same env as the server, for consistency. Otherwise I'd make you test on Pandora
Its not a large time invest, but it _is_ a time invest.. a couple hours to test everything, after you've picked the titles (or through the picking process.)
Anyone want to do it? It might work out. Not saying we'll go this way, but like to see how interested people are.
jeff
Ultimately I could move the server to EDs box, but I wouldn't want ED to take over admin of it anyway, he's got enough on his plate.. so those issues are down the road.
But right now, I'm thinking.. for adding new 'MAME games' for the c4a-mame side of compo4all, its not too hard to add them; the time sink is in the stages of testing..
- picking a game(s) to add
- testing they work in PanMAME and on Pandora (valid controls, performs well, and playing well enough to register high scores as needed)
- testing they work in the decoders for the compo4all server (I have written custom decoders, but now we also have an automated decoder from other people, but its hit or miss.)
---> once good, I can just add the conf to the server over 30s and the game goes live (and milkshake does artwork etc.)
Normally I do it all myself of course, takes a few hours to test the games, and then I release them to production environment,.
This month I've had _no time_ at all hardly, so I just said hell with it and tosses out a dozen games to see 'what sticks', let people try them out; pretty poor odds.. maybe 3 out of 10ish worked.. annoying Mostly its annoying to players.. and as you know, I like to present a pretty solid front to users; I don't mind having it rough in alpha, but even ebta shoudl be pretty solid. So throwing some out is not happy for me
Maybe next month I'll have more time, but right now its looking like June will be rough (RL madness.)
So it occurs to me..
- Given a sufficiently Linux talented person (not a coder, but someone not afraid of command line), I could show them how to do the decoder test pretty easily
- They could do the PanMAME testing (or coordinate crowd sourcing, like I do.. get our boards to do it!)
- And they could also then pick the games to add each month; again, I crowdsource it, get people to suggest, etc.
---> I would still do the server side conf work/etc; the topic here is the actual game picking and testing, which can be done under an x86 linux install easily enough (even other OSes, but I'd like testing to be done in the same env as the server, for consistency. Otherwise I'd make you test on Pandora
Its not a large time invest, but it _is_ a time invest.. a couple hours to test everything, after you've picked the titles (or through the picking process.)
Anyone want to do it? It might work out. Not saying we'll go this way, but like to see how interested people are.
jeff
Last edited by a moderator: