PowerMAME - New Derivative Build

[MikeQ]

Maybe this has been beat to death but I still feel like I needed to add my two cents. Flame on if you don't like it 

Why not build and distribute a bunch of diffs that anyone can apply to the main MAME trunk?  It would be very easy, and resource-light to build a PHP script that can zip a bunch off selected diffs and provide that for download instead of the entire source and/or compiled versions with every option.

It a well known experience that the older MAME is the slower it gets running games.  Will any of your code effect the gameplay even if the option is turned off?  If so, I don't think people will want that code in there if they're not using it and it would be worth providing diffs for people to selectively compile in the code to avoid different compiled versions.

This is similar to how gentoo ebuilds work. The simplicity of it is what keeps it running so smoothly.

You could provide both a fully compiled build and a complete zip (and a batch file/shell script that will apply them selectively)/PHP interface of all the diffs and then even save on some bandwith.

Not gonna happen, ever.  I'd end the project if this is what people wanted.  Someone else could deal with the hassle.  It will be too much work to independently test all the permutations that this will produce.  Also, many of the modifications won't be compatible with older builds.  Additionally, most people don't compile their own version they just take the executable.  This would cater to a very small percentage of the MAME users.

Also, having each feature be coded so it could be a standalone feature would require more code than coding them to cohabitate.  If someone wanted all the features, likely it would be bigger than what we have now.

Mame getting slower has little to do with the quantity of code too.  It is because the architecture has changed.  Mame has been redesigned to make it easier to add new games.  This means compartmentalizing code and making things more modular.  The very thing that has made MAME slower is what you are proposing.

[Tiger-Heli]

Mame getting slower has little to do with the quantity of code too.  It is because the architecture has changed.  Mame has been redesigned to make it easier to add new games.  This means compartmentalizing code and making things more modular.  The very thing that has made MAME slower is what you are proposing.

They are also moving the "Sweet Spot" to faster games. http://www.mame.net/mamefaq.html#m14

[Popcorrin]

I have a question that might have already been covered but I didn't see it mentioned anywhere.    If a game is set to cocktail mode does powermame instruct the ledwiz to light both the player 1 and player 2 buttons? 

This was kind of a side thought.
I thought I would be kind of cool if it only lit the buttons of the player that was currently active but I think that would be something more apt to be controlled by glitter's LSE which I don't think is being worked on.

[MikeQ]

I have a question that might have already been covered but I didn't see it mentioned anywhere.    If a game is set to cocktail mode does powermame instruct the ledwiz to light both the player 1 and player 2 buttons? 

This was kind of a side thought.
I thought I would be kind of cool if it only lit the buttons of the player that was currently active but I think that would be something more apt to be controlled by glitter's LSE which I don't think is being worked on.

The current version does not but the next version does.  I just ran into this when implementing the 49way stuff.  If the dip setting for cocktail is NOT set, a 49way menu selection only appears for 1 player if you set the dip to cocktail, then all the 49way menu selectors appear.  I realized when I did this that I needed to do the same for the lighting code to.  It has been implemented and is ready to go.  I'm probably going to do a minor revision release shortly (week or 2).

[screaming]

Eh well it was just an idea. Great work regardless!

[MikeQ]

Eh well it was just an idea. Great work regardless!

Thanks for taking it so well.  I expected  a flame war.   It is a nice concept but very difficult to maintain and test for the developer.  Looking at the downloads of PowerMAME too, I can tell you most people are just taking the .exe.

[screaming]

Eh well it was just an idea. Great work regardless!

Thanks for taking it so well.  I expected  a flame war.   It is a nice concept but very difficult to maintain and test for the developer.  Looking at the downloads of PowerMAME too, I can tell you most people are just taking the .exe.

  haha!  Well if I didn't wait the 2 days to reply then it probably would have been   PowerMAME is your baby and I respect your decision. If the gentoo-style diff files was worth looking into then I figured it would benefit the project more if you started thinking about it now rather than later!

  So how will you manage the modifications through the constant changing versions of MAME?  You're going to have to keep track of your changes somehow.  I figured you'd be using a diff repository anyway so why not make it public (in a sense)?  At least, that's how I initially came up with the idea.

  Rock out man. I support your project (as long as it doesn't slow things down )!

[Buddabing]

  So how will you manage the changing versions? 

CVS maybe?

[AllisterFiend]

  So how will you manage the changing versions? 

CVS maybe?

I've also noticed some people moving over to Subversion, they like it better then CVS, I have no clue which is better, but here is a link to read up on it if interested.

http://subversion.tigris.org/


Allister Fiend

[screaming]

  So how will you manage the changing versions? 

CVS maybe?

  I amended my original question to hopefully clarify what I meant.

[MikeQ]

  So how will you manage the changing versions? 

CVS maybe?

  I amended my original question to hopefully clarify what I meant.

Perforce.  I baseline each mame drop.  Make my changes.  Check each change in.  New version of Mame comes out, baseline it.  Merge .

I started this project on .102 and have merged 3 times since with little problem.  If a drastic rewrite of some component occurs, I'll have to do a drastic rewrite too.

I've got another developer that works with me coming on the project so we are going to be setting up a server so we are both working out of the same development tree.  Merging with his changes won't be any different than merging with the next rev of mame.   Although his and my changes will affect less of the overall source.

[RetroBorg]

I've only just come across this thread and haven't read everyone yet but has anyone suggested support for playing against other people on the net? e.g. Kaillera

[Popcorrin]

Then there is always force feedback and network linking for the games that supported it.

[Popcorrin]

MikeQ,  I thought I read a post awhile back that you were working on 3d support for mame.  Is that something you were still planning on or did you put that on the back burner?

[MikeQ]

MikeQ,  I thought I read a post awhile back that you were working on 3d support for mame.  Is that something you were still planning on or did you put that on the back burner?

Something I still want to do.  Need to get all the core features in that we want and then start working on it.

[Tiger-Heli]

MikeQ,  I thought I read a post awhile back that you were working on 3d support for mame.  Is that something you were still planning on or did you put that on the back burner?

Something I still want to do.  Need to get all the core features in that we want and then start working on it.

http://powermame.joymonkey.com/pmwiki.php?n=Features.ModernVidCard

But as Mike said, it's still kinda backburner

[arcadegeezer]

Hi All,

Just come across this thread, tonight and I am excited in what I read. Unfortunately the Homepage for Powermame shows site being updated so I cannot see all of the features that are currently planned or available.

Reading through the threads, am I right in thinking that this will also be a frontend ? or just a super version of MAME for use in a frontend ?

I am thinking about building a new USA style cocktail table and was wondering if a configurable frontend could be designed to rotate to the side of the controls that the user touches ?

i.e. specify how many sides/orientation the controls are fitted to on the Cocktail Table.

For instance:- 3 sided controls, vertical Orientation @ ends for verical games (screen flipping), horzontal orientation for horizontal games (2 players on same side.

Also the menu Choices could change or flip depending on the controls that were touched.

Show horizontal games when horizontal controls moved, etc.

Thanks 

Mark

[Tiger-Heli]

Hi All,
Just come across this thread, tonight and I am excited in what I read. Unfortunately the Homepage for Powermame shows site being updated so I cannot see all of the features that are currently planned or available.

http://powermame.joymonkey.com/pmwiki.php?n=Main.HomePage

Reading through the threads, am I right in thinking that this will also be a frontend ? or just a super version of MAME for use in a frontend ?

Yes!    The command line version is super for use in a frontend - the MAME32 version basically includes a frontend - just like regular MAME.

I am thinking about building a new USA style cocktail table and was wondering if a configurable frontend could be designed to rotate to the side of the controls that the user touches ?
i.e. specify how many sides/orientation the controls are fitted to on the Cocktail Table.
For instance:- 3 sided controls, vertical Orientation @ ends for verical games (screen flipping), horzontal orientation for horizontal games (2 players on same side.

That can pretty easily be done now with standard MAME.  See this thread: http://forum.arcadecontrols.com/index.php?topic=45609.0

Although I suppose it could be added to PowerMAME - hardest thing would be deciding whether it should rotate right or left for Player 1 on the Vertical games.

It would be up to MikeQ to decide whether or not to implement this, and I don't think it would be top priority for him.

IMHO - since it can be done with most any MAME build, I think a better solution would be for you (or Com[Plex] or SpyStyle or . . .) to come up with a small program that would ask:
1.  Input your MAME directory:
2.  Do you want the front-end to rotate also, or just the games?
3.  If #2 was Front-end also - are you using I-rotate, or Pivot Pro, or just the graphics HotKeys and input the keys you use.
4.  Should MAME start with a vertical game orientation or a horizontal orientation?
5.  Should MAME rotate CW or CCW for single player vertical games (if horizontal default) or horizontal games (if vertical default)?

Then once all this was inputted, the program would parse MAME and generate -ror or -rol entries for the vertical or horizontal games.

The good thing would be you could run this program with any version of MAME and it would set things up for you.  It should be pretty simple to code, although I couldn't figure out how to do it.   

Also the menu Choices could change or flip depending on the controls that were touched.
Show horizontal games when horizontal controls moved, etc.

That's a new idea that I hadn't heard before.  Tougher to implement, though!

[swindus]

I am thinking about building a new USA style cocktail table and was wondering if a configurable frontend could be designed to rotate to the side of the controls that the user touches ?

i.e. specify how many sides/orientation the controls are fitted to on the Cocktail Table.

For instance:- 3 sided controls, vertical Orientation @ ends for verical games (screen flipping), horzontal orientation for horizontal games (2 players on same side.

Also the menu Choices could change or flip depending on the controls that were touched.

Show horizontal games when horizontal controls moved, etc.

Thanks 

Mark

MaLa can rotate itself (and MAME if you want) and load a game list depending on the controls that were touched.
http://mala.arcadezentrum.com/controllerdetection.html

[Tiger-Heli]

MaLa can rotate itself (and MAME if you want) and load a game list depending on the controls that were touched.
http://mala.arcadezentrum.com/controllerdetection.html

Wow - that's pretty cool! 

[mr.Curmudgeon]

Wow. Hadn't seen MaLa before....(not that I've been trolling the software forum much lately)

Looking good.



mrC