I now have a preliminary version of MAME 35 Final, where the major flaw I referred to is now impossible to do. The file format of the INP file has been alterred, the INP files created with this version will only playback with this version.

So Chris, I'll email you this version if you like. Please refer to this version as MAME 35 TG version, or 35TG if you like... if you could add this to your script that would be nice :-)

I'd like to see Ben Jos play this version for Donkey Kong, and German to play Frogger also, only to show the nay-sayers that their games were for real. I'd also like to see Phil Lamat play League Bowling and get a perfect game with this version... only to confirm their great skills are for real.

I'll set up a link to my web page tonight with the new MAME35TG version. This is only the beginning, later on I'll add a zip file with all the cfg files set to TG settings.

Mark

-- Mark Longridge (cubeman@iname.com), October 30, 1999

Answers

Good to see Mark and now perhaps this version could be approved, for those who will be participating, for use in T2. Naturally Mame32 0.35 Final for those who have trouble running the DOS version of Mame such as I do on my Windows NT at work - only when I'm on my lunch break naturally !!!!! ;-p

Good luck to all T2 participants and remember it's not only about winning it's also about having fun and building from other peoples playing styles to achieve new PB's on all these games. Everyone who achieves PB's is a winner in their own right if not the winner of the tournament.

BeeJay.

-- BeeJay (brianjohnstone@cardinal.co.nz), October 30, 1999.

I actually used MAME32 .35 final (the Windows version, which doesn't have the flaw) for my Donkey Kong recording, but I'd be happy to make a recording using the version you just made if you want me to. What's the minimum score I should get? :-)

A minor point regarding the name for your version: The latest readme.txt (for 36b8) states:

(Start of quote) VII. Derivative Works --------------------- Derivative works are allowed (provided source code is available), but discouraged: MAME is a project continuously evolving, and you should, in your best interest, submit your contributions to the development team, so that they are integrated in the main distribution.

There are a certain number of drivers in the source which are disabled in the official distribution. Enabling them is NOT considered a derivative work, and distribution of executables with those drivers enabled is strictly forbidden.

If you make a derivative work, you are not allowed to call it MAME. You must use a different name to make clear that it isn't an official distribution from the MAME team. (End of quote)

Ben Jos.

-- Ben Jos Walbeehm (walbeehm@walbeehm.com), October 30, 1999.

Actually, I just re-read the quote and you may have a problem because of the "(provided the source code is available)" part...

Ben Jos.

-- Ben Jos Walbeehm (walbeehm@walbeehm.com), October 30, 1999.

Yes, of course, and I knew you would see that immediately. But I think this is a special case... now if anyone thinks *I'M* cheating the only way I can prove myself is inviting all the players down to Funspot in New Hampshire to watch me play!

But as for the MAME aggrement about releasing the source code, I think this is a worthy exception, and also the new clause with "derivative works" is new to MAME368.

Don't you think it's better if the oportunity to cheat is decreased as much as possible? This is good way to do it.

Mark

-- Mark Longridge (cubeman@iname.com), October 30, 1999.

OK. No problem. I'll record a game using the version you created. I already downloaded it, and I noticed that some of the sound is missing (most notably, Mario's "quacking" noise when he walks), but that shouldn't matter. I just hope you're not expecting me to score 893400 or even more. And I'll try to squeeze it in soon, but I can't promise anything since the MARP tournament is starting in less than a day. I had not planned on playing any Donkey Kong during that tournament, EXCEPT if someone were to beat any of my CK/DK scores on MARP. So I can't guarantee that I'll send you anything soon, but I'll try.

As for the MAME agreement, I personally have no problem at all with you keeping the source code for yourself, but maybe it's a good idea if you at least let the MAME team know about it. And, yes, I fully agree with you that a special version of MAME that is more cheat-proof than the regular version is a valuable asset.

Ben Jos.

-- Ben Jos Walbeehm (walbeehm@walbeehm.com), October 30, 1999.

I think derivative work pertains to a new project based on the mame code. Mess for instance.

I think this exception could fall under modifying for personnel use. Unless its being distributed as an official marp version, then the source may have to be released.

-- Dave Kaupp (info@kaupp.cx), October 30, 1999.

I can't come up with any suggestions that aren't illegal, impossible, and/or downright disrespectful on how to get the need for MAMETG to NOT be open source across to the MAME Team.

Aqua

-- Aquatarkus (aquatarkus@digicron.com), October 31, 1999.

There is also no way to verify someone hasn't compiled there own version of mame34 with the re-playback hack put in. Nor if they played it on an ultraslow machine or other methods of cheating.

The only way I can think of truely verifying a recording is using a camcorder.

-- Dave Kaupp (info@kaupp.cx), October 31, 1999.

Aqua: I don't think there is any point in releasing the source for this "TG" version of MAME35 final. All I changed was a few lines of code which makes cheating a lot less likely, and I modified the INP file format a bit. If I distribute the source of this minor patch then the whole point of patching the source is negated, and the ability to EASILY cheat is once again a problem.

Dave Kaupp: Sure, you could recompile mame34 and cheat your brains out... but with mame35tg you can't do it and you don't know what I changed in the inp file format to make it incompatible with ALL other versions of MAME. Now it's pretty tough to cheat, which is the point of the exercise. May as well make cheating as difficult as possible, don't you think?

Mark

-- Mark Longridge (cubeman@iname.com), October 31, 1999.

I agree that a MARP or a TG version of Mame is a good thing, it inhibits some of the flaws in Mame that someone could use to manipulate a recording.

I'm just pointing out that if someone wants to cheat, they will find a way to cheat.

The only way to truly verify a recording is by visual inspection, everything else is just a deterrent.

-- Dave Kaupp (info@kaupp.cx), October 31, 1999.

Well you can nearly guarantee that MAME32 35 FINAL or TG MAME 35 FINAL or however you want to call it will be the version we use in the next tournament. I think all of us who are honest players at least appreciate what you have done Mark, thanks :)

Oh yeah... and if the MAMEDEV team gives you any problems with that source code junk, I have a great feeling that quite a few of us, including myself, will step up and defend. :)

-- Gameboy9 (goldengameboy@yahoo.com), October 31, 1999.

Mark, thanks for taking the initiative and coming up with a tourney version of MAME! As I told you last week, working with the MAME development team is extermely challenging as they are unresponsive to excellent suggestions from MAME users. They really need to start paying attention more to the program itself rather than just trying to crank more games out.

Anyway, since you did not mention this, I was wondering if you addressed my (and yours too!) #2 concern which is the infamous PAUSE key. I'm hoping that your final version disables this feature since we'd want the tourney version to closely mimic and emulate the way actual machines operate in an arcade setting.

-- Pat (laffaye@ibm.net), October 31, 1999.

First I'm happy someone else besides me is trying to make a secure mame, thanks Mark for starting the ball rolling.

Now, You can't fault the MAME team at all for not responding to the MARP people, their job is NOT to serve our desires. I think they worry about and improve the program tons more than we can ever thank them for. This particular problem we face (to debunk forged inp files) is an incredibly difficult problem in cryptology, and it is our communities responsibility to make right.

For our problem, you need to pass information between a program run by user A through to user B (the inp authority) with out user A knowing what the program is doing. A virtual impossiblity since A can see everything the program consists of since it must exists on A's hard drive. I am hopeing Mark has thought of everything in his version because I will try to hack out his modification of the inp file to see if anyone else can too (i'm pretty lame at hacking so if i can do it anyone else could.)

I encourage others (better than me) to attempt to cheat with Marks program, and if they are successful in doing so mail him how they did it. Alhtough, you can see i don't have faith that Mark's first version will be hack proof, I indeed hope we can eventually make something that no one can cheat with.

-- Chad (churritz@cts.com), October 31, 1999.

I have to sincerely thank Mark for providing this "Competition Release" of Mame !!! (BTW, where's the URL !?!?)

As Pat suggested, if the PAUSE key could be desabled, at least while recording, well....the job will be complete ...

Thanx again, Mark.

Cicca

-- Cicca (cicca@writeme.com), November 02, 1999.