72 replies to this topic

[niallquinn]

Cheers for the update TGA.

 

Or is it David H...........who knows....

 

Or Penry the mild-mannered janitor?  Could be.......

 

NQ-BF.

[tga]

more like penry the mild mannered janitor, before you ask questions over who i am, im an outsider. If you played counter-strike back in 1999-2005 you will have probs have seen me and remember the nick

been busy with a few things tonight, was hoping to spend a bit more time with mame but no joy yet.  i might as well use this thread as a reference guide for myself, i need to map the hardware int's to what chip/part of the board. 

 

notice when playing sphinx, the reels dont even make a full spin.  This is a little odd.  i still havnt had the time to dig through and remove all the mame drivers i dont need and recompile it....

to be honest , i should just do the ones i do need and compile it like that. still its a bit of a job.

[tga]

urghghghg things are very slow. hopefully going to sit down later and get something done.  Should be a new mame release soon hopefully, if the work ethic thing is resolved.

[niallquinn]

151 is out with all it's lovely relicenced code...

 

Let us know how you get on.  This really is great stuff.

[tga]

i had a quick look at .151 a few days ago and not much has changed relating to the electrocoin stuff. have noticed something interesting, every so many spins the reels actually line up fine, i got a funny feeling i might beable to multiply the reel spin  to this and get a quick result.  At the moment i need to move back to win 7 from 8.1 and plough on with my c# development, as thats pretty slow at the moment, define'ing int's , i can do that , understand debug info ( epic fail at moment ) looking at port emu and understanding the connections and how that works ( epic win ) , i need to focus more on the memory and dbug and also pulling errors from mame and investigating.  my general knowledge with c# is useless at the mo, and that's what is slowing this down horribly......

 

current todo list.

 

find stepper pattern, reflect in steppers or remake new stepper call, check pattern and spin accuracy.

find hardware calls to reel index or motor and determin final output and map to above..  check emu speed - ( running a bit quick ????? )

find a way to double/triple check all the above.

 

going to be a while before i make any progress...... :/ hopefully can sit down with it over the weekend.

[tga]

things have been grinding to a halt for me at the moment, will post something soon hopefully....

[niallquinn]

Thanks for the updat5e TGA, stick at it!

 

Cheers.

[tga]

well 3 months 2 late...... not had time to sit down with mame, uber uber busy :/ will look at new mame source tonight and hopefully see if any new changes added.  Still need to rebuild mame for fme to make dev'ing easier, still looking with squinty eyes at the electrocoin driver, made small progress but nothing to shout about....... will post more soon

[niallquinn]

Yep, thanks for the update, it's much appreciated.

[Daryl]

I do shudder at the thought of all those that supported the 'FME God' and are probably now thinking...

 

..."I told you so"!

[adb]

Not really many changes since .150

 

Think there may have been 1 or 2 stability improvements but still plays like crud & freezes for me

[tga]

Im still knocking about, list a few reasons why much hasnt been done.... no work, no bene's no nothing, so im spending all my time running around tryin to stay afloat.... :/

 

Mame has had some issues and rewrites for the layout code..... some of the versions 151 etc mean no layout apears..... and cause of the bugs , you cant add one unless you re-compile mame.  

 

Understanding the reel emulation is the biggist problem and probably why after the inital emulation it became difficult to complete,  most fm makers use there own reel setup, power distrabution to the reels, a reel index check, line up to start, then possibly hardware reel stop, and/or rotation check/modifier. maybe start / stop hardware if the reels are primitive.

 

tryin to pickup the code in mame and make it work has been difficult. its defo possible, but as mentioned my c/c++ coding is very basic at moment.

 

i wanted to do a seperate builf of mame for fme, but as mentioned the layout issues may or may not have been sorted. you will load say , sphinx and theres no layout and mame wont let you sue yours either......

 

without the reel accuracy theres no possible way to get the emulation working at any playable level , i have skinned a small sphinx layout , with feature board , but useless without semi complete emulation....

 

the reel pattern thats feeds the power and montiors the reel,s still unsure if theres hardware not mapped or correctly hooked up , 50% of the attempts i made to modify the reel pattern lookin for stability failed epicly......

 

i have a funny feeling

 

reel input power <> reel change <> reel stop is currently how it works, but maybe its more

reel input power <> reel change <> software check <> hardware change <> reel stop

 

but unsure......

 

im no fme god im afriad, just someone who wants to play spinx or pyramid etc lol

[tga]

going to pull the latest mame now and see if i can get a layout to apear

[tga]

ive had no luck with mameui64 0.152 and layouts, they dont work :/ nothing apears. other versions, .150 seem fine

[tga]

i dunno if theres a massive bug with layouts, or if theve been changed because i mentioned i was going to make a workin build of mame just for fme
, i dont think the code has changed from 150>152. but i could be wrong.  i'l try and investigate more.

[tga]

the steppers.c hasn't been modified since 2012

http://mamedev.org/s...steppers.c.html

since my eval copy of 8.1 ran out , im back on win 7 and need to install vc , and quite a few other bits and bobs to dev mame once again. So i'l crack on with that now..

[tga]

on gettin any cash win  on sphinx, the reels 1 + 3 go down , while 2 go's up.  This would have some bearing on why the nudges in mame could seem to make the reel go up or down randomly... but nothings random in mame.   Think the reverse call needs tobe looked at and debuged, this is easy as when collecting a win, the second reel should run in reverse.  making me think the reel input is giving clues to when the reel needs forward or reverse and i dont think this is relfected right in the steppers config. also as the reels are 16 16 16 16(number reel has features and a maybe a blank or 2), 

getting the forward / reverse calls accurate is now top priority, and then working out the reel pattern/input should be second. 

 

i think the video posted is for a sphinx v1, but unsure. 

 

also once the motor input is debuged and reflected properly , that should tie up the reels to be good.  I think whats happend is the steppers.c is processing incorrect reel reflections from the input and causeing mame to not have a clue whats going where. also i dont think these machines half step, rather a full step only...... i could be horribly wrong...........

 

looks like the big part of this is gonna be trial and error, if anyone wants to give me a hand please drop me a msg ro reply to this thread

 

TGA.

[tga]

one of the things i should do is take save states for .150,  so can save state on a win to look at motor input debug info, save state on feature board.  might be issues with spinx v2 and v1 , suttle changes to maybe the software calls and maybe hardware difference. dunno yet......