Interesting But Unwanted Behavior: When I was playing Viper Phase One on .172 I got a high score... I exited then entered and noted that the high score said 0...
I then went in to .251 Viper Phase 1 to get the high score there to see if it saved and guess what... my high score was there.
It seems that .172 is SAVING high scores to the same file that .251 and .269 are pointing to, but LOOKING at a different highscore file that is NOT getting updated.
Also of note: Cave games are just about unplayable on .269. They are quite playable on .172.
ARpiCADE_5 - for Pi4 or Pi5
Moderator: dee2eR
-
FrizzleFried
- Legendary Contributor
- Posts: 522
- Joined: Sat Nov 25, 2017 5:35 pm
- Location: Idaho
- Contact:
Re: ARpiCADE_5 - for Pi4 or Pi5
Visit my arcade blog ... www.idahogaragecade.com (Updated: 10/28/21)
Re: ARpiCADE_5 - for Pi4 or Pi5
Interesting. I guess that's because 172 is running the 176backports patch by ozfalcon, but it wasn't doing that before I set up the later MAME hiscores though, the hiscore folder wasn't there... possibly fixable with some ini editing - I'll look into it.
-
cool_factor
- Legendary Contributor
- Posts: 409
- Joined: Tue Sep 26, 2017 10:55 am
- Location: Fort Worth TX
Re: ARpiCADE_5 - for Pi4 or Pi5
Just a thought we may not want to hear but maybe just make version 5 with 172 as the highest version. Seems to be the most stable. Even in version 4 the majority of my ROMS were in 172, very few in 203 or 230. 106 had about 5. I've got about 200 ROMS on that build.
-
FrizzleFried
- Legendary Contributor
- Posts: 522
- Joined: Sat Nov 25, 2017 5:35 pm
- Location: Idaho
- Contact:
Re: ARpiCADE_5 - for Pi4 or Pi5
... not sure if there is a "substantial" update between .172 and .251... like .184 (for example... just a number I pulled out of my hat) that would be more appropriate with this CPU than .269+? My thoughts are back to running primarily .172 with few games running later versions as necessary (versus running primarily .269 and running older version for some games).
Visit my arcade blog ... www.idahogaragecade.com (Updated: 10/28/21)
-
FrizzleFried
- Legendary Contributor
- Posts: 522
- Joined: Sat Nov 25, 2017 5:35 pm
- Location: Idaho
- Contact:
Re: ARpiCADE_5 - for Pi4 or Pi5
What is really weird is that .172 is supposed to use a different VERSION of highscores.dat file. Remember?
I am going to go back out there in a minute to DOUBLE CHECK and make sure I am 100% correct in that when I make a high score in .172 it's saving to the high score dat file that .251 and .269 are using because that just seems... improbable.
Visit my arcade blog ... www.idahogaragecade.com (Updated: 10/28/21)
-
FrizzleFried
- Legendary Contributor
- Posts: 522
- Joined: Sat Nov 25, 2017 5:35 pm
- Location: Idaho
- Contact:
Re: ARpiCADE_5 - for Pi4 or Pi5
OK... a bit of an update.
First off, I was incorrect... Version .172 MAME is, in fact, saving and DISPLAYING high scores. You have to coin-up and start playing before the high score appears... it does not do so on attract at the beginning. My apologies for the confusion but we did discover something...
.172 ... somehow ... is looking at the same highscore.dat as .251... because .251 displays the high score I made in .172 as well! I need to go out and put a high score in on .251 to confirm that .172 is READING the same file though... that's next.
Now for an oddity that has only one answer that I can think of... .269 NO CLONES (list made from Italia Arcade website) must use a different ROM because this ROM made me go through the whole 999 -ticks UPDATE at the beginning thing... and no high scores at all. Since I already know that .251 and .269 are sharing this has to be one of the other of TWELVE total roms to choose from...
As mentioned above though... .172 uses a different highscore.dat file and I worry it's going to corrupt this newer version if we continue to use it with .172??
FWIW, moving to .174 is an option... as it's compatible with the new HS file.
First off, I was incorrect... Version .172 MAME is, in fact, saving and DISPLAYING high scores. You have to coin-up and start playing before the high score appears... it does not do so on attract at the beginning. My apologies for the confusion but we did discover something...
.172 ... somehow ... is looking at the same highscore.dat as .251... because .251 displays the high score I made in .172 as well! I need to go out and put a high score in on .251 to confirm that .172 is READING the same file though... that's next.
Now for an oddity that has only one answer that I can think of... .269 NO CLONES (list made from Italia Arcade website) must use a different ROM because this ROM made me go through the whole 999 -ticks UPDATE at the beginning thing... and no high scores at all. Since I already know that .251 and .269 are sharing this has to be one of the other of TWELVE total roms to choose from...
As mentioned above though... .172 uses a different highscore.dat file and I worry it's going to corrupt this newer version if we continue to use it with .172??
FWIW, moving to .174 is an option... as it's compatible with the new HS file.
Visit my arcade blog ... www.idahogaragecade.com (Updated: 10/28/21)
-
FrizzleFried
- Legendary Contributor
- Posts: 522
- Joined: Sat Nov 25, 2017 5:35 pm
- Location: Idaho
- Contact:
Re: ARpiCADE_5 - for Pi4 or Pi5
OK... first off I can confirm that .172 and .251 are sharing the same HS for sure. Scores made on each are seen on each and vice-versa.
The behavior of this ROM is that upon load-up it counts down from 999 to 0 for an "update"... it takes about 5 minutes or so... When I initially loaded .172 ... it counted down and loaded and ran and I got the high scores. I then went to .251 and noted that it did NOT go through the update... which I found odd. It was then that I realized that the high scores from .172 appeared on .251.
.269 is a strange one. I just went through all 12 ROMs and NONE of them had the UPDATE applied. Also, the NOCLONES ROM (assuming it's the parent... and it DOES have the same name and tags (world) and what not)... it didn't have the update applied. I let it run through the 999 ticks. The high scores aren't there?!?
I thought I'd verified on other games (Strikers) that .251 and .269 indeed shared highscore.dat... but?
EDIT: Time to check my edits.
The only thing I note is that .251 and .269 point to .mame (.251) as a homepath whereas .172 has no homepath but does mention the same path as inipath.
To my layman eyes it looks like the plugin path should be fine but what do I know?
The behavior of this ROM is that upon load-up it counts down from 999 to 0 for an "update"... it takes about 5 minutes or so... When I initially loaded .172 ... it counted down and loaded and ran and I got the high scores. I then went to .251 and noted that it did NOT go through the update... which I found odd. It was then that I realized that the high scores from .172 appeared on .251.
.269 is a strange one. I just went through all 12 ROMs and NONE of them had the UPDATE applied. Also, the NOCLONES ROM (assuming it's the parent... and it DOES have the same name and tags (world) and what not)... it didn't have the update applied. I let it run through the 999 ticks. The high scores aren't there?!?
I thought I'd verified on other games (Strikers) that .251 and .269 indeed shared highscore.dat... but?
EDIT: Time to check my edits.
The only thing I note is that .251 and .269 point to .mame (.251) as a homepath whereas .172 has no homepath but does mention the same path as inipath.
To my layman eyes it looks like the plugin path should be fine but what do I know?
Visit my arcade blog ... www.idahogaragecade.com (Updated: 10/28/21)
-
FrizzleFried
- Legendary Contributor
- Posts: 522
- Joined: Sat Nov 25, 2017 5:35 pm
- Location: Idaho
- Contact:
Re: ARpiCADE_5 - for Pi4 or Pi5
OK... ready for this one Dee2er?
Running .172 I load up Strikers 1942. It has it's own High scores. I verify it. It is a separate high score list from .251 and .269 which do share the same high score board. I verified all this.
Then I move over to Viper Phase 1... running .172 it SHARES the high score board with .251... and .269 and .251 DO NOT share high score boards?!?! (at least not the parent roms).
What... The... F....?? Um...
How does that work exactly?
EDIT: I've tried 5 or 6 random older and newer games thus far and none have behaved the same as Viper ... all the way Striker does...
Running .172 I load up Strikers 1942. It has it's own High scores. I verify it. It is a separate high score list from .251 and .269 which do share the same high score board. I verified all this.
Then I move over to Viper Phase 1... running .172 it SHARES the high score board with .251... and .269 and .251 DO NOT share high score boards?!?! (at least not the parent roms).
What... The... F....?? Um...
How does that work exactly?
EDIT: I've tried 5 or 6 random older and newer games thus far and none have behaved the same as Viper ... all the way Striker does...
Visit my arcade blog ... www.idahogaragecade.com (Updated: 10/28/21)
Re: ARpiCADE_5 - for Pi4 or Pi5
Snow Bros definately shows different high scores in 172 compared to 251/269. A SnowBros JAMMA board does not remember high scores when powered down so it is definately using MAMEs score saving system, I suspect this may be the origin of the difference in behaviours you're seeing...
Currently waiting the 5 minutes for Viper Phase 1 in 251... (sounds like it's not quite full speed and would run better in 172 anyway... EDIT: actually once the 999 timer finished it sounds full speed) - I don't know if VP1 saved high scores on the original board...
It's possible games that originally remembered high scores when powered off behave different to those that are using MAME to remember high scores... that's about all I can come up with to explain this behaviour, at least while I'm waiting for 999 seconds...
--------
unrelated to that: compiling day went poorly but I did encounter a way to hide more of MAMEs pop up boxes in 251 and 269 with a couple of lines added to the bottom of their respective ini files (I think they work in conjunction with the skip_gameinfo line):
skip_disclaimer 1
skip_validitychecks 1
not perfect but may help until I get the compiling issues sorted.
-------
More EDIT: I thought maybe 269 isn't saving everything properly, as it would also explain why 251 remembers you've read warnings and 269 doesn't... Nope, both are saving NVRAM and cfg to their own folders... this is weird.
Currently waiting the 5 minutes for Viper Phase 1 in 251... (sounds like it's not quite full speed and would run better in 172 anyway... EDIT: actually once the 999 timer finished it sounds full speed) - I don't know if VP1 saved high scores on the original board...
It's possible games that originally remembered high scores when powered off behave different to those that are using MAME to remember high scores... that's about all I can come up with to explain this behaviour, at least while I'm waiting for 999 seconds...
--------
unrelated to that: compiling day went poorly but I did encounter a way to hide more of MAMEs pop up boxes in 251 and 269 with a couple of lines added to the bottom of their respective ini files (I think they work in conjunction with the skip_gameinfo line):
skip_disclaimer 1
skip_validitychecks 1
not perfect but may help until I get the compiling issues sorted.
-------
More EDIT: I thought maybe 269 isn't saving everything properly, as it would also explain why 251 remembers you've read warnings and 269 doesn't... Nope, both are saving NVRAM and cfg to their own folders... this is weird.
Re: ARpiCADE_5 - for Pi4 or Pi5
I set a VP1 score in 269, then loaded it in 251 (noscore, I think - this is starting to make me feel a bit crazy...), then tried 172 (not scaled properly on hori, couldn't see the score - will look into that later if I remember), hit the input issue and rebooted, reloaded VP1 in 172 to check the scaling, loaded it again in 251 and the score I set in 269 was there... WTF
Swappped to vert ror to check 172 again, no high score. 251 again, high score. 269 again, same high score. Reloaded 172 with the intention of setting a score to test but failed due to rotated monitor/joystick mismatch hand/eye confusion (I'll try again, it's just one level...).
IDK what's going on but it seems to be working as expected for me here... I used the same rom (in the same directory) to check each MAME, just swapping which MAME launched it using the 'e' edit menu from Attract-Mode. Pretty sure it's the parent rom as I only have it in my list once, but I didn't check.
EDIT: 172 score set. reload in 251 and the score is from... 172! Try 269 again and score is from 269. Facinating.
MORE EDIT: figured out whats happening. MAME 172 is saving NVRAM in .mame, which is loaded by 251... I guess emulated NVRAM overrides .hi files hiscores(?). I've changed the paths in mame172/mame.ini to:
cfg_directory $HOME/mame172/cfg
nvram_directory $HOME/mame172/nvram
input_directory $HOME/mame172/inp
state_directory $HOME/mame172/sta
snapshot_directory $HOME/mame172/snap
diff_directory $HOME/mame172/diff
hopefully that will avoid it happening. Testing it now in 251, have to wait 999 again due to deleting the NVRAM...
AND AGAIN: yep, that fixed it. 172 score is different from the shared 251/269 hiscore in VP1. I wonder how important it is to have 172 do it's own thing, conversly would we be better off if 251 and 269 were fully separated as well? For now I'm going to leave it as is, maybe any answer will present itself with time.
Swappped to vert ror to check 172 again, no high score. 251 again, high score. 269 again, same high score. Reloaded 172 with the intention of setting a score to test but failed due to rotated monitor/joystick mismatch hand/eye confusion (I'll try again, it's just one level...).
IDK what's going on but it seems to be working as expected for me here... I used the same rom (in the same directory) to check each MAME, just swapping which MAME launched it using the 'e' edit menu from Attract-Mode. Pretty sure it's the parent rom as I only have it in my list once, but I didn't check.
EDIT: 172 score set. reload in 251 and the score is from... 172! Try 269 again and score is from 269. Facinating.
MORE EDIT: figured out whats happening. MAME 172 is saving NVRAM in .mame, which is loaded by 251... I guess emulated NVRAM overrides .hi files hiscores(?). I've changed the paths in mame172/mame.ini to:
cfg_directory $HOME/mame172/cfg
nvram_directory $HOME/mame172/nvram
input_directory $HOME/mame172/inp
state_directory $HOME/mame172/sta
snapshot_directory $HOME/mame172/snap
diff_directory $HOME/mame172/diff
hopefully that will avoid it happening. Testing it now in 251, have to wait 999 again due to deleting the NVRAM...
AND AGAIN: yep, that fixed it. 172 score is different from the shared 251/269 hiscore in VP1. I wonder how important it is to have 172 do it's own thing, conversly would we be better off if 251 and 269 were fully separated as well? For now I'm going to leave it as is, maybe any answer will present itself with time.
