Roman,

Just thought I have something nice here for when you get back! :)

I was just going through, re-scanning my sets, when I came across a strange error. After playing around, trying to replicate the problem, I came up with the following sequence :

If I scan my Testdrivers set (dat file version 20030904), and then I try & use the 'official mame32 cabinets', 'icons' or 'screenshots' (from John IV's site), I get the following message :
------------------------------------------------------
ROM-Path :

D:\Program Files\Mame32\cabinets\

matches a setname.

This is mostly done by mistake. Normally you have a rompath which includes setfolders or setzipfiles but it doesn't point to a setfolder.

It's recommended to change that romfolder !
------------------------------------------------------

Then, when I click OK, it comes up with all of the BIOS files from the Mame set (decocass.zip, neogeo.zip, etc) as having problems & missing files.

I can replicate this every time by doing the following :
- Load the 'MAME Test Drivers v0.73' datfile.
- Go straight back to the Profiler & load the 'official mame32 cabinets' datfile.
- Run a Scan & get the error.

When it happens, if I load one of the other datfiles (the 'icons' or 'screenshots'), I get the same problem.

If I load another datfile (in this case, I am using the 'MAME Samples 20030904' file from your site), then go back & load the datfile I was having problems with, the problem goes away, and the set scans without a problem.

I'm sure it's just something simple, like not clearing the list of BIOS Sets under certain conditions, but I hope this information helps you to fix the problem.

Thanks, and I hope you had a good break!

> If I scan my Testdrivers set (dat file version 20030904), and then I try & use
> the 'official mame32 cabinets', 'icons' or 'screenshots' (from John IV's site),
> I get the following message :
> ------------------------------------------------------
> ROM-Path :
>
> D:\Program Files\Mame32\cabinets\
>
> matches a setname.

Why do you use a rompath (...cabinets) when you scan testdrivers? You should use rompaths carefully. Only add rompaths which are used for the current profile. And in these rompaths you should only have files which belong to the currently used profile.



Roman Scherzer
ClrMamePro

> Why do you use a rompath (...cabinets) when you scan testdrivers? You should use
> rompaths carefully. Only add rompaths which are used for the current profile.
> And in these rompaths you should only have files which belong to the currently
> used profile.
>
>
>
> Roman Scherzer
> ClrMamePro
>

Sorry, but this is what I said :

If I scan my Testdrivers set (dat file version 20030904), and then I try & use the 'official mame32 cabinets' ......

I got that message when scanning the cabinets set, NOT the testdrivers. If you look at the sequence I outlined, you will see that if I load the 'Testdrivers' datfile, THEN go back to the Profiler and load the 'cabinets' datfile, then run a scan, that is when I get the message.

All of my paths in all of my profiles are correct (i.e. only the '\cabinets' dir for the cabinets datfile), so I'm sorry if you misunderstood!

I know it was a long message, but I wanted to put as much detail in it to help you find the error. :)

Edit : BTW, welcome back, hope you had a good holiday!

This was probably the most important part of the message :

I can replicate this every time by doing the following :
- Load the 'MAME Test Drivers v0.73' datfile.
- Go straight back to the Profiler & load the 'official mame32 cabinets' datfile.
- Run a Scan & get the error.

> - Load the 'MAME Test Drivers v0.73' datfile.
> - Go straight back to the Profiler & load the 'official mame32 cabinets'
> datfile.
> - Run a Scan & get the error.

Nope. Works fine here. Can't repeat your mentioned issue.
Take a look at the belonging .cmp files in cmpro's settings folder and check if there are wrong rompathentries in there. Or create new profiles.

And be sure you're using correct rompaths for your cabinets profile:

If you're using ZIPPED cabinet files, use:
D:\Program Files\Mame32\cabinets

If you're using UNZIPPED cabinet files, use:
D:\Program Files\Mame32


Roman Scherzer
ClrMamePro

Sorry Roman, it's definatly NOT a settings issue.

I was pretty sure the sequence I outlined came up with the error every time, but I was just trying it again, and I think there is more to it.

If I do the following EXACTLY, I get the message every time :

- Open ClrMamePro (so obviously make sure it isn't open to start with)
- Load the 'official mame32 cabinets' datfile
- Close ClrMamePro immediatly
- Open ClrMamePro again
- Load the 'MAME Test Drivers v0.74' datfile
- Go back to the Profiler & load the 'official mame32 icons' datfile
- Run a Scan

Doing this, I get the error message EVERY time. If I then go back to the Profiler and load another 'official mame32 *' datfile, I get the same message.

To fix it, I do the following :

- Go back to the Profiler
- Load the 'Modeler v0.9.3a' datfile
- Go back to the Profiler
- Load the datfile I got the message on (so, using my example, I would load the 'official mame32 icons' datfile)
- Run a Scan

I no longer get the error message.

That's what makes me think that it is a problem with not clearing the BIOS list in memory, as the Modeler datfile doesn't have BIOS sets.

If I do the entire process from the beginning again, but load the 'Nebula v2.23c' datfile instead of the 'MAME Test Drivers v0.74' datfile, I get the same error message, but only get 'neogeo.zip' & 'pgmbios.zip' files showing as missing, instead of all of the Mame BIOS's, as happens with the Test Drivers file.

I hope that this helps! If you need me to try anything else, or want me to clarify anything, please let me know.

Edit : Should have mentioned that I DID check the .cmp files for anything funny, but couldn't find anything. All of my paths are 100% correct.

I also deleted the datfile from the Profiler screen, and made sure any associated file for that datfile was gone. I then re-downloaded the datfiles from your site and tried again, but still get the message following the sequence above.

I just deleted the datfiles again, re-downloaded them & the problem has gone away for now.

It's a wierd one, for sure!

It's got to be some sequence of events that STARTED it initially, and then did not fix itself until just now.

Rest assured though, it's not a problem with the way I have set the program up. The paths I have setup are not in error, I have checked them 15 times while looking at this issue.

Don't worry about it, it's just as a software developer myself, I hate having little problems like that in my programs. The harder to find, the more satisfying it is when you squash them! :)

It wasn't a huge issue, just thought I would let you know about it. Since I can't replicate it 100% of the time, there's nothing you can do.

I will do some further investigations next week.

Roman Scherzer
ClrMamePro

> I will do some further investigations next week.
>
> Roman Scherzer
> ClrMamePro
>
Aha! I think I have found the problem (where have I heard THAT before?!?!)

I just started getting the message again, and I thought I would look at the settings files in each step.

When I load the 'official mame32 icons' datfile (after loading the 'test drivers' file & going back to the Profiler), the following line is written to the .cmp file (as soon as the datfile is loaded) :

Misc_BIOSSets = decocass|*|1|pgm|*|1|playch10|*|1|stvbios|*|1|konamigx|*|1|tps|*|1|taitofx1|*|1|cpzn1|*|1|cpzn2|*|1

If I go back to the Profiler & load the Modeler datfile, then go back & load the icon datfile again, the line in the .cmp file read as :

Misc_BIOSSets =

If I then load the Nebula v2.23c datfile, then load the icon datfile, the line is changed to :

Misc_BIOSSets = neogeo|*|1|pgmbios|*|1

Load the Modeler file, then the icon file again, and the line is clear again.

(the .cmp file in question is 'j4_rc_icon.cmp' in the 'settings' folder)

All of these changes are made WITHOUT ever touching any settings. All I was doing while testing this was to open ClrMamePro, load a datfile, go back to the Profiler, load another one, etc. & checking the .cmp file at each step (I use UltraEdit, so it would tell me if the file had changed & if I wanted to reload it)

Interestingly, this does not occur when I use the internally created Mame datfile, only seems to be the additional files that contain BIOS's.

Here's hoping that this info leads you in the right direction!

The settings of the currently loaded profile are saved each time you visit the profiler or quit clrmamepro.
Thanks for the "how to repeat the problem". I will take a look at it next weekend.


Roman Scherzer
ClrMamePro

> The settings of the currently loaded profile are saved each time you visit the
> profiler or quit clrmamepro.
> Thanks for the "how to repeat the problem". I will take a look at it next
> weekend.
>
>
> Roman Scherzer
> ClrMamePro
>
Thanks Roman.

So you are saying that the settings shouldn't be changed when you are simply loading the existing datfile? It's doing it at this point, I'm 100% positive of that.

> So you are saying that the settings shouldn't be changed when you are simply
> loading the existing datfile? It's doing it at this point, I'm 100% positive of
> that.

Of course it shouldn't change the settings. right. It should simply rewrite the currently loaded settings....If it writes something else...it may be a bug. Well...thanks for letting me know....I won't be able to code before next sunday....guess you can expect a release then ;)

Roman Scherzer
ClrMamePro