I downloaded the latest nightly build (17.0.342, which incorrectly includes catapult v16 instead of v17, BTW)
That should be fixed now, Vampier told me.
So, apparently you had that TDC-600 extension combined with a non-matching openMSX... how did you do that?
I don't know, I just had it...I probably downloaded -at one point or another- an archive full of machine configurations and/or ROMs an it just stuck...
By the way, is there any way to speed up hardware checking in Catapult?
Yes, there would be a way, but I don't think anyone is interested to implement it (as no one is really interested to work on Catapult these days).
You can do the same check also in openMSX console: test_all_machines
and test_all_extensions
. Try it, it's quite fast.
The way Catapult is doing it now is by starting a new openMSX instance for each machine and extension. That makes it so slow. Catapult could also keep openMSX running and test the items with a command for each item, for instance.
Yes, there would be a way, but I don't think anyone is interested to implement it (as no one is really interested to work on Catapult these days).
You can do the same check also in openMSX console: test_all_machines
and test_all_extensions
. Try it, it's quite fast.
The way Catapult is doing it now is by starting a new openMSX instance for each machine and extension. That makes it so slow. Catapult could also keep openMSX running and test the items with a command for each item, for instance.
Or catapult could use these 2 commands, right? That would make Catapult’s job here quite simple…
Then you would need some kind of mechanism and parser to keep track of the output. In any case, you need a live connection.
I'm shocked about this, I had no idea this functionality existed, despite taking pride in being thorough with documentation. I'd like to say this is another nail in Catapult's coffin, but I actually prefer it since it uses standard system dialogs to open files.
I tried test_all_machines
and test_all_extensions
and I was quite pleased with its speed. @Manuel, correct me if I'm wrong, but does OpenMSX actually try connect each extension to the running machine? I spotted some weirdness from time to time (like OpenMSX complaining about a port conflict in the Video 9000 extension when I used a custom machine which already had an internal Video 9000, or complaining about an absent <secondary>
tag in the KB-7 extension when I used a machine which had an expanded slot 0, or the aforementioned KB-7 extension being disabled after reset when I used the Casio PV-7 together with it).
https://openmsx.org/manual/commands.html#test_machine is the documentation. As you can see here: https://github.com/openMSX/openMSX/blob/master/share/scripts... you can actually tell it which machine to use. By default the current machine is used.
The help
command tells you this as well.
Very nice, thank you. I see it has existed for almost a decade now. Since I've never used it before, I just ignored it. Kinda embarrassing.
I used the test_all_machines and test_all_extensions commands and, to my surprise (all machines/extensions pass the check in catapult), two extensions fail the check:
Casio_KB-7
Casio_KB-10
both say that "slot is already an external slot"
But, If I run a different machine (japanese MSX2+), then the error changes to:
Casio_KB-7 "Missing secondary tag for DeviceKB7 RAM" (sic).
Could it be because there's a machine already running that the check behaves weirdly?