Which MSX1 has which VDP exactly?

Página 7/8
1 | 2 | 3 | 4 | 5 | 6 | | 8

Por ARTRAG

Enlighted (6828)

Imagen del ARTRAG

17-10-2014, 08:02

Por Manuel

Ascended (18704)

Imagen del Manuel

17-10-2014, 10:55

At least the test does not work in openMSX. The tiny timing difference which it is using is unknown/undocumented, so it is not emulated. openMSX always reports "no Toshiba" even on a Toshiba. And the next openMSX release will have the differences of the Toshiba implemented. This means software that will rely on this test will fail in openMSX for machines with a Toshiba VDP Sad

Por dioniso

Champion (479)

Imagen del dioniso

17-10-2014, 13:55

Manuel: "This means software that will rely on this test will fail in openMSX for machines with a Toshiba VDP".

Quite the opposite: openMSX will fail with this software.

Por PingPong

Prophet (3885)

Imagen del PingPong

17-10-2014, 15:40

It does appear that this kind of detection is somewhat too much "fragile" or prone to errors to be reliable.

Por dioniso

Champion (479)

Imagen del dioniso

17-10-2014, 17:11

PingPong, could you please post the results of your tests? It would be nice if you could write the name of the MSX1 computers you have been testing the ROM with: name of the MSX1 computer, VDP (TMSxxxx, T6950), and result of the ROM.

Por PingPong

Prophet (3885)

Imagen del PingPong

17-10-2014, 21:18

dioniso wrote:

PingPong, could you please post the results of your tests? It would be nice if you could write the name of the MSX1 computers you have been testing the ROM with: name of the MSX1 computer, VDP (TMSxxxx, T6950), and result of the ROM.

to do this i need a wav version of the sw, my machine does not have any other interface than cassette tape. i do not know if your rom works with loadrom or similar tool
my machine is a sony hb-10p equipped with the infamous toshiba clone. (pal)

Por dioniso

Champion (479)

Imagen del dioniso

17-10-2014, 21:46

I can't help you with that. Don't you know anyone with a Megaflashrom or any other device like that who lives in your city? I guess that's pretty difficult.

Anyways, I have tried the ROM in that very same computer (I have one myself) and the detection works flawlessly.

Por ARTRAG

Enlighted (6828)

Imagen del ARTRAG

18-10-2014, 09:21

l_oliveira wrote:

I did, yesterday, when it was suggested that Uridium could be used as a VDP benchmarking and that it was unlikely it would work on NTSC TI VDPs:

TMS9128:

TMS9928:

It's very curious that only sprites did glitch ...

Don't feel bad as for example, this is what happens when you try Guru Logic on a NTSC MSX1:

If I put the original PAL VDP back on the machine it looks like this:

For that kind of thing, Open MSX is starting to work towards emulating chip behavior down to access timing. Aiming to be a useful reference for development.

Edit:
Oh and of course, if I put the PAL VDP back:

@l_oliveira
About the free version of uridium, the glitch you've spotted depends on fast I/O: it appears only on NTSC msx1 machines because the vertical border at 60Hz lasts less time and that sprite update is done in the ISR.
In the cart version of the game we fixed in the problem by changing the order the ISR does things getting critical VDP I/O always during the tracing of the border.

So the glitch was not related with the use of undocumented features or with Toshiba/TMS incompatibilities.

Por Manuel

Ascended (18704)

Imagen del Manuel

21-10-2014, 22:51

dioniso wrote:

Manuel: "This means software that will rely on this test will fail in openMSX for machines with a Toshiba VDP".

Quite the opposite: openMSX will fail with this software.

Yes, if openMSX fails with this software, it means the software will fail in openMSX. That's quite the same isn't it? Smile

Anyway, I'm also very interested whether it works reliably on real machines. When I have some more time I will try it on some of my own.

Por dioniso

Champion (479)

Imagen del dioniso

21-10-2014, 23:25

Manuel wrote:
dioniso wrote:

Manuel: "This means software that will rely on this test will fail in openMSX for machines with a Toshiba VDP".

Quite the opposite: openMSX will fail with this software.

Yes, if openMSX fails with this software, it means the software will fail in openMSX. That's quite the same isn't it? Smile

Anyway, I'm also very interested whether it works reliably on real machines. When I have some more time I will try it on some of my own.

It's not the same. A MSX cannot fail; an emulator, yes. It's good that emulators aim for perfection Wink and I would like to see the day a PC or MAC runs all programs like in an MSX (no irony, I really mean it). But in the meantime, it makes more sense to improve emulators to behave like an MSX than to "repair" all MSX to behave like emulators. That's what I ment with "openMSX will fail with this software" and not "the software [...] will fail in OpenMSX", because as long as the software works in an MSX, the "problem" is with the emulator, and not with the software or the MSX computers.

Thanks for wanting to try the detection routine, Manuel. Up to now, I only have had the results of two MSX1 computers ... Quite poor. Whenever you have done it, please, tell me the results.

Página 7/8
1 | 2 | 3 | 4 | 5 | 6 | | 8