Raven060 (luciodra build).

[luciodra]

I purchased a network card on eBay that now seems to respond to incoming and outgoing pings. I don't know if I'll be able to surf sooner or later...
The problem I can't solve is that Raven is always unstable. Sometimes even moving a window or opening a menu in Emutos. I seem to find solutions but when I turn it off and on again everything goes back to the way it was before. But does this happen to anyone else even occasionally?
I keep thinking about the MC68150FN40 and the video card...
What I would like to understand: is the instability due to the hardware, the immature software or both?

[luciodra]

I would like to find emutos over serial connection to try the card with the bare minimum.

raven_emutos3.png

[agranlund]

I would like to find emutos over serial connection to try the card with the bare minimum.

Try this one:

emucon-raven.rom.zip

(I have not tested the serial-only build in a while, hopefully still works)

[luciodra]

Try this one:
emucon-raven.rom.zip

(I have not tested the serial-only build in a while, hopefully still works)

In these last hours I'm making another attempt and as I tell you I'm crossing my fingers... I replaced the MC68150FN40 with the new Freescale 68150FN33 that I had. I left only the video card and placed it in slot 4. On Emutos it seems to work quite well, and I can play the games you posted. Every now and then the screen freezes and I have to turn it off and on again, but less than before. If I also place the network card (which I purchased and it makes pings) the system becomes noticeably more unstable :(
There is one thing that is not clear to me: but is isa_bios.prg still necessary with the latest release you published?
Thanks for emucon-raven ;)

[agranlund]

In these last hours I'm making another attempt and as I tell you I'm crossing my fingers... I replaced the MC68150FN40 with the new Freescale 68150FN33 that I had. I left only the video card and placed it in slot 4. On Emutos it seems to work quite well, and I can play the games you posted. Every now and then the screen freezes and I have to turn it off and on again, but less than before. If I also place the network card (which I purchased and it makes pings) the system becomes noticeably more unstable :(

I wonder if it has something to do with the graphics card. I do not have one with 100ns dram to test with.
Does it make any difference with other screen resolutions? Or without NOVA driver and just the built-in 640x480x2 color?
If I was going at it without hardware debugging I'd try to somehow find a pattern in what you where doing and then adjust the test actions deterministically to try and pinpoint some kind of pattern.

Is it always in relation to something that is about to draw or otherwise manipulate graphics?
Does the issue show itself if you just leave it at the desktop doing nothing?
what if it just plays a midi over and over with mxplay but otherwise does nothing?
and then the same test with an ym2149 tune, playing in sndplay.
Doom running in self-demo mode over and over again?
(Tyrian will crash to desktop if left in demo mode for a couple of levels, so perhaps not the best one for that kind of test :)

Otherwise, what you last posted makes it feel like something is not great on the ISA side, or possibly one of the other parts that share databus with ISA?

"Every now and then the screen freezes and I have to turn it off and on again,"

Wow, I haven't seen that happen here. Nor the other issues you have mentioned.
I'm still puzzled by the fact that MiNT 68060 kernel isn't working but the 68000 one is and somehow it's probably a piece of the same puzzle but it's very difficult to see the pattern.

There is one thing that is not clear to me: but is isa_bios.prg still necessary with the latest release you published?

Yep isa bios is still a separate executable for now.

[luciodra]

I wonder if it has something to do with the graphics card. I do not have one with 100ns dram to test with.
Does it make any difference with other screen resolutions? Or without NOVA driver and just the built-in 640x480x2 color?
If I was going at it without hardware debugging I'd try to somehow find a pattern in what you where doing and then adjust the test actions deterministically to try and pinpoint some kind of pattern.

Is it always in relation to something that is about to draw or otherwise manipulate graphics?
Does the issue show itself if you just leave it at the desktop doing nothing?
what if it just plays a midi over and over with mxplay but otherwise does nothing?
and then the same test with an ym2149 tune, playing in sndplay.
Doom running in self-demo mode over and over again?
(Tyrian will crash to desktop if left in demo mode for a couple of levels, so perhaps not the best one for that kind of test :)

Otherwise, what you last posted makes it feel like something is not great on the ISA side, or possibly one of the other parts that share databus with ISA?
Wow, I haven't seen that happen here. Nor the other issues you have mentioned.
I'm still puzzled by the fact that MiNT 68060 kernel isn't working but the 68000 one is and somehow it's probably a piece of the same puzzle but it's very difficult to see the pattern.
Yep isa bios is still a separate executable for now.

If the problem is only for me, then it's something electric. Everything I could replace with something else I did. The current situation, which is the most stable, is with only the graphics card in the isa slots. In this way, blocks are less frequent, but they are there. When do they occur? While using graphics, games, Gembench, which are the applications I use the most to test. Sometimes it also happens to me while using sndplay and rarely the pointer is stuck for no reason. If I hadn't bought the isa slots and sockets from Mouser, I'd start to think they're of poor quality. As I wrote a few days ago I also tried with a CL GD5420 but it freezes.
I think I'll disassemble the motherboard and take it to the friend who assembled it for me to check if everything is okay, even if I have some doubts that it falls within his competence...

[Steve]

Could you have a fake chinese re-badged 060?

[luciodra]

Could you have a fake chinese re-badged 060?

https://www.amibay.com/threads/lc060-at ... 4.2444716/

It could be... that?

[Steve]

Could you have a fake chinese re-badged 060?

https://www.amibay.com/threads/lc060-at ... 4.2444716/

It could be... that?

The CPU has factory marking of the EC, but this weirdness has long been known by Motorola ;-)

Ok I think this is your answer. Motorola would mark any CPU's with 'faulty' aspects in the FPU/MMU as LC/EC respectively. I would say that while your CPU may be reporting an MMU and/or FPU, there might be problems. Obviously that Amibay seller hasn't done thorough enough tests, but as far as I am aware, Motorola would have known if they needed to mark an 060 CPU as LC/EC, why else would they lose money. This would explain why you can't run 68060 builds of MiNT, and then anything else requires a fully functioning MMU due to how agranlund designed the Raven, which explains your instability. (These probably run fine in an Amiga, but a Raven needs the real deal)

[luciodra]

On Amiga I have never tested it, but I have already bought a rev 6 that I am waiting for, at this point with anxiety ;)