News:

Printed Amstrad Addict magazine announced, check it out here!

Main Menu
avatar_llopis

Amstrad Diagnostics ROM

Started by llopis, 17:35, 24 February 21

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Animalgril987

Hi Noel.
I've run ADR a couple of times, and each time I do the upper ram test I get a different sequence of red areas.
I have a 464 with M4 board and Revaldinho's 512k ram (C3 mode enabled).


I have the latest version of ADR.


Alan.

llopis

Thanks for letting me know, Alan. I also have Revaldinho's expansion, but I haven't tried it with the diagnostics program yet.
Did you have the ADR on the upper ROM or the lower ROM on the M4? Could you post a pic in case it gives me a hint?
Thanks!


Animalgril987

Hi Noel.
I just ran the file in the dsk.
I'll try and get some screenshots of the ram test, but it won't be until next weekend.


Alan.

llopis


Hi Alan,


I just tested the diagnostics program with the M4 and the 512K RAM expansion and I didn't have any weird errors. The only weird thing is that it reports 576KB, but I suppose that's correct since I have 64KB of upper RAM already on a 6128 and the expansion probably maps itself after that.


Is it possible that you're not getting quite enough power to the expansion card and it's causing those intermittent errors?




--Noel

Quote from: Animalgril987 on 22:25, 21 March 21
Hi Noel.
I just ran the file in the dsk.
I'll try and get some screenshots of the ram test, but it won't be until next weekend.


Alan.

Animalgril987

Hi Noel.
The MX4 compatible backplane has a separate supply from the CPC, and I only have the RAM expansion and Duke's M4 plugged in.




Alan




llopis

Quote from: Animalgril987 on 00:03, 23 March 21
Hi Noel.
The MX4 compatible backplane has a separate supply from the CPC, and I only have the RAM expansion and Duke's M4 plugged in.
Right. I thought that maybe you hadn't powered it on and it maybe wasn't enough.
When you get a chance, try version 1.0 and see if that fixes it. I fixed how I was addressing the upper RAM ports. I don't think it would have caused that, but you never know.

pelrun

I used to have constant crashes with my M4 and other cards, and it turned out to be the 5V to the CPC. Didn't matter that I had a separate high-power supply going to the MX4, or directly to the M4, things crashed *all the time* until I put 5.1-5.2V into the CPC and removed the second supply altogether. Now it just works.

Animalgril987

Hi Noel.
I can't attach JPGs of the screen, for some reason, but I still get various different ram faults being recorded.
In one test, the graphic showed everything was ok, but the summary line at the bottom reported faults.


Alan.

Audronic

Hi Noel


Just burned a  Version 1.01L  Eprom for the Lower Rom Board
It works well in the CPC464

with the exception that in the CPC6128 it thinks the Machine is a 464 and the Keyboard layout reflects that.
Apart from that it seems to work well

Thanks for all your good work

Keep Safe

Ray
Procrastinators Unite,
If it Ain't Broke PLEASE Don't Fix it.
I keep telling you I am Not Pedantic.
As I Live " Down Under " I Take my Gravity Tablets and Wear my Magnetic Boots to Keep me from Falling off.

llopis


Hi Ray,


Great to hear! The detection of the model has been fixed (should be ready in the next release tomorrow probably). The problem is that it shows CPC 464 instead of "It's a CPC but I have no idea which model it is" because the only reliable way I know of detecting the model is looking at the lower system ROM (which is not accessible if you're using a low rom board).


I could try looking at RAM and FDC presence, but a 464 can have all those things too. Maybe in the future I can be smarter about it.


Anyway, glad it's working better and thanks for the update!




--Noel

Quote from: Audronic on 04:10, 10 April 21
Hi Noel


Just burned a  Version 1.01L  Eprom for the Lower Rom Board
It works well in the CPC464

with the exception that in the CPC6128 it thinks the Machine is a 464 and the Keyboard layout reflects that.
Apart from that it seems to work well

Thanks for all your good work

Keep Safe

Ray

gerald

Quote from: llopis on 09:02, 10 April 21
The detection of the model has been fixed (should be ready in the next release tomorrow probably). The problem is that it shows CPC 464 instead of "It's a CPC but I have no idea which model it is" because the only reliable way I know of detecting the model is looking at the lower system ROM (which is not accessible if you're using a low rom board).
For poeple using the XMEM for overriding the lower ROM, you might just ask to re-enable the internal ROM after copying the code in available RAM (ie after checking it).Then you could test the internal lowerROM.

llopis

Quote from: gerald on 09:57, 10 April 21
For poeple using the XMEM for overriding the lower ROM, you might just ask to re-enable the internal ROM after copying the code in available RAM (ie after checking it).Then you could test the internal lowerROM.
Oooh, I didn't know you could do that with XMem. I'm already doing it with the Dandanator and the M4. Could you point me in the direction of the simplest way to re-enable the internal ROM and disable it again for the XMem? Thanks!

gerald

Quote from: llopis on 10:56, 10 April 21
Oooh, I didn't know you could do that with XMem. I'm already doing it with the Dandanator and the M4. Could you point me in the direction of the simplest way to re-enable the internal ROM and disable it again for the XMem? Thanks!
It's the user that has to physically toggle a switch  ;)

CraigsBar

Hi @Noel


I see on the Github a fix for Crash in the DSK version from 19 days ago. Any idea when this will be released as I currently have a new Tall Key CPC 464 that I want to give the a once over and the DSK woudl be be better w3ay of loading it. The version in releases however crashes


Thanks
Craig

IRC:  #Retro4All on Freenode

llopis

Quote from: CraigsBar on 14:12, 26 April 21
I see on the Github a fix for Crash in the DSK version from 19 days ago. Any idea when this will be released as I currently have a new Tall Key CPC 464 that I want to give the a once over and the DSK woudl be be better w3ay of loading it. The version in releases however crashes
Argh, yes! :-) I kept waiting to fix one thing... and then I got busy. I'll release a version tonight. Sorry for the delay.

CraigsBar

Quote from: llopis on 14:21, 26 April 21
Argh, yes! :-) I kept waiting to fix one thing... and then I got busy. I'll release a version tonight. Sorry for the delay.


Awesome news, thanks. This tool is amazing and I am now using it on most machines. But it is easier to use the disc version I find than openeing machines and swapping the lower rom ;)


No need to appologise, the test suite you ahev made is amazingly good.

IRC:  #Retro4All on Freenode

llopis

Quote from: CraigsBar on 20:02, 26 April 21

Awesome news, thanks. This tool is amazing and I am now using it on most machines. But it is easier to use the disc version I find than openeing machines and swapping the lower rom ;)
Glad to hear! I plan to keep updating it, so if you have some suggestions of things to check, let me know. Especially now that I have a new main menu and status screen, there's a lot more room for other diagnostics (the problem is starting to be fitting it in the 16KB ROM!).


New release is up by the way: https://github.com/llopis/amstrad-diagnostics/releases

CraigsBar

Quote from: llopis on 08:30, 27 April 21
Glad to hear! I plan to keep updating it, so if you have some suggestions of things to check, let me know. Especially now that I have a new main menu and status screen, there's a lot more room for other diagnostics (the problem is starting to be fitting it in the 16KB ROM!).


New release is up by the way: https://github.com/llopis/amstrad-diagnostics/releases


Yep, grabbed it and tested this morning. Yay the full 512kb in the Expansions is OK :) Interestingly it reports the CPC Version on this Tall Key as GX4000/Plus on the menu screen but correctly identifies the Roms on the Rom Check. Is this a bug? or is it the "fault" of the DDI-4 expansion (that clones a DDI-1, 256k Ram and 256k Silicon disc in a single interface) from Piotr (Zaxon).


Next test: I'll get the CDT and load that and see if this has the same issues with no disc interface fitted.

IRC:  #Retro4All on Freenode

CraigsBar

Quote from: CraigsBar on 09:14, 27 April 21

<snip>

Next test: I'll get the CDT and load that and see if this has the same issues with no disc interface fitted.


OK completed this test on the tall key... and the result is "Unknown CPC"
IRC:  #Retro4All on Freenode

llopis

Quote from: CraigsBar on 09:14, 27 April 21
Yep, grabbed it and tested this morning. Yay the full 512kb in the Expansions is OK :) Interestingly it reports the CPC Version on this Tall Key as GX4000/Plus on the menu screen but correctly identifies the Roms on the Rom Check. Is this a bug? or is it the "fault" of the DDI-4 expansion (that clones a DDI-1, 256k Ram and 256k Silicon disc in a single interface) from Piotr (Zaxon).

OK completed this test on the tall key... and the result is "Unknown CPC"
Oh wow, that's unexpected!


It attempts to detect GX4000/Plus by turning on the ASICS control registers. No normal CPC should do that, and I didn't think the DDI-4 did either, right? That's quite odd. I have a DDI-3 and I have a DDI-5 coming my way, so I can try it with those.


As for the unknown CPC... that at least it's more normal. It says that if it's not a Plus model but can't make sense of the system ROM. What does it say for ROM test? Does it detect the lower ROM correctly?

CraigsBar

Quote from: llopis on 09:46, 27 April 21
Oh wow, that's unexpected!


It attempts to detect GX4000/Plus by turning on the ASICS control registers. No normal CPC should do that, and I didn't think the DDI-4 did either, right? That's quite odd. I have a DDI-3 and I have a DDI-5 coming my way, so I can try it with those.


As for the unknown CPC... that at least it's more normal. It says that if it's not a Plus model but can't make sense of the system ROM. What does it say for ROM test? Does it detect the lower ROM correctly?


OK some tests done! All of these were loaded from tape. Even when the cpc had AmsDOS to ensure the exact same software was used.


First 2 are taken with stock configuration, no expansions just the cpc 464 tall key.


Second 2 are with the DDI-4 attached.
IRC:  #Retro4All on Freenode

llopis

Quote from: CraigsBar on 10:15, 27 April 21

OK some tests done! All of these were loaded from tape. Even when the cpc had AmsDOS to ensure the exact same software was used.


First 2 are taken with stock configuration, no expansions just the cpc 464 tall key.


Second 2 are with the DDI-4 attached.
That's really useful. Thanks! I'll see if I can fix it and release a new version soon.

CraigsBar

It works fine with the Z-MEM 1mb ram expansion tho. So that make this an awesome tool.

IRC:  #Retro4All on Freenode

Audronic

Hi Noel

Just tested the Lower rom version v1.1l.
It works well

Hows the progress on the next version.
Keep up the good work

LowerRom eprom data in the attachment.

Keep Safe

Ray

Procrastinators Unite,
If it Ain't Broke PLEASE Don't Fix it.
I keep telling you I am Not Pedantic.
As I Live " Down Under " I Take my Gravity Tablets and Wear my Magnetic Boots to Keep me from Falling off.

llopis

New release of the diagnostics ROM is up: https://github.com/llopis/amstrad-diagnostics/releases/tag/v1.2
The big changes are fixing the model detection and changing the RAM detection algorithm to be able to tell bad RAM from no RAM. It sounds dumb, but it's harder than it sounds. Before I had someone show me that their CPC 6128 with one bad upper RAM IC was reported as just having 64KB of RAM, but this new approach should fix that.

Powered by SMFPacks Menu Editor Mod