Oric Atmos resets after a few seconds ... when not grounded

[Godzil]

Oh yes I do remember theses ROMs, but theses ROM are way more recent than the Oric production, where does that german keyboard come from? Custom made?

What I can say is that IC11 and the two ROM socket was probably soldered at the same time as the rest of the computer, so even if it is an issue 4 it is an early produced Oric 1. (that have been upgraded to an atmos in some way)

If the ROM is prooved to be a modified version of Pascal's rom ( http://pleclerc.free.fr/Oric/OricPage.html ) that mean it is a recent change on that computer, but does not explain the keyboard it self.

And Pascal ROM are not going higher than 1.21

And the file with all known version and CRC: http://oric.free.fr/ROMS/romident.txt does not list any 1.3"a"

That's weird it is really a non known version version

I will also have a look at it a bit later

[Steve M]

There are several tracks cut that I can see on the board. The blue wires may be connecting differently - I'm not sure what is going on there. I'd be inclined to make a note of the connection and then remove and reinstate the cut tracks. (Is this a bad idea ??)

There's a capacitor stuck on the 74ls04 . I wonder if that could be causing a problem?

[Godzil]

The blue wires are mostly (and weirdly connected) to use IC11 to buffer Phi2, the use of wire like this will not really improve the signal quality in the end but yet.

I don't remember, btu there were one wire that was strange for me and didn't get it's reason to be there, problem I haven't documented that (or lost it) since I had a look :/

Had a really quick look at the rom file, as iss says it seems to be mostly the same as our old good 1.1(b?) but there are changes, they changed the copyright text to make it shorter (and it is starting 2 bytes earlier in some "padding" bytes (two 00 bytes have disappeared) I suspect some code has been stuffed there, also, on the hundred bytes of the ROM, there is two empty zones at 0x3F98 (from the start of the ROM not Oric address) and 0x3FD8 which contain "Normal" in the first place and "Version" on the other.

Some bytes around has been modified too, but I don't have the code, just the hex code so I'm not sure what the meaning of the modification.

[iss]

Hi ...
When i checked the mainboard i noticed that there was some work on it before: ...

Mmmm strange, an Atmos with two EPROMS, It will not work with your controller.

BTW, I think there is good chance that this Atmos will work with the Silicebit's controller. In my opinion this is the reason for the modifications with the blue wires on the bottom side.

[TomPhobos]

BTW, I think there is good chance that this Atmos will work with the Silicebit's controller. In my opinion this is the reason for the modifications with the blue wires on the bottom side.

Well, i'll definitely check out the controller with my Atmos. Maybe it works or, if not, it solves the problem with the resets because the Atmos then uses another power source (from the expansion connector instead of the power connector).
Also i'll dump the EPROMs of the Atmos, maybe it's another unknown version.

[Steve M]

The twin EPROM was a feature on the Oric-1. Once a manufactured ROM was fitted it was in a single case and IC11 wasn't needed. There should be such a thing as a twin ROM Atmos. I have seen people burning there own ROMs to get an Atmos working - but it's an oddity.
https://www.thefuturewas8bit.com/index.php/oricrepairs

Not sure if the reset circuitry is at fault - it's pretty basic. There's a mod to improve it used here -
https://www.thefuturewas8bit.com/index.php/oricrepairs

But if the capacitor has been changed I would think the problem is elsewhere(?)

Again - the capacitor on 74ls04 ???

[TomPhobos]

OK, here is a dump of both EPROMs. I dumped them with a Willem Eprommer.
Maybe it's the same ROM as the other german dump, but maybe it's another unknown ROM. We will see

[Godzil]

/me having a look at the ROM

So, I'm attaching the concatenated file (which work in Oricutron) :

Oric_German.rom.zip

(12.66 KiB) Downloaded 60 times

The CRC32 is 242799D0 and SHA1 is 101ec597c09c5c387132b0bc477965bcf23412d4

So this ROM report itself as 1.1, but the keyboard is clearly a QWERTZ (pressing Y tape a Z and vice-versa)

Also the charset is slightly different:

To be compared with the normal one:

Haven't seen any differences for now, but can't say for sure, will look at the ROM iteself, the difference may only lie in the charset and the keyboard table

edit: ok have done some really quick check, the only difference is in the charset and keyboard table:

Code: Select all

< 00003e70  0e 10 10 10 3c 10 3e 00  0c 12 2d 29 29 2d 12 0c  |....<.>...-))-..|
---
> 00003e70  08 08 08 08 08 08 08 08  0c 12 2d 29 29 2d 12 0c  |..........-))-..|
1014,1016c1014,1016
< 00003f50  0e 18 18 30 18 18 0e 00  08 08 08 08 08 08 08 08  |...0............|
< 00003f60  38 0c 0c 06 0c 0c 38 00  2a 15 2a 15 2a 15 2a 15  |8.....8.*.*.*.*.|
< 00003f70  3f 3f 3f 3f 3f 3f 3f 3f  37 ea ed eb 20 f5 f9 38  |????????7... ..8|
---
> 00003f50  12 00 1c 02 1e 22 1e 00  12 00 1c 22 22 22 1c 00  |....."....."""..|
> 00003f60  12 00 22 22 22 26 1a 00  2a 15 2a 15 2a 15 2a 15  |.."""&..*.*.*.*.|
> 00003f70  3f 3f 3f 3f 3f 3f 00 70  37 ea ed eb 20 f5 fa 38  |??????.p7... ..8|
1019,1020c1019,1020
< 00003fa0  31 1b fa 00 08 7f e1 0d  f8 f1 32 5c 0a 5d f3 00  |1.........2\.]..|
< 00003fb0  33 e4 e3 27 09 5b f7 3d  26 4a 4d 4b 20 55 59 2a  |3..'.[.=&JMK UY*|
---
> 00003fa0  31 1b f9 00 08 7f e1 0d  f8 f1 32 5c 0a 5d f3 00  |1.........2\.]..|
> 00003fb0  33 e4 e3 27 09 5b f7 3d  26 4a 4d 4b 20 55 5a 2a  |3..'.[.=&JMK UZ*|
1022,1024c1022,1024
< 00003fd0  56 46 24 5f 0b 50 45 3f  00 00 00 00 00 00 00 00  |VF$_.PE?........|
< 00003fe0  21 1b 5a 00 08 7f 41 0d  58 51 40 7c 0a 7d 53 00  |!.Z...A.XQ@|.}S.|
< 00003ff0  23 44 43 22 09 7b 57 2b  d0 01 47 02 8f f8 44 02  |#DC".{W+..G...D.|
---
> 00003fd0  56 46 24 7b 0b 50 45 3f  00 00 00 00 00 00 00 00  |VF${.PE?........|
> 00003fe0  21 1b 59 00 08 7f 41 0d  58 51 40 5f 0a 7d 53 00  |!.Y...A.XQ@_.}S.|
> 00003ff0  23 44 43 22 09 7c 57 2b  d0 01 47 02 8f f8 44 02  |#DC".|W+..G...D.|

You have to add $C000 to the addresses (as it is file related and not memory here), the left part (line starting with <) as the standard basic 1.1b, the right side (list starting with ">") are the German one.

$3Exx + $C000 == FExx so the first three diffed chunks are in the charmap (starting at $FC78) and the last one is in the keyboard table (starting at $FF78)

[Symoon]

I compared too TomPhobos' set of ROM have the exact same content than the one Fabrice and Thierry analysed a few years ago!
I suspect the v1.3A is a home-modified one?