Testing Tetroid backplane and GC

Nagging hardware related question? Post here!
User avatar
Doug
Chuggy Microdrive
Posts: 52
Joined: Tue Apr 04, 2017 11:25 am

Re: Testing Tetroid backplane and GC

Post by Doug »

I've since tried the Gold Card in the backplane (with the buffer board) without the 68008, and it still won't work, alas!


User avatar
mk79
QL Wafer Drive
Posts: 1349
Joined: Sun Feb 02, 2014 10:54 am
Location: Esslingen/Germany
Contact:

Re: Testing Tetroid backplane and GC

Post by mk79 »

RWAP wrote:The ATR code is also attached - this probably needs some tidying up too - again is suitable for use with q-emulator
I think that's the same that is already on Dilwyn's page. But there were also driver ROMs with integrated ATR device, which I have parts of but not all.

Marcel


User avatar
Dave
SandySuperQDave
Posts: 2765
Joined: Sat Jan 22, 2011 6:52 am
Location: Austin, TX
Contact:

Re: Testing Tetroid backplane and GC

Post by Dave »

QLObi wrote:The Qimi Interface Eagle files and firmware for a PS/2 mouse Driver is already published and given free by me for non-Commercial reproduction.
Just do !

Detlef
Detlef,

Would you permit me making the PCB available for a charge that just covers my costs? Under €5? And/or putting together a kit of all the parts? At least it would be generally available.

If I unintentionally made a 'profit' I could roll it into other hardware....


User avatar
Doug
Chuggy Microdrive
Posts: 52
Joined: Tue Apr 04, 2017 11:25 am

Re: Testing Tetroid backplane and GC

Post by Doug »

Great to see the SerMouse software available here! It has inspired me to look out for a serial mouse on the listings. (And a bus one, should some Qimi stuff become available too)
afx wrote:Yes, but keep in mind that ser1 and ser2 connections are different. I had to change the connections of the retrocomputershack adapter to connect my SerMouse through ser2_ port .
Forgive my ignorance, but does this mean that the retrocomputershack adaptor would be ok for connecting to a mouse, if it was in SER1? And I guess the follow-up question is, is it awkward for it to be in SER1, based on what programs like SerMouse (which I've no experience of) are expecting? (Or is it more a case of needing SER1 free for other things?)

Oh and a further follow up question! Would the retrocomputershack adaptor be suitable for joysticks?

And finally, does anyone have anything else I could try with the Gold Card and the Tetroid backplane kit? I get the feeling that like Pr0f said, it is only going to work with a shorter backplane where the signals have less distance to travel. Though that still doesn't explain why the Tetroid clone GC is supposed to work with the Tetroid backplane and mine doesn't, if they are pretty much the same...

Is there a diagnostics program for the GC which could throw up any issues, memory problems etc?


User avatar
dilwyn
Mr QL
Posts: 2750
Joined: Wed Dec 01, 2010 10:39 pm

Re: Testing Tetroid backplane and GC

Post by dilwyn »

RWAP wrote:
mk79 wrote: Albin has given me permission to release SerMouse, but he doesn't find either the code or the manual. So if somebody could come up with the stuff it can be put online.
Marcel
I missed that announcement.
Gret news - the Sermouse code (v3.04) is attached (this is suitable for unzipping under Windows and attaching to q-emulator). I should have a printed manual here somewhere to be scanned in.
I've attached SerMouse instructions lifted from my SMSQ/E manual. Unfortunately, I don't have a printed manual for SerMouse 3.04 to know how appropriate this version is - could someone compare them?

I have a Word .docx version of this I could send if anyone wishes to update it, unfortunately the Forum won't let me attach that type of file.
SERMouse.pdf


User avatar
BSJR
Trump Card
Posts: 182
Joined: Sun Oct 18, 2015 12:53 pm
Location: Amsterdam
Contact:

Re: Testing Tetroid backplane and GC

Post by BSJR »

dilwyn wrote: I've attached SerMouse instructions lifted from my SMSQ/E manual. Unfortunately, I don't have a printed manual for SerMouse 3.04 to know how appropriate this version is - could someone compare them?
Hi Dilwyn,
I have checked the pdf against my SMSQ/E manual revision 7, which is the last printed version I have and dated 1998.
It seems identical and as my database dates SeMouse 3.04 as 1997 we can assume it's the last version.

Bob


User avatar
Doug
Chuggy Microdrive
Posts: 52
Joined: Tue Apr 04, 2017 11:25 am

Re: Testing Tetroid backplane and GC

Post by Doug »

Hi guys,

I've had a very interesting weekend tinkering and learning more about how to set things up on the QL and what the limitations are. My issue with the Gold Card and Tetroid backplane remains, but I've made progress in other areas - just wanted to share those thoughts/findings which may be of benefit to other returning or new users.

Understanding the memory limitations with multiple peripherals:
Actually reading through the QubIDE documentation properly, especially the bit about the jumpers, really helped me here, for things even beyond QubIDE. It makes a lot more sense when wondering why things plugged into the ROM slot stop working when you have certain stuff plugged in to the expansion port, and why QubIDE may or may not get recognised. As the manual makes clear, there is apparently only one memory address (and associated jumper setting) which allows QubIDE to work when using a Trump or Gold card.

But this principle also applies to the Tetroid rom switcher - for a while I didn't think it is was working correctly as I had set the jumpers to load Minerva with Toolkit II. However, with the ROM slot occupied with my mouse software the toolkit didn't come up on the boot screen. Pulling the mouse out, it still didn't show up, because my Trump Card (actually the lesser 'Disk Card') also carries a slightly older version of Toolkit II that was, well, trumping it! Only when using the QL with just the QubIDE plugged in and nothing else, could I see the version of Toolkit II from the rom switcher show up on the boot screen.

If any of the more seasoned QL-ers are reading this, is Toolkit II on the Miracle Systems 'Disk Card' just on a chip I could pull off, so that the better version of Toolkit II with the rom switcher would work with the card? And even if pulling it off is possible, would the Trump Card still be occupying that memory space and so the newer toolkit still would not load? Can all this be avoided with some kind of boot file??

Trial and error with the Habi image editor:
My biggest breakthrough this weekend was getting the Habi image editor to successfully transfer files directly to my QL-hard-disk-formatted SD card.

I'd initially had a lot of trouble with this and given up. Some SD cards simply didn't seem to work with it, even though they would format on the QL. I also had no luck dumping the raw image of the card to the PC, in order to manipulate the file there. However, loading the partition directly from the SD-card reader in my PC did work. I was able to transfer some files, following the instructions, but sometimes the file names would become strange or garbled, and even if they didn't, they would just crash the QL when trying to load them there. This was true for both normal files and zips. I was following the 'Import programs from QubIDE IDE unit' instructions from here: http://hardware.speccy.org/temp/qubide4-i.html

However, when doing the opposite of what it says about the headers - so instead of selecting "Options" > "Headers" > "None", I did "Options" > "Headers" > "QL" - transferring files from PC worked perfectly - both zipped and unzipped files!

Old disks and other bits:
Having successfully set up a WIN1_ partition with the unzip software transferred and installed on it, along with some things to test like CHESS, I spent the rest of the time looking into the QubATA driver and the associated Starter Kit (that comes with the Tetroid interface), and whether or not it would be a good thing to try out with my setup. However, there were two problems. The Starter Kit seems to assume QubATA is already installed and available (won't run without it, calls commands I think are specific to QubATA), and in addition, it wants to format and re-partition the card as part of the setup.

Now with only one good SD card set up which I didn't want to wipe, and finding it tricky to mirror this to another SD card, I did find a blank card which seemed ok but then I had to resort to floppies to transfer QubATA and the Starter Kit, so that I could run them in conjunction with the blank SD card ready to be formatted. This is where I ran out of time, because of the handful of DD disks I could find kicking around, they were nearly all squeaky, stiff and cranky and wouldn't work properly. Even after cleaning the heads on the drives. I found one that was reliable...

I had zero luck transferring files to floppy using emulators. I had zero luck formatting floppies using emulators. Maybe the HD disk drive in my old PC isn't that great. Everything had to be done via the QL and my old 720k double disk drive - so I needed to first put things on the partitioned SD card using the Habi image editor, and then once back on the QL, copy them off to a floppy.

The only other disks I had knocking around were HD, and after taping the hole, I found the same issue as above - they would format and copy files just fine using the QL DD drive, but using emulators and my PC's floppy drive just wasn't happening. It seemed to work, but in the QL, they just threw up a 'not found'. I hope I can get it all done without buying more floppy disks. I rather resent the prices that even used floppies are going for on eBay and Amazon at the moment!

That's it for now. On the whole, it has been satisfying to finally make some progress and my QL is gradually whirring back into life, with abilities I'd never thought possible back in the day!


User avatar
Pr0f
QL Wafer Drive
Posts: 1297
Joined: Thu Oct 12, 2017 9:54 am

Re: Testing Tetroid backplane and GC

Post by Pr0f »

So, the trumpcard rom chip is a 32K device, with approximately one half being the disk driver and other driver support software, and the other half being the toolkit II.

I say approximately half, because some of the TK II images are actually larger than 16Kb.

From my understanding of the ROM scan logic (and this depends on the OS you are using), it looks in the ROM slot first, so whoever responds to that code between 48K and 64K will run it's initialisation code, If you have the ROM switcher with the top 16K enabled for TK II to load, it will load at this point, provided no other interface overides it. The ROM socket / expansion ROM port, are both controlled by the I/O decoder on the main board, so if any other device that's plugged into the main expansion slot decides it wants that 16K block between 48 and 64, your ROM switcher or external ROM cartridge will be switched out and never seen. This is what happens when the QUBIDE interface is set for that address.

If you are using Minerva, then it will also scan the 2 blocks of 16K above the ROM, which is where Trump/Gold and SuperGold cards wiill use. Again these all come with a version of TKII so will attempt to load it in. Then scanning passes to the first address in the memory map designed for expansion peripherals, but only if that space is not already occupied by RAM (which it would be in the face of the above 3 cards). I am not sure any of the ROM variants attempt to scan past the RAM limit. The Gold and Super gold cards do map part of their ROM and the system ROM to beyond the upper RAM limit, in order to be able to copy them back into RAM which starts at zero for these cards. The SuperGold card with the Aurora and a backplane will look at one last address which looks like it is at the very top of the 1Mb memory map, but in fact is remapped to much higher. That's becuase the Aurora card leaves that last 16K free, mapping only 240K of video RAM into high memory and not 256K. Qubide can be set to operate there, in theory I would image any other card with correctly written driver should also work there - but never tried it.

One final option open to you is to load and run TK II from a file, which you can place in the boot file - this leaves your ROM areas as free as possible, and does mean you can load the latest version of it. There is no 16K constraint on that file, so you can load the 'full version' of the toolkit.


RWAP
RWAP Master
Posts: 2833
Joined: Sun Nov 28, 2010 4:51 pm
Location: Stone, United Kingdom
Contact:

Re: Testing Tetroid backplane and GC

Post by RWAP »

Doug wrote: I had zero luck transferring files to floppy using emulators. I had zero luck formatting floppies using emulators. Maybe the HD disk drive in my old PC isn't that great. Everything had to be done via the QL and my old 720k double disk drive - so I needed to first put things on the partitioned SD card using the Habi image editor, and then once back on the QL, copy them off to a floppy.

The only other disks I had knocking around were HD, and after taping the hole, I found the same issue as above - they would format and copy files just fine using the QL DD drive, but using emulators and my PC's floppy drive just wasn't happening. It seemed to work, but in the QL, they just threw up a 'not found'. I hope I can get it all done without buying more floppy disks. I rather resent the prices that even used floppies are going for on eBay and Amazon at the moment!
HD disks with the hole taped over are ALWAYS a source of potential issue when moving them between disk drives. Do you have a disk drive built into the PC or a USB disk drive - as they are very different beasts (the latter has no disk controller). I am stuck with a USB disk drive and find that if I format DD disks on the PC it is hit or miss as to whether that works when I put the disk in the QL (Windows claims to no longer support DD disks) - however, there do not tend to be issues if you format them on the QL first (the QL is also much quicker at formatting them than either QPC2 or q-emulator).

The other trick is to force Windows to see the disk as a DD disk before you try using it (once you do one you can carry on with other disks during the same session). Open a cmd window and use

Code: Select all

FORMAT a: /t:80 /n:9


User avatar
Doug
Chuggy Microdrive
Posts: 52
Joined: Tue Apr 04, 2017 11:25 am

Re: Testing Tetroid backplane and GC

Post by Doug »

Pr0f wrote:So, the trumpcard rom chip is a 32K device, with approximately one half being the disk driver and other driver support software, and the other half being the toolkit II.

I say approximately half, because some of the TK II images are actually larger than 16Kb.

From my understanding of the ROM scan logic (and this depends on the OS you are using), it looks in the ROM slot first, so whoever responds to that code between 48K and 64K will run it's initialisation code, If you have the ROM switcher with the top 16K enabled for TK II to load, it will load at this point, provided no other interface overides it. The ROM socket / expansion ROM port, are both controlled by the I/O decoder on the main board, so if any other device that's plugged into the main expansion slot decides it wants that 16K block between 48 and 64, your ROM switcher or external ROM cartridge will be switched out and never seen. This is what happens when the QUBIDE interface is set for that address.

If you are using Minerva, then it will also scan the 2 blocks of 16K above the ROM, which is where Trump/Gold and SuperGold cards wiill use. Again these all come with a version of TKII so will attempt to load it in. Then scanning passes to the first address in the memory map designed for expansion peripherals, but only if that space is not already occupied by RAM (which it would be in the face of the above 3 cards). I am not sure any of the ROM variants attempt to scan past the RAM limit. The Gold and Super gold cards do map part of their ROM and the system ROM to beyond the upper RAM limit, in order to be able to copy them back into RAM which starts at zero for these cards. The SuperGold card with the Aurora and a backplane will look at one last address which looks like it is at the very top of the 1Mb memory map, but in fact is remapped to much higher. That's becuase the Aurora card leaves that last 16K free, mapping only 240K of video RAM into high memory and not 256K. Qubide can be set to operate there, in theory I would image any other card with correctly written driver should also work there - but never tried it.

One final option open to you is to load and run TK II from a file, which you can place in the boot file - this leaves your ROM areas as free as possible, and does mean you can load the latest version of it. There is no 16K constraint on that file, so you can load the 'full version' of the toolkit.
Many thanks Pr0f, very useful to know! Is there example of a boot file available which loads in the standard sort of things that people are interested in these days? TKII, QubATA, etc? I guess this is how SMSQ/E would be fired up too?


Post Reply