31st Mar 2021, 9:11 am | #1281 |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: Non-working Commodore PET 3016
I don't think the current drain is an issue at all, "modern" (i.e. made in the 80s or later) EPROMs have significantly less power draw than the mask-programmed ROMs of the 70s. The ROMs in my 2001-8 would get uncomfortably hot, so they must have been dissipating several watts each. I've never encountered an EPROM that becomes even noticibly warm.
|
31st Mar 2021, 9:25 am | #1282 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,583
|
Re: Non-working Commodore PET 3016
I agree with Owen's observation that it would be 'nicer' if the 2716 CE / OE pins could be swapped but as this is for short-term testing I think I'll just be grateful that the relevant pins on a 2716 land where they do.
It looks as though I'll be putting Slothie's code in a 2716 in the adaptor (because it needs to work when A11 = high, address range F800-FFFF) and we'll just let the other code (address range E000-E7FF) go in a 2716 to be plugged in directly to the UD8 socket? Fingers crossed. |
31st Mar 2021, 11:58 am | #1283 | |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,820
|
Re: Non-working Commodore PET 3016
I have spare 24 pin turned pin sockets - is that what's required?
Colin. Quote:
|
|
31st Mar 2021, 12:01 pm | #1284 | |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,820
|
Re: Non-working Commodore PET 3016
I'll re-do the tests for all the RAM chips and post images later on today for comparison.
Colin. Quote:
|
|
31st Mar 2021, 12:13 pm | #1285 | |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,583
|
Re: Non-working Commodore PET 3016
Quote:
We'll cross that bridge if we come to it. |
|
31st Mar 2021, 12:18 pm | #1286 | ||
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,820
|
Re: Non-working Commodore PET 3016
OK - I'll make sure I have some conventional sockets here for the weekend.
Colin. Quote:
|
||
31st Mar 2021, 12:28 pm | #1287 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,583
|
Re: Non-working Commodore PET 3016
Rather than rush off and order them just try fitting one of the TP sockets you have there into UD9' socket and meter all the socket 'holes' to see if they are all making contact with the pads on the mainboard (you might have to measure on the underside). If they do then you won't need another socket.
This repair may cost you a bit more yet, so don't go buying parts unnecessarily. It all adds up. |
31st Mar 2021, 2:46 pm | #1288 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,820
|
Re: Non-working Commodore PET 3016
OK - UI2-9 and UJ2-9 pins 14 (green) and 15 (yellow) attached.
There will be four of these posts to attach all screen shots - not as a ZIP so all can read them. 1 of 4 Colin. |
31st Mar 2021, 2:48 pm | #1289 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,820
|
Re: Non-working Commodore PET 3016
OK - UI2-9 and UJ2-9 pins 14 (green) and 15 (yellow) attached.
There will be four of these posts to attach all screen shots - not as a ZIP so all can read them. 2 of 4 Colin. |
31st Mar 2021, 2:48 pm | #1290 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,820
|
Re: Non-working Commodore PET 3016
OK - UI2-9 and UJ2-9 pins 14 (green) and 15 (yellow) attached.
There will be four of these posts to attach all screen shots - not as a ZIP so all can read them. 3 of 4 Colin. |
31st Mar 2021, 2:50 pm | #1291 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,820
|
Re: Non-working Commodore PET 3016
OK - UI2-9 and UJ2-9 pins 14 (green) and 15 (yellow) attached.
There will be four of these posts to attach all screen shots - not as a ZIP so all can read them. 4 of 4 Colin. |
31st Mar 2021, 6:58 pm | #1292 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,583
|
Re: Non-working Commodore PET 3016
Thanks for doing all that. Remember when we were closing in on the video circuit problem and we were getting very similar output from all of the bits of every address of the original video RAM instead of the random output we were expecting? This is looking a bit like that, but then this is dynamic RAM so I don't know if we can expect the bits of this memory to initialise in random states.
I think the thing to do now, before going any further down the hardware route, is to try running some test code. I'll get those items off to you as soon as I can, hopefully tomorrow. Last edited by SiriusHardware; 31st Mar 2021 at 7:26 pm. |
31st Mar 2021, 7:18 pm | #1293 | |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,820
|
Re: Non-working Commodore PET 3016
Would we see any activity here if I pressed keys on the keyboard? Presuming the keyboard actually works that is....
Colin. Quote:
|
|
31st Mar 2021, 7:27 pm | #1294 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,583
|
Re: Non-working Commodore PET 3016
With things as they are, no, because the keyboard scanning is run by the code in the PROMs and the CPU is currently only able to get so far before stopping, so it doesn't set up the periodic interrupt which does the keyboard scanning. (Or maybe it does, but doesn't work because interrupts, like subroutine calls, need working system RAM in order to work).
One of the two test PROMs includes code which directly accesses the keyboard scanning hardware so you can see things changing on the screen when you press keys - if the keyboard scanning hardware is working, that is. |
1st Apr 2021, 1:50 am | #1295 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,298
|
Re: Non-working Commodore PET 3016
I wonder if #1288 to #1291 is that the memory content FF FF 00 00 FF FF 00 00 etc.
At least it is confirming that every ram can drive both high and low levels. Also as the level stays high or low until driven during the next read cycle is also a good sign. As soon as we get the test code running I’d like to see what RAM RW looks like. Hopefully not seeing activity on this RW before was just that there are so few write cycles when the PET code is running. |
1st Apr 2021, 1:26 pm | #1296 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,583
|
Re: Non-working Commodore PET 3016
I put the two test EPROMs in the post to Colin today. I have not studied their operation in detail but I would think the most likely behaviour is that they keep going until the first time a RAM read / write operation fails, stop, and report the failure.
From a diagnostic point of view it would be useful if, on failure to write to and read from a RAM location, it just kept on continually trying the same operation. Then you would be able to follow the signals, hopefully to the point of failure. |
4th Apr 2021, 12:23 pm | #1297 | |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,820
|
Re: Non-working Commodore PET 3016
Hi - I've just been told I had a delivery yesterday and found the EPROMs - apologies. I have had words....
So - ready when you are. Colin. Quote:
|
|
4th Apr 2021, 1:24 pm | #1298 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,583
|
Re: Non-working Commodore PET 3016
Hi Colin, glad they got there. I am tied up for this afternoon, heading for my first vaccine shot at long last. Back later.
|
4th Apr 2021, 3:25 pm | #1299 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,820
|
Re: Non-working Commodore PET 3016
|
4th Apr 2021, 3:28 pm | #1300 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,583
|
Re: Non-working Commodore PET 3016
Right then,
It's very simple. Restore the machine to normal configuration (CPU directly fitted). Remove UD9 (only) and try the 'Slothie' EPROM - fitted in the adaptor - in that position, turn on, see what happens. If it works at all it will most likely report an error message of some sort as long as the machine is working enough for the EPROM to be able to work at all. Slothie, being the person who kindly wrote that code, will know more about the detail of what it does but I gather it does tests and reports the results in plain english on the screen (which of course is why we went to such trouble to get the display hardware working first). If it does work take a screen shot of the screen it stops on / error it reports. Regardless of whether it does or doesn't work, then refit the original UD9, remove UD8 and fit the Daver2 EPROM in the UD8 position without the adaptor and again just see what happens and if it runs at all, show us or tell us what happens. The UD8 replacement EPROM does quite a few tests and there is a well written manual for it which I will try to find the link to again, but that will be academic if it just doesn't work anyway. |