UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Computers

Notices

Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment.

Reply
 
Thread Tools
Old 1st Apr 2024, 10:40 am   #21
Phil__G
Octode
 
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,186
Default Re: Karen's most recent SC/MP

I dont have one to hand but I think its cts. Its been a while but iirc the problem was that it continued to send one or more characters after the cts input should have stopped it, it did stop but not right away, it was enough to confuse xmodem. Or something similar, I forget xmodem always worked perfectly in the days of polled uarts, I think xmodems troubles started when interrupts and buffers became the norm!
Phil__G is offline   Reply With Quote
Old 1st Apr 2024, 1:38 pm   #22
Phil__G
Octode
 
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,186
Default Re: Karen's most recent SC/MP

I've done a bit of a 'package' for this project, the zip has the pic code with updates & NIBL and also a few examples
including improved versions of BASYS and DISASSEM, Ians 'fruit machine', a diagram etc.
A cheap, two-chip 32k NIBL with a monitor & disassembler cant be a bad thing can it?
Zip last updated this morning: https://www.mccrash-racing.co.uk/phi...picl.htm#k2019
Phil__G is offline   Reply With Quote
Old 7th Apr 2024, 5:52 pm   #23
Phil__G
Octode
 
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,186
Default Re: Karen's most recent SC/MP

Not sure if anyone is following this project, I've been using it extensively this week & uncovered a couple of emulation bugs. Neither is new, they were first noticed during the PICLv2 project - one was senseA & B reset, and the CSA instruction was corrupting STAT. The zip has been updated just now with 'fixed' pic code:
https://www.mccrash-racing.co.uk/phi...picl.htm#k2019
I've a spare PIC16F877 if anyone fancies a go, no charge

Last edited by Phil__G; 7th Apr 2024 at 6:09 pm.
Phil__G is offline   Reply With Quote
Old 7th Apr 2024, 6:42 pm   #24
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,341
Default Re: Karen's most recent SC/MP

Yes definitely following this and appreciate the effort you are putting in to fix bugs. I just received a few more PIC16F877 yesterday, not had chance to test them yet.

Hoping to make an adapter to 8060 pinout and also a PICL pcb to fit in a small case. I’m trying to finish a PIC14 pcb and an 8060 NIBL/MK14 expansion, both to fit in cases, so its going to take some time yet.
Mark1960 is online now   Reply With Quote
Old 7th Apr 2024, 7:12 pm   #25
Phil__G
Octode
 
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,186
Default Re: Karen's most recent SC/MP

Thanks Mark, these bugs dont affect it in its intended use, eg NIBL doesnt even use DAE
and unless you wired senseA you wouldnt even notice.
In Karens diagram she shows SenseA grounded, and since NIBL doesnt
use SenseA at all I assumed it was a "drawing typo" (is there a word for that?)
but now I wonder if it was to temporarily mask the problem. In a nutshell the
Original project works perfectly as a NIBL computer and the updates are only necessary
if you go a bit deeper with some machine-code calls. But then you'd build a 'real' SC/MP

Last edited by Phil__G; 7th Apr 2024 at 7:25 pm.
Phil__G is offline   Reply With Quote
Old 7th Apr 2024, 7:17 pm   #26
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,541
Default Re: Karen's most recent SC/MP

I'd definitely be interested in being able to accurately emulate an SC/MP at hardware / pinout level, given price that SC/MP's now typically are (having only recently bought 3, all from different sources, to save having to keep swapping the only one I originally got a few years ago when they were half the price).

And I did get a few PIC16F877's (and later PIC16F18877's) to use on one of Phil_G's PICLv2 PCB's, so should have some around to try.

I might even try to assemble one of these 2-chip systems, but many other SC/MP boards to do some finishing-off on at the moment.
ortek_service is offline   Reply With Quote
Old 8th Apr 2024, 10:54 pm   #27
Phil__G
Octode
 
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,186
Default Re: Karen's most recent SC/MP

Quote:
Originally Posted by ortek_service View Post
I'd definitely be interested in being able to accurately emulate an SC/MP at hardware / pinout level, given price that SC/MP's now...
When I get a mo I'll have a fresh look at the 'raw' sc/mp emulation and do all the bugfixes, make up an adapter and try it in an MK14, we wont know till we try it
My gut feeling after playing with the "2019 NIBL project" is that it will work fine
Phil__G is offline   Reply With Quote
Old 9th Apr 2024, 12:03 am   #28
Phil__G
Octode
 
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,186
Default Re: Karen's most recent SC/MP

Translation:
Only A0-A11 for the MK14 of course. Anything >4k will need either latch changes or pic code changes to do NADS and to multiplex data & address
Attached Thumbnails
Click image for larger version

Name:	SCMPEMU.jpg
Views:	47
Size:	81.0 KB
ID:	296073  
Attached Files
File Type: zip SCMPEMUpinout.zip (679.9 KB, 36 views)

Last edited by Phil__G; 9th Apr 2024 at 12:09 am.
Phil__G is offline   Reply With Quote
Old 9th Apr 2024, 12:17 am   #29
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,541
Default Re: Karen's most recent SC/MP

Quote:
Originally Posted by Phil__G View Post
Translation:
Only A0-A11 for the MK14 of course. Anything >4k will need either latch changes or pic code changes to do NADS and to multiplex data & address
I presume Xtal from MK14 can't be connected to the PIC's oscillator pins, if that needs to run at 20MHz to emulate a 4MHz clock SC/MP? (As I recall you waiting for a 20MHz Crystal, to run it at 4MHz Emuation speed)
- Unless this particular PIC has a built-in clock multiplier (rather than just a divider?), that works from an Ext. Xtal ?
ortek_service is offline   Reply With Quote
Old 9th Apr 2024, 2:01 am   #30
Phil__G
Octode
 
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,186
Default Re: Karen's most recent SC/MP

The pic would run at 4mhz but very slowly, needs its own separate 20mhz crystal.
I need to check the MK14 schematics to see if it needs NADS or any of the missing signals...
Phil__G is offline   Reply With Quote
Old 9th Apr 2024, 2:32 am   #31
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,541
Default Re: Karen's most recent SC/MP

From your pinout picture, It had seemed to me, that you were going to connect the PIC's Xtal pins to those of the SC/MP socket. So that's why I queried this.

But maybe just for test purposes at least, it might work.

I presume the PIC's built-in RC etc oscillator won't allow running at that high a frequency (as well as accuracy & stability probably not being being good enough on that).
ortek_service is offline   Reply With Quote
Old 9th Apr 2024, 1:29 pm   #32
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,890
Default Re: Karen's most recent SC/MP

Quote:
Originally Posted by Phil__G View Post
I need to check the MK14 schematics to see if it needs NADS or any of the missing signals...
Mark's 7474 fix for the OrtonView memory corruption problem needs an input from NADS to the VDU.
SiriusHardware is online now   Reply With Quote
Old 9th Apr 2024, 1:35 pm   #33
Phil__G
Octode
 
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,186
Default Re: Karen's most recent SC/MP

I dont expect it to work with the VDU anyway G as theres no way to release the bus,
there is no nenin, nenout, nbreq, nhold, cont, nads... not sure if the bare MK14 needs any of these, I need to check...

I've started an adapter, just making up 40 wire jumpers...

One further thought I had is that we'll need to be more careful with contentions like writing to ROM, the PIC outputs can
sink & source 25mA or so, in a fight my money's on the PIC...

Last edited by Phil__G; 9th Apr 2024 at 1:44 pm.
Phil__G is offline   Reply With Quote
Old 9th Apr 2024, 2:24 pm   #34
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,890
Default Re: Karen's most recent SC/MP

Can't you turn the bus lines into inputs in order to 'tristate' them? That is actually what Ortonview does, since at times it needs to be able to drive the MK14 address lines, but at other times has to let them go.
SiriusHardware is online now   Reply With Quote
Old 9th Apr 2024, 3:44 pm   #35
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,341
Default Re: Karen's most recent SC/MP

The 74ls74 fix for memory corruption with orton view was only needed due to the way the 8060 releases the bus when NENIN is raised at certain times in the memory cycle. On the PIC the dificult part is probably to detect the NENIN and stop the memory cycle without changing the normal cycle times. Maybe an interupt could do that. First step is just to see if it even runs on a basic MK14.
Mark1960 is online now   Reply With Quote
Old 9th Apr 2024, 3:56 pm   #36
Phil__G
Octode
 
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,186
Default Re: Karen's most recent SC/MP

Quote:
Originally Posted by SiriusHardware View Post
Can't you turn the bus lines into inputs in order to 'tristate' them?
It could but as it stands theres no NENIN input
As a trial I just swapped the NIBL code in the PIC for KB+ at 2400, changed to copy length to suit, set the write protect to below 0x800 and everything works perfectly except the G command. Serial fast-loader works, modify, copy, list, dump, register set & display, etc all perfect. But not 'Go', it immediately comes back to the KB+ prompt. The program counter image in the register area is correct (ie the Go address). Most odd, but we like puzzles
Phil__G is offline   Reply With Quote
Old 9th Apr 2024, 4:14 pm   #37
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,341
Default Re: Karen's most recent SC/MP

This might be due to the wrap around address being different to the intended address. Wrap around pc relative addresses will go to top of memory instead of 0Fxx, if you have chip enable of ram connected to A15 there is no ram at FFxx.
Mark1960 is online now   Reply With Quote
Old 9th Apr 2024, 4:21 pm   #38
Phil__G
Octode
 
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,186
Default Re: Karen's most recent SC/MP

thats the line I'm chasing Mark, when EXIT: picks up the registers from 0FF8 onwards using PC relative addressing with a negative offset its actually reading FFF8 which on the 2019 emulation is unoccupied, so it runs through back to zero and the KB+ prompt. Unfortunately the emulation has true 16 bit pointers rather than 12-plus-4.
Houston...
Phil__G is offline   Reply With Quote
Old 9th Apr 2024, 4:29 pm   #39
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,341
Default Re: Karen's most recent SC/MP

Try just grounding chip enable to the ram first. If that doesn’t work then also grounding A14, A13 and A12 would see if it works within a 4k system, similar to MK14.
Mark1960 is online now   Reply With Quote
Old 9th Apr 2024, 5:06 pm   #40
Phil__G
Octode
 
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,186
Default Re: Karen's most recent SC/MP

Some progress.
Within the emulation I patched Karens GEA (get effective address) macro to AND EAHI with 0x0F so mimicking the 12-bit EA thing.
This was a quick-fix, not fully considered, and there will no doubt be other implications...
but... it does work, kinda KB+ appears to run perfectly, it fast-loads & runs programs perfectly, but 1000h is an image of 0000h, ie we only have one page.
I'm a bit dim and it takes me a while to follow Karens methods, with their heavy use of macros and almost complete lack of
comments but if nothing else, at least it confirms where the problem lies.

Last edited by Phil__G; 9th Apr 2024 at 5:29 pm.
Phil__G is offline   Reply With Quote
Reply

Thread Tools



All times are GMT +1. The time now is 4:50 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2024, Paul Stenning.