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.

Closed Thread
 
Thread Tools
Old 28th Oct 2020, 11:52 pm   #681
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,264
Default Re: Mk14 vdu

Quote:
Originally Posted by SiriusHardware View Post
Quote:
Originally Posted by Mark1960 View Post
Is it possible to scope chip enable that is unique to the block that should not be written and NWDS. Trigger on the chip enable and monitor NWDS.
Will do Mark, but I'm going to have a rest from this for the rest of the evening and pick it up again tomorrow evening.
Of course. I’m on a different time zone so by the time I catch up its bedtime for you guys.

I’m wishing I had a copy of the hardware to join the fun, but sure that between you this is going to be solved soon.
Mark1960 is offline  
Old 29th Oct 2020, 12:09 am   #682
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

Quote:
Originally Posted by Mark1960 View Post
Quote:
Originally Posted by SiriusHardware View Post
Quote:
Originally Posted by Mark1960 View Post
Is it possible to scope chip enable that is unique to the block that should not be written and NWDS. Trigger on the chip enable and monitor NWDS.
Will do Mark, but I'm going to have a rest from this for the rest of the evening and pick it up again tomorrow evening.
Of course. I’m on a different time zone so by the time I catch up its bedtime for you guys.

I’m wishing I had a copy of the hardware to join the fun, but sure that between you this is going to be solved soon.
You still have time to build one...
Timbucus is offline  
Old 29th Oct 2020, 12:33 am   #683
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

I'm still here, despite trying to escape and think about something else for a while at least a couple of hours ago. Mark has a real MK14 but as we know the early versions don't have connections for the VDU, nor do they have the useful memory hole at 0200-07FF, that area being occupied by 'images' of the OS in all revisions prior to issue V.

If I did consider connecting a VDU to the MK14 now I think I would solder a 40-pin turned pin socket to the CPU pads on the underside of the PCB, put a 40-pin header plug on the end of a wide bit of ribbon cable and use that to convey all the necessary signals back towards and beyond the rear end of the PCB. The trouble is that's not the only problem to overcome, there are a few minor mods to the PCB and the small matter of having to change the crystal from its original 4.43MHz to a 4.00MHz one - all modifications which I did make to my original machine back in the day and subsequently later reversed. I wouldn't think of connecting the VDU back up to it now.

What you really need is one of Slothie's famed issue VI replica PCBs, if you look back to the very early pages of this thread Slothie posted the design files for his PCB and you could actually have a small batch made yourself if you wanted to.

I did that and got the minimum five made - one is the one I am using, one I sold to Tim for exactly what it had cost me (5 GBP), one I gave to Phil over in France for his generosity in providing many people here with Scrumpi, PIC14 and other PCBs and the remaining two I sent to Slothie to do with as he saw fit, so I no longer have any spares myself but the process of getting them made was surprisingly painless and quite quick.
SiriusHardware is online now  
Old 29th Oct 2020, 3:03 am   #684
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,264
Default Re: Mk14 vdu

I did download the files for Slothie’s replica and still thinking about getting a few made. I have most of the components. Displays are in UK at my Mum’s, since last year, I guess nobody expected international travel to become so difficult this time last year when I ordered them from the Netherlands. Only thing that would be a problem is the proms, I was considering replacing with a 2716 or 2732, possibly connected via the vdu connections on the edge connector.

I also started laying out a keypad and display board, to be compatible with RC2014 bus, and I have an SC/MP board designed for the same bus that I haven’t had manufactured yet. I got a bit diverted playing with an INS8073 wired to the same bus. Maybe I should post some details in the homebrew section.
Mark1960 is offline  
Old 29th Oct 2020, 10:14 am   #685
Slothie
Octode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
Default Re: Mk14 vdu

If you are thinking of getting the PCBs made, I found JLCPCB to be considerably cheaper for a board this size than PCBWAY. The quality is very good too, both in the circuit and silkscreen. I recall I tried a couple of the European manufacturers and either had trouble getting the website quote to work or the cost was huge, albeit this was a few years ago now and I didn't make an exhaustive search so it might be worth checking again - you may even find it worth trying Oshpark, Seeed or other US/Canadian manufacturers, I didn't mainly because shipping costs have always been high for the UK.
As for the PROMs, I was going to make an adaptor from matrix board that plugged into one of the PROM sockets with the 4 data bits going via a bit of ribbon cable to the other socket, but fortunately I managed to get some Tesla PROMs programmed with help from people in the UK, France and Germany - they were more travelled that I was, so not a route I'd recommend. There are people on here that can program genuine National Semi DM74S571 chips and these do pop up on eBay from time to time.

I would imagine memory is probably going to be your worst problem, they are available but recently seem to have spiked in prices - perhaps you'll have more luck with chips from the USA if shipping is less than the rediculous amount it is to europe.

It would be interesting to see your experiences with the 8073, they are very rare! Information on the SC/MP 3 is also pretty hard to find.
Slothie is offline  
Old 29th Oct 2020, 5:43 pm   #686
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,264
Default Re: Mk14 vdu

I bought 10 of AM9111BPC from utsource last year at about $2.30 each and used two of these to fix my original MK14. I resisted the temptation to test the other 8 in the MK14 in case of damage to the sockets. Shipping and import cost might still be an issue but as this was part of a larger order it wasn’t much of an impact on the cost of the ram chips.

I have used Jlcpcb a few times. They have a 102.4 x 102.4 limit on the low cost offer compared to 100 x 100 from most of the others. It doesn’t sound much different but does allow 40 pin headers to be used.
Mark1960 is offline  
Old 29th Oct 2020, 5:55 pm   #687
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

JLCPCB was who I ordered mine from, Slothie's 'package' uploaded to them without any problems, the only change I made was to order them in black rather than the default green. I must say they did a nice job.
SiriusHardware is online now  
Old 29th Oct 2020, 6:04 pm   #688
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

It's also worth remembering that 74S571s are not the only pin compatible PROMs which can be used in the MK14, here, courtesy of mikesarcade.com, is a cross reference list showing a number of compatible types. The 'TS' at far left signfifies these are all tri-state output types, there are others with the same pinout which have open-collector outputs and would not be suitable for the MK14.

Code:
   Signetics     MMI     TI       Harris  Raytheon  AMD      National  Intel
   ---------     ---     --       ------  --------  ---      --------  -----
TS 82S131 (50ns) 6306-1  -        7621-5  29611     27S13    74S571    3622
   82S131A(30ns) 63S241  -        7621A   -         27S13A   74S571A   3622A
The Tesla version of the 74S571 (=MH74S571) presents a particular problem because hardly any current or historic device programmers can programme them.
SiriusHardware is online now  
Old 29th Oct 2020, 6:57 pm   #689
Slothie
Octode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
Default Re: Mk14 vdu

I got a quote today from JLCPCB and it was $13.20 + $8,54 for shipping to the UK for 5 MK14 boards that are too big for their low cost boards. I tried other places - OSH park wanted $247 for 3! Most of the UK and european sites didn't have online quotes aside from PCB cart whose I couldnt get to work! PCBWay was $72+$24 shipping for 5 which seems odd because theyre being pushed as a cheap PCB house...
Slothie is offline  
Old 29th Oct 2020, 7:07 pm   #690
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

The price for my batch of 5 from JLCPCB back in mid-year was a tiny bit over 25 GBP for 5 including packing and post to the UK. I don't think they incurred any duty, but whether that was by accident or design I don't know.
SiriusHardware is online now  
Old 29th Oct 2020, 8:23 pm   #691
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

Mamma Mia!

(Here I go again..)

First check for tonight will be to run the same version of CHARSET again but this time with OrtonView running the most recent pre-optimisation firmware (which I call V1.01). The rationale behind this is that we were so preoccupied with the very obvious system slowdown in that version that we did not really test and experiment with it very much, so I just want to see if the pre-op version also does this - if it doesn't we can surmise that it is something specifically in the optimised version which causes writes to one page to occasionally also cause a write to the same location in another page. I think this is already our default theory, but I prefer to prove things one way or the other and not just make assumptions as to what is more likely.

After that I'll revert to the latest mainstream firmware and have a go to capture what is going on around the occasional _CE pulses for the main RAM 0Fxx when the charset program is continually writing to 0Bxx. This is one case where I could do with being able to capture more than two channels, but it is what it is.
SiriusHardware is online now  
Old 29th Oct 2020, 9:05 pm   #692
Karen O
Rest in Peace
 
Join Date: Jul 2011
Location: Bridgnorth, Shropshire, UK.
Posts: 787
Default Re: Mk14 vdu

I've got one more trick to try - I've widened the margins for NENIN release during the active display region. This will mean the SC/MP running a bit slower but worth it if it works.

I'm really struggling to understand how the PIC VDU can alter memory contents. It has to be something to do with abortive write cycles...?
Attached Files
File Type: zip mk14vdu - latest.zip (6.5 KB, 43 views)
Karen O is offline  
Old 29th Oct 2020, 9:14 pm   #693
Slothie
Octode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
Default Re: Mk14 vdu

Quote:
Originally Posted by SiriusHardware View Post
This is one case where I could do with being able to capture more than two channels, but it is what it is.
You could try one of those £12 Saelee 8 channel clone logic analysers on eBay et al.
Slothie is offline  
Old 29th Oct 2020, 9:17 pm   #694
Slothie
Octode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
Default Re: Mk14 vdu

Quote:
Originally Posted by Karen O View Post
I'm really struggling to understand how the PIC VDU can alter memory contents. It has to be something to do with abortive write cycles...?
Yes, and its strange how it is at specific memory addresses too. It should be possible to get it to work though, since the 1st version worked so well. I don't think you need worry about consuming a fair bit more bus time as the revised software is so fast (too fast!).
Slothie is offline  
Old 29th Oct 2020, 9:41 pm   #695
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,264
Default Re: Mk14 vdu

Quote:
Originally Posted by SiriusHardware View Post
It's also worth remembering that 74S571s are not the only pin compatible PROMs which can be used in the MK14, here, courtesy of mikesarcade.com, is a cross reference list showing a number of compatible types. The 'TS' at far left signfifies these are all tri-state output types, there are others with the same pinout which have open-collector outputs and would not be suitable for the MK14.

Code:
   Signetics     MMI     TI       Harris  Raytheon  AMD      National  Intel
   ---------     ---     --       ------  --------  ---      --------  -----
TS 82S131 (50ns) 6306-1  -        7621-5  29611     27S13    74S571    3622
   82S131A(30ns) 63S241  -        7621A   -         27S13A   74S571A   3622A
The Tesla version of the 74S571 (=MH74S571) presents a particular problem because hardly any current or historic device programmers can programme them.
That’s very useful, I might be able to get 82s131 or 7621. I didn’t check the price yet. Any ideas which might be easier to program?
Mark1960 is offline  
Old 29th Oct 2020, 9:58 pm   #696
Karen O
Rest in Peace
 
Join Date: Jul 2011
Location: Bridgnorth, Shropshire, UK.
Posts: 787
Default Re: Mk14 vdu

Quote:
I don't think you need worry about consuming a fair bit more bus time as the revised software is so fast (too fast!).
I'm happy to remove the cycle scavenging (hand-over of bus back to SC/MP while generating display, at which time the VDU doesn't need the bus and...). I could make it a compile time option too.

I'm so sorry that I can't seem to crack this one, my friends.

We're so close, too.
Karen O is offline  
Old 29th Oct 2020, 10:06 pm   #697
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

Quote:
Originally Posted by Karen O View Post
I've got one more trick to try - I've widened the margins for NENIN release during the active display region. This will mean the SC/MP running a bit slower but worth it if it works.

I'm really struggling to understand how the PIC VDU can alter memory contents. It has to be something to do with abortive write cycles...?
Well I put that on a chip and it performs as well as the others but, MINEFIELD still corrupts sorry - I am unable to do much more tonight now but, I will like Sirius take a closer look at relative timings tomorrow after work.

I do not think it is the VDU doing the write. The level of corruption is definitely proportional to the number and type of writes in a program - the writes are related in place to the original ST statements which is why I am supporting the corruption of the address bus while NWDS is still low and maybe _CS for the RAMS is changing to the new address - meanwhile the SC/MP is merrily retrying its write now it has been allowed access.
Timbucus is offline  
Old 29th Oct 2020, 10:09 pm   #698
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

Quote:
Originally Posted by Karen O View Post
Quote:
I don't think you need worry about consuming a fair bit more bus time as the revised software is so fast (too fast!).
I'm happy to remove the cycle scavenging (hand-over of bus back to SC/MP while generating display, at which time the VDU doesn't need the bus and...). I could make it a compile time option too.

I'm so sorry that I can't seem to crack this one, my friends.

We're so close, too.
Well in real terms we have a working one in 352 (on an A or non A chip) - what is struggling is the highly optimized one. I call that tuning.... blame your test team for not isolating the fault.
Timbucus is offline  
Old 29th Oct 2020, 10:11 pm   #699
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

Quote:
Any ideas which might be easier to program?
I'm not quite sure as to the thrust of your question, you mean which might be the easiest to make a homebrew programmer for? All the different manufacturers have a different programming regime, some connect VPP to the bits to be programmed, some connect VPP to a single pin and you ground the pins / bits you want to program.

If you get actual National Semiconductor DM74S571 PROMs there is a documented Arduino project for programming those,

https://hackaday.io/project/25953-pr...rom-programmer

but be warned that neither it (nor any programmer capable of programming the Nat Semi parts) can programme the Tesla version - despite the misleading similarity between the part numbers the programming regime is quite different - there are a couple of threads about that somewhere.

I can tell you that I have a device programmer which can programme the Nat Semi DM74S571, Signetics 82S131 and AMD 27S13 and I'm happy to programme them for anyone who can get a pair of blank PROMs to me, although that offer will be less useful to you than most since you happen to be in Canada.
SiriusHardware is online now  
Old 29th Oct 2020, 10:22 pm   #700
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,264
Default Re: Mk14 vdu

Sorry, I was wondering which one might be easier to build a homebrew programmer for. It looks like the signetics has more information available on programming. I’ll try giving the local shop a call tomorrow to check the price.
Mark1960 is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 9:09 am.


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 - 2023, Paul Stenning.