|
Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment. |
|
Thread Tools |
4th Jul 2020, 4:45 pm | #1 |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
MK14 NIBL adapter
Well the subject of adapting the MK14 to run NIBL has been talked about in other threads and I've been thinking about it. At first I was considering the changes necessary to modify my series VI MK14 replica to support paged memory so I could create a board that plugged onto the edge connector and support the ROM and RAM needed for the NIBL-E interpreter, leaving page 0 unmolested for the standard memory and peripherals.
This approach however didn't appeal because it basically requires the redesign of most of the address decoding logic because SOC took shortcuts with the 'purity' of the design that worked but essentially didn't always use the NRDS and NWDS signals to Marshall the keyboard and display devices (for example). An alternative approach that I believe Timbucus first dug up in his research was to make a board that plugged onto the processor socket and intercepts the control and address lines as required. Another advantage of this approach is that the adapter would work on any MK14 not just my replica. Based on the article Tim found I sketched out the following circuit that latches the upper 4 address bits, and redirects access to pages 1-7 to on-board memory chips and pretends to the MK14 that a memory hole in page 0 is being accessed. Luckily on most versions of the MK14 the NRDS and NWDS strobe the RAM IO chip so by intercepting those signals and forcing them high while changing the address lines to fall in the range decoded for the chip creates a reliable memory hole so when pages 1+ are addressed nothing on the MK14 main board responds. The links on the diagram set bits 8-11 to Axx with U4 which on the main board is the ram IO that won't respond because U3a,b force the main board and edge connector NRDS and NWDS signals high (U2a,b tristate the outputs in the case that the SC/MP is disabled by, for example the VDU). U5 is the latch that latches the upper address bits during the NADS cycle. The components are listed as LS types but I can't see any reason why HC or HCT wouldn't work. I must emphasise that this is completely untested, bit I hope in the next couple of weeks to be getting my Laptop and I'll be able to use my logic simulator to check things out because it's so easy to get things wrong! I'm presenting this for comment. Things missing are some kind of on-board serial interface that takes into account the strange flow control requirements as is currently being discussed on other threads, and some way of arranging mass storage for those 24k BASIC programs! |
4th Jul 2020, 6:43 pm | #2 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,363
|
Re: MK14 NIBL adapter
This is great work lets hope you can get your gear and progress this - I will help where I can of course as it has been one of my goals. Since I was unable to share my copy of the article due to it being a licensed reprint from the Museum someone has scanned the issue to archive.org so you can see it here:
https://archive.org/details/Personal...ge/62/mode/2up One thought it might be nice to support an 8K ROM though to allow Karen O's Page 2 addon on-board Assembler. https://www.vintage-radio.net/forum/...d.php?t=168405 I have been toying with an i2c interface for something like a 24LC256 on small removable cards so with some code squeezed into the Page 2 ROM you would have a relatively cheap mass storage device that should be able to be used fairly easily. |
4th Jul 2020, 9:19 pm | #3 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,294
|
Re: MK14 NIBL adapter
I’m taking a slightly different approach, designing a pcb for RC2014 to fit an sc/mpII. This would then allow use with a range of existing modules.
I think it could share the bus with a z80, so I can use one of my existing modules to load code to ram, then halt the z80 and enable the sc/mpII. As I couldn’t find a transparent latch with tristate output, I’m going to try a 74hct257. This also avoids inverting NADS. Also want to experiment with multiprocessing two sc/mpII. I’ll add a schematic when I get a bit further, if anyone is interested in reviewing for mistakes. |
4th Jul 2020, 9:45 pm | #4 | |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,363
|
Re: MK14 NIBL adapter
Quote:
|
|
4th Jul 2020, 10:08 pm | #5 | |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,294
|
Re: MK14 NIBL adapter
Quote:
I’m planning to use the hct257 as a latch, similar to the 157 in the mk14 display circuit, but with tristate output controlled by nenin of the first sc/mpII. Not sure if the 81ls95 could easily be used as a latch. |
|
4th Jul 2020, 10:17 pm | #6 | ||
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,363
|
Re: MK14 NIBL adapter
Quote:
https://www.vintage-radio.net/forum/...splay.php?f=99 |
||
4th Jul 2020, 10:44 pm | #7 |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: MK14 NIBL adapter
The address latch doesnt need to be transparent as the data bus is stable on the rising edge of NADS so you can use a 74x173 which is positive edge triggered latch without needing to invert it. Using a multiplexer as a latch is not reliable according to some people as was discussed in the early parts of the mk14 schematics thread.
|
4th Jul 2020, 11:59 pm | #8 | |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,294
|
Re: MK14 NIBL adapter
Quote:
Anyway sorry to divert the topic of this thread. |
|
5th Jul 2020, 9:24 am | #9 |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: MK14 NIBL adapter
Not at all, often important and interesting things come from the tangents, as long as their vaguely related
|