2nd Aug 2021, 8:55 pm | #61 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 Programming Interface
On a serious note Phil's created an uploader for the Elf which can be seen here....
http://www.mccrash-racing.co.uk/phil...elf_keyer2.pdf This uses the same technique as I did when I adapted this one for the Triton of making the lines pullups that you pull down so I can modify my extended board to do the same thing. If and when we get a larger version of this with more opto's then it will be another machine in its armoury if I can get it working. If not then I will build one of Phil's devices. |
2nd Aug 2021, 8:58 pm | #62 | |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 Programming Interface
Quote:
I suspect the H9 will be a longer term project. |
|
2nd Aug 2021, 9:01 pm | #63 | |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 Programming Interface
Quote:
I know this is one thing I won't be able to persuade you on, though, and it's true that anything like that would be a decidedly non-period feature. |
|
2nd Aug 2021, 9:18 pm | #64 | ||
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 Programming Interface
Quote:
|
||
2nd Aug 2021, 9:19 pm | #65 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 Programming Interface
In fact it could repeat the edge connector for a PI uploader or even have one built in...
|
2nd Aug 2021, 9:48 pm | #66 | |||
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: MK14 Programming Interface
Quote:
Thinking about it for the MK14 I was thinking you could make a SCRUMPI3 style keyboard with ~40 keys and several shifts to save on keys and scanning time while retaining dedicated MEM TERM ABORT and GO keys. I have a funky coloured PS/2 keyboard so another idea was to make a PS/2 to MK14 interface using one of my uploader boards and a microcontroller to make a simple keyboard interface, with a programmable mapping of PS/2 keys to MK14 keys. My only worry about all this is that the MK14 has a very limited memory so how much use I could make of it. |
|||
2nd Aug 2021, 10:05 pm | #67 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 Programming Interface
As a first step, do you know the 'key14' test utility which comes with the uploader 'software' package - if you run that up and use the Pi's keyboard (actual or remote, via SSH) to 'press' MK14 keys, do they ever bounce in the same way as your keys do when you are using the physical keys?
Connect any other momentary contact switch across the 'Mem' row and column on the edge connector and use that to step through the memory locations - does it bounce - skip locations sometimes - if you do that? Does the extent to which this happens change when the interface is / is not connected? I have only ever used the TLP185 optos in my own interfaces so I can't say whether your PC817s will completely let go of the keypad lines when turned off. However, I think you originally had this problem before you even built the uploader. As far as I know I'm using the exact same hardware in my issue VI -except- that I am using Omron branded low Newton force tact switches which were bought through Farnell. I can't say I have ever experienced problems with switch bounce. Didn't Mark have something similar until he replaced ...mmmm... the 7445? Are you using a 'plain' 7445? |
2nd Aug 2021, 10:08 pm | #68 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 Programming Interface
Yes I also wonder with so many fakes around is it a real 7445?
|
2nd Aug 2021, 10:15 pm | #69 | |||
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 Programming Interface
Quote:
I had not planned on repeating the Function keys externally just a proper ASCII (well MK14 VDU charset) for the MK14 long term i.e. what they would have needed to run a version of NIBL in Page1 - the MK14 remains a number/function pad beside it. I assume it would be a routing nightmare to map the existing keys so it acts as both a full ASCII and supported in SCIOS... it would be interesting though. |
|||
2nd Aug 2021, 10:26 pm | #70 |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: MK14 Programming Interface
Well you could have the MEM TERM GO keys be M T and G with ESC for ABORT... 0..9 mapped as before and symbols as shifted numbers. Maybe use RETURN as TERM rather than T.
|
2nd Aug 2021, 10:27 pm | #71 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 Programming Interface
You could map 0-9 A-F as they are on the standard keypad and map all the other alpha characters etc into the currently unused row / column positions and sort it all out using a lookup table embedded in whatever RAM based application will be reading the full keyboard. Assuming outboard RAM mapped at 0200-07FF, of which 512 bytes is dedicated as screen RAM, the system will have 1K + 640 bytes of RAM for your word processor / terminal program to run in. Slightly more than a basic ZX81.
|
2nd Aug 2021, 10:56 pm | #72 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 Programming Interface
Assuming we could make enough space in the NIBL ROM to have the lookup table and the scan code those might be a really interesting way to do it. RETURN as TERM is a good idea but, ESC as ABORT may not be depending on how we map the full ASCII set as we will need two SHIFTS (CTRL and SHIFT maybe) so maybe DEL as being a useful key to stand alone doubles as ABORT.
We need to scribble on a copy of the current grid layout... oh look |
2nd Aug 2021, 11:01 pm | #73 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 Programming Interface
How we map that to a physical layout that can be done with a double sided board though....
I assume the Shifted chars would be best using the Spectrum layout as muscle memory would remember where they were as otherwise custom keycaps would be needed. |
2nd Aug 2021, 11:04 pm | #74 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 Programming Interface
Blast missed SPACE - maybe instead of DEL (So is Abort) - using the ZX81 layout as Keycaps are likely to be around for that.
|
2nd Aug 2021, 11:14 pm | #75 | |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: MK14 Programming Interface
Quote:
Do we have a reasonable picture of the SCRUMPI3 key layout? I tried looking about on the forum but couldn't see anything. |
|
2nd Aug 2021, 11:47 pm | #76 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 Programming Interface
|
3rd Aug 2021, 12:27 am | #77 |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: MK14 Programming Interface
Hmm I wonder what the "User" key was for - some kind of 'Fn' key? Its pretty impressive to get all that into 21 keys. I presume the "Int" key was connected to interrupt, or was the equivalent of a break key.
|
3rd Aug 2021, 12:30 am | #78 |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: MK14 Programming Interface
As for your diagram above, you could get away with only 1 shift key and use the other as space. You could always have 2 physical shift keys wired in parallel for convenience.
|
3rd Aug 2021, 8:00 am | #79 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 Programming Interface
I prefer that idea to the SC-3 keypad which looks hardly any better than a dumbphone keypad. I wouldn't want to be typing text on that for any length of time.
I agree there hardly seems a need for two electrically separate shift keys when there is only upper case available in the character set. |
3rd Aug 2021, 8:47 am | #80 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,298
|
Re: MK14 Programming Interface
Writing NIBL programs using T9 predictive text would be an interesting challenge. Maybe a version optimized for NIBL keywords. That would also avoid using multiple shift keys and might even be easier than using the scrumpi 3 keypad.
|