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 6th Jun 2020, 1:23 am   #61
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

The connections are (if recommended)

Code:
Set	Clear	Name	VDU	Desc
810	800	PA0	b13	Video on/off (the manual recommends FLAG2 so could be that - I have)
811	801	PA1	b11	PS3 (likely _BEST_ Not connected - wired to b17 Top Page... see notes)
812	802	PA2	b10	PS2
813	803	PA3	b15	Reverse Pages - as b15 is INTR - it should be re-routed here (b8)
814	804	PA4	b9	PS1
815	805	PA5	b12	PS4
816	806	PA6	b14	Graphics/Text - when 0 is text
817	807	PA7	b16	Invert video
You can of course just write the whole lot to 0x0820 in one go...

Last edited by Timbucus; 6th Jun 2020 at 1:31 am.
Timbucus is offline  
Old 6th Jun 2020, 2:01 am   #62
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

I'm not sure that leaving any of PA4, PA2, PA1, PA5 connected to the b9-b12 page select inputs is a good idea because 'top page' may potentially end up being connected to any of the page selects. My feeling is that it would be better to decide what block of the memory is going to be the 'screen RAM', hardwire b9-b12+'top page' for that configuration, stick with that choice and keep the port A pins safely out of the way.

If you want to go with your configuration above then I suggest always having a low value resistor (~100R?) in between 'top page' and anything it is ever connected to so that if Port A pins do get switched to output mode, the voltage difference can happen across the resistor and the opposing outputs will not damage each other. On your non-original VDU you might consider either cutting the 'top page' track coming from the 74L86 output and laying a SM resistor across the gap, or perhaps you can cut a gap in the DIN connector pin where it curves up away from the PCB and insert a SM or very small conventional resistor into that gap.
SiriusHardware is online now  
Old 6th Jun 2020, 4:25 pm   #63
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

Well I finally managed to get the beast under control so started typing in some of the software for it like the SHOWCHAR demo from the VDU manual and a nice LISTER program (shows the contents of memory in hex a page at a time correctly formatted as one or two bytes by instruction) from the FurtherApplicationPrograms book from SoC by DJD.

And I took some photos of the Character sets displayed by the two chips:

The original DM8678CAB: Click image for larger version

Name:	DM8678CABoriginal.jpg
Views:	115
Size:	86.5 KB
ID:	207826

and the pin compatible Martin found DM86S64CAB: Click image for larger version

Name:	DM86S64CAB.jpg
Views:	115
Size:	86.2 KB
ID:	207827
Timbucus is offline  
Old 6th Jun 2020, 4:26 pm   #64
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

Quote:
Originally Posted by SiriusHardware View Post
I'm not sure that leaving any of PA4, PA2, PA1, PA5 connected to the b9-b12 page select inputs is a good idea because 'top page' may potentially end up being connected to any of the page selects. My feeling is that it would be better to decide what block of the memory is going to be the 'screen RAM', hardwire b9-b12+'top page' for that configuration, stick with that choice and keep the port A pins safely out of the way.

If you want to go with your configuration above then I suggest always having a low value resistor (~100R?) in between 'top page' and anything it is ever connected to so that if Port A pins do get switched to output mode, the voltage difference can happen across the resistor and the opposing outputs will not damage each other. On your non-original VDU you might consider either cutting the 'top page' track coming from the 74L86 output and laying a SM resistor across the gap, or perhaps you can cut a gap in the DIN connector pin where it curves up away from the PCB and insert a SM or very small conventional resistor into that gap.
I will find a suitable place to put one - there is a Via near B1 that I use for a flying lead so I will probably insert it in that lead.
Timbucus is offline  
Old 6th Jun 2020, 4:41 pm   #65
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

Might be best to arrive at the right value for the limiting resistor in rough form before fitting one nicely. The value needs to be high enough to be able to take up to 5V across it without burning up, but barely any higher than that so that it does not unduly weaken the output drive from the 'top page' output. It may turn out that there is no such compromise and the resistor will just have to be allowed to cook - better to destroy a 2p resistor than a £12 or £35 chip.

Other than a few extraneous bits I don't think I could see any difference between the output from the 'real' character generator and the output from the substitute, so that's all good?

That 'lister' program from the book was probably the one which was running in that first VDU video by Milan, so that's another small mystery (probably) explained.
SiriusHardware is online now  
Old 7th Jun 2020, 10:03 pm   #66
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

This is another option I discovered a long time ago but forgot about until I went hunting through my notes. It is a MK14 Issue 5 with tape and single step cards and a VDU card from Reddich Electronics RE67 which uses the Video chip from the Dragon/Electron etc M6847.

https://www.***********/photos/retrop...57622778055029

For those who have not seen the box the MK14 kit was delivered in it also has photos of that.
Timbucus is offline  
Old 7th Jun 2020, 11:05 pm   #67
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

Only problem with that VDU is that, like the SOC VDU, it does not have its own RAM. Otherwise, you could probably reproduce it even today. Where's the character generator I wonder? Built into the 6847 perhaps?

I think I've come across that particular MK14 on Flickr before but there seem to be a few more images since I was last there some time ago. It illustrates very well the (wire) lengths needed in order to get the bus and control signals to the rear edge connector on issue IV and issue V PCBs.

I didn't even have that option because on the issue II there weren't even any lower side contacts on the edge connector to solder wires to. I had flying leads coming off the VDU and landing on tracks all over the PCB. Horrible.

Slothie and I are both pretty impressed by Tony's 'Teletext decoder' VDU with its I2C interface, although, given that the MK14 is already quite slow, anything slower than parallel access to the video RAM may make it impractical to use for any kind of game which aspires to a fast frame update.
SiriusHardware is online now  
Old 8th Jun 2020, 5:50 pm   #68
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

I took the opportunity to add a character invert option from Practical Electronics August 1980 which uses a 74LS32 to allow individual characters to be inverted controlled by Bit6 in Text Mode. As I will never use the Invert option over the whole screen as my TV's cannot cope without a Back Porch it seemed a useful addon.

In truth it changes a little before the character so it is only of use if you can put space around them. I did attempt to delay the signal a little by sticking it through a few gates but it did not help. But, having a 74LS00 on the circuit did allow me to use them as a NOT gate to invert D6 so the default of a character without Bit 6 set is normal was possible in case of any software issues.

Found a bug in the LISTER program already that it was setting that bit on some characters when calculating the Hex numbers...
Timbucus is offline  
Old 8th Jun 2020, 6:31 pm   #69
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

I started building Slothie's quick memory add-on which will just clag onto one side of the MK14-issue-VI-to-VDU interface 'card' I've made from two back to back bits of stripboard.

Have to remember to take a 6116 to work tomorrow. Edge connectors still have not turned up, hopefully will tomorrow.

I've taken the precaution of putting SM 330R resistors between VDU b9-b16 and the corresponding I/O port pins - not just the Page Select inputs but all the rest as well just in case I accidentally connect 'top page' to any of the others while their corresponding PAx pin is set as output. A lesson well learned (from Tim).

For now, I've left the VDU on / off control line driven from PA0 and I'm going to put a weak pullup on that so that after reset, with Port A defaulted to all inputs, the VDU will be held off until I turn PA0 around and drive it low.

I've also cut the link connecting 8154 INTR to 'Reverse Pages' (b15), cut the link connecting PB0 to 'top page' (b17) and jumped PB0 across to 'Reverse Pages' (b15).

And as already mentioned, I have rerouted the clock signal from the lower side to the upper side as it makes its way across the card, and have also provided the NRDS pullup as well.

I haven't (Yet) put a separate regulator for the VDU on the 'card', but that will be the next step after I get it basically working.
SiriusHardware is online now  
Old 8th Jun 2020, 6:48 pm   #70
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

I don't get why you have cut the link to PB0 - as that can be used as an input to see if you are still in the top page... I have patched my b15 over to b8 so it appears on PA3 and lets me control that through software on the INS - well it will do when I have a working IC1 again. I think what you have done won't work as you have just shorted Pin 1 and 3 on IC1... linking its output back to one of its inputs...

Edit:All the software I am writing is setting both the Flag and the PA0 to enable the VDU so it will cover either approach.
Timbucus is offline  
Old 8th Jun 2020, 6:53 pm   #71
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

In real terms you need to hook TopPage to b11 PS3 as that allows you to select F00 and B00 as the paged addresses. However, as I think you plan to use contiguous space in the 1.5K expansion that will not be vital.
Timbucus is offline  
Old 8th Jun 2020, 7:05 pm   #72
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

For those who are interested in the Reverse Video addon - you could just build it into the board between the MK14 and the VDU and just tap off the the one via which is not on the bus with a header pin and a flying lead that provides handy access to the buffer enable on IC1 Pin 9 - I pick it up near IC15 the white wire in the photo.

I have soldered the connections for the moment (all of them so the Blue, Red and Black - the yellow is D6 which is picked up below but, I will likely change to that approach when I make the 1.5K memory expansion.

There is also a picture of it in action...
Attached Thumbnails
Click image for larger version

Name:	IMG_3398.jpg
Views:	95
Size:	72.9 KB
ID:	208030   Click image for larger version

Name:	IMG_3399.jpg
Views:	90
Size:	71.6 KB
ID:	208031   Click image for larger version

Name:	IMG_3397.jpg
Views:	90
Size:	48.9 KB
ID:	208032  
Timbucus is offline  
Old 8th Jun 2020, 7:45 pm   #73
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

The VDU halts the SC/MP while it is slurping data from the RAM so I'm not sure whether the system will ever be able to poll 'top page' to see if the VDU is currently rendering the top page... since you are further ahead than me you can investigate this.

The main use for 'top page' is, as you have suggested, to auto-switch between two 256-byte banks of RAM, most likely 0F00-0FFF and 0B00-0BFF if you don't have more RAM, or some 512-byte continuous section of the area 0200-07FF if there is RAM there. Once I do have RAM there I will hardwire top page + the Page Select lines to use that area only.

I think my brain is running a bit slow tonight as I don't quite understand this

Quote:
I think what you have done won't work as you have just shorted Pin 1 and 3 on IC1
IC1 of... the VDU I think you mean? I haven't explicitly connected 'top page' (b17), IC1 pin 3, to anything. I have DISCONNECTED it from PB0, that's all. I plan to connect it to one of the Page Select inputs though.
SiriusHardware is online now  
Old 15th Jun 2020, 6:06 pm   #74
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

The 32-way edge connectors have finally arrived, I have two spares if anyone needs one on an emergency basis. The plastic used in them is surprisingly hard, as I found when I tried to open up the ends with a scalpel - that's one scalpel blade I probably won't ever see again. It went up, but didn't come back down. Probably embedded in a ceiling tile somewhere.

I'll focus on getting it working in the most basic way with the VDU first before attempting to graft the 1.5K memory upgrade onto it.

Tim, if you can remind me... when using the existing RAM which 256-byte block are you using as upper screen RAM and which block as lower? Intuitively, I would guess 0fXX as upper and 0Bxx as lower.
SiriusHardware is online now  
Old 15th Jun 2020, 6:25 pm   #75
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

Excellent - that is what I did to move forward slowly.

It is best to have F00 on the top as the OS uses a few bytes at the start which is on the invisible first line which is offscreen on my two TV’s

The Horse Race game though uses the B00 for the code as it runs i from the IO shadow at 0A80 so it varies depending on the software - it just means the horses are on the top with it in default that I use.
Timbucus is offline  
Old 15th Jun 2020, 6:31 pm   #76
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

Righto, I think I'll do the rounds with a meter first, and then I'll go through basic startup - I've got Arduino style female connectors tacked on to B9 through B16 and similar connectors on +5V and 0V so as long as I leave all the I/O ports in the default input mode I can play just by using arduino jumper leads. I have a resistor in series with the 'top page' b17 output so that no harm will be done if I accidentally connect it to an opposed output.

I spent a while last night trying to understand enough about the .BMP file format to 'rip' a 64 * 64 pixel image created in, ie, Paint to .DB statements - so far I've succeeded in using a very crude bit of python code to extract the single plane bitmap and display it as binary numbers arranged 8 bytes / 64 binary digits across and 64 lines down, so if I squint I can see the test image picked out in '1's rather than white pixels.

The idea is then to output the bitmap as a source file full of .DB tables which can be incorporated into source and then the simplest thing to do would be to have the first half of the image ORG at 0f00 and the second half of the image ORG at 0b00, assemble that into Intel Hex code and let the uploader 'type' the image into the screen RAM, a bit like watching a loading screen load from cassette.

Of course it's not quite that simple, you wouldn't be able to 'type' code into 0F00-0F11 because the monitor is actively using that RAM as scratchpad even as it is accepting key presses and displaying the results, so, the first three pixel lines of the image would have to be 'typed' into the I/O RAM (ORG at 08C0 perhaps) during the typing in phase, and then there would be a bit of auto-run code which would copy the displaced first three pixel lines from their temporary home in the I/O RAM to their proper place in the first three lines of screen memory. The code would then have to go into a busy loop and not return control to the monitor, otherwise the monitor would resume using 0F00-0F11 again and corrupt the first part of the image.

Last edited by SiriusHardware; 15th Jun 2020 at 6:58 pm.
SiriusHardware is online now  
Old 15th Jun 2020, 9:07 pm   #77
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

First hurdle cleared, using links to take b9, b10, b12 to +5V via pullups and b17 (top page) to b11 I get a nice rendition of the contents of 0f00-0fff (upper half) and 0b00-0bff (lower half). I won't bother with images just yet as Tim has pretty much been where I am now and posted the pictures to prove it. I must say I got a little bit of a rush from seeing my previously very neglected original MK14 VDU connected to an MK14 for the first time in more than 40 years - and very neatly too.

For the display I used my Philips CM8524 (colour) monitor and just jumped the composite-in signal going to the modulator straight across to CVBS-in on the monitor. The (mono) picture is very steady and quite sharp.

I think I am going to have to put an independent regulator on the 'bridge board' because the onboard one gets too hot too fast for my liking, even with only 7.6V regulated input to the onboard regulator - I'm not really surprised, that was the reason the regulator on my original machine was moved offboard and onto a much bigger heatsink.

Having reached this checkpoint I'll call it a night for tonight and tomorrow I'll probably try running 'Falling Man'.
SiriusHardware is online now  
Old 15th Jun 2020, 10:01 pm   #78
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

Excellent news, I bet you did - I get the same feeling each time I get one of my old machines running again - it is why we do it the nostalgia rush.

I had come to the same conclusion about the Regulator as well! I have burned my finger a few times disconnecting the MK14 and the VDU alone let alone with the RAM powered as well.

I shall look forward to a photograph of the setup running falling man as seeing an original VDU in use is definitely a major contribution... Like you I am calling it a night as well.
Timbucus is offline  
Old 16th Jun 2020, 7:35 pm   #79
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Mk14 vdu

Thought you might want some of the other software for the VDU I have typed in... I have a few more in progress as well.
Attached Files
File Type: zip VDUSW.zip (8.5 KB, 85 views)
Timbucus is offline  
Old 16th Jun 2020, 8:54 pm   #80
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Mk14 vdu

Well, I've just had 'Falling man' running, first time in ~41 years that my original Science Of Cambridge VDU has displayed those moving images... although I don't remember 'Brutal Deluxe' standing watching from above...

In the second image I happened to catch the moment where the falling man transitions to the standing man.

Looking closely I observe some artefacts or effects which I believe were present when I had the VDU hanging off a PIC as well, firstly very narrow vertical bands of black between each group of 8 pixels, and secondly the two small white lines, bottom left, which are no more than a quarter of a pixel high. Both of these effects may just be the Philips monitor's mild response to the lack of back porch, I'll have to bring down the contemporary B&W TV I keep in running order for this and the ZX81s to see if those artefacts are visible there as well.

Thanks for the entertainment package, I've been corralled for the rest of the evening but I'll give them a go closer to, or possibly at, the weekend.
Attached Thumbnails
Click image for larger version

Name:	FallMan_Original_SOC_VDU.jpg
Views:	109
Size:	53.5 KB
ID:	208658   Click image for larger version

Name:	FallMan_Original_SOC_VDU_2.jpg
Views:	109
Size:	47.4 KB
ID:	208659  
SiriusHardware is online now  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 9:42 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.