Computers Overview
Commodore PET
Sinclair ZX80
Sinclair ZX81
BBC Micro
Sinclair Spectrum
Memotech MTX
    About
    Library
    Manuals
    Options
    Photos
        Disk Drives
        FDX
        Keyboards
        SDX
        Smörgåsbord
        Videowalls
            Dave 1
            Dave 2
            Mike
            Peter
    Projects
    Repairs
    Software
    Tools
    User Groups
    Video Wall
Memotech CP/M
Atari ST
Commodore Amiga
PDAs
DEC 3000 AXP
OpenVMS
Raspberry Pi

 

 
 
 

The Memotech MTX Series

Memotech Photos - Multi-Effect Video Wall

Dave's Video Wall

System 1

 

If you click on the images below, the full size picture will open, be aware, many of these are quite large and may be unsuitable for display on mobile devices or with a slow internet connection, but I wanted to include as much detail as possible in the larger photos

 

My Video Wall Hardware

This photo is of Video Wall hardware, listed on ebay in 2013 as a "Memotech Video Wall Controller 1980s / 1990s".

The advert described the item as having the 4 "control boxes" shown in the photo and advised that the seller believed that the Video Wall did not have a computer with it (the system being totally self contained).

After I had bought it, the seller found two additional "control boxes" and an MTX512S2 which he generously threw in with the other parts and delivered it all to me for no additional charge - thanks very much Kenny!

[Although these items were bought as a what the seller believed to be a complete system, the components appear to be from two Video Wall systems.]

This is the rear view of the four "control boxes" initially included in the auction.

The two bottom "boxes" appear to be similar to the two modules that make up Mike and Peter's Video Wall systems, these are described in the Cameron Video Wall manual and summarised on my MTX Options page.

The larger of the modules is the Video Wall controller itself, this houses the majority of the electronics, including the Frame Memory Buffer boards. In the Cameron manual, the controller is shown as having 8 monitor output connections, this one, like Mike and Peter's, appears to have 9.

The bottom module appears to be a Video Decoder module, this decoded composite video inputs and output RGBS to the Video Wall controller.

The two modules at the top of the photo have the large Memotech logos on the front, and at this point, were unrecognised - by me anyway.

Although Kenny had bought the Video Wall as complete and working, he had never actually tried to run it, but thought he had the interconnecting cables etc.

Part way through the auction, Kenny found this crate with a host of video and ribbon cables, as well as an MTX computer PSU.

Wow !! The kit that actually arrived! (not including the cables)

Apologies for the poor quality photo, I will take more once I get the items cleaned up a bit, they have been stored in a workshop, undisturbed, for the past ten years or so, and are somewhat dusty!

Close up / detailed photos - the full size pictures are quite large!

 

Computer - MTX512S2

(Working : missing two key tops)

The MTX512S2 computer used to control the Video Wall system.

It looks to be in reasonable condition, but unfortunately, it is missing a couple of keys, "3" from the main keyboard and "4" from the numeric keypad, the keyboard itself feels very good though.

Pushing the resolution of the graphics processor to its limit, the MTX512S2 starts up in 56 column text screen mode and boots into CP/M to run the Video Wall software.

(A description of his modifications to the SDX Boot ROM to allow it to run the Video Wall software from ROM are described on Andy's software page).

This screen is only displayed for the briefest of moment - taking lots of iPhone snaps to get this poor quality image!

And just to show that it actually works, this is the Video Wall software Main Menu - after pressing <ESC> (with no hardware connected, the software initially displays a "No Sync" warning and prompts you to hit <ESC> to continue).

Although you would obviously not be able to drive the Video Wall at this point, display sequences could be created, edited, etc.

This is a better quality capture (from Andy's MEMU page) showing the Menu more clearly - this is from a later software version than mine.

This version of the software allows you to drop into MTX BASIC if you quit the Video Wall application - mine reboots CP/M and reloads the Video Wall system.

The internals of the MTX512S2.

From left to right, the 4000-04 computer board, the Video Wall ROM board and the RS232 interface board.

The computer board.

The obvious differences from an original MTX are the socketed 256k RAM chips, the modified PAL (the chip on the left with the white label used for memory decoding) and the second black capacitor at the extreme right hand side of the board, this has been fitted in place of "D19" fitted to the earlier MTX models.

The ROM board containing two 32k ROMs holding the Video Wall software. This allowed the Video Wall to operate without the disk drive that was required for earlier versions of the Video Wall.

The ROM card has also been modified with additional yellow wiring.

The RS232 Interface board, this example has been fitted with the necessary chips to support the two available RS232 channels.

The rest of the board is unpopulated and was designed to hold the additional components and IDC header for the interface to the, by now discontinued, FDX.

Solder Side of the MTX512S2 computer, showing the RS232 Interface board, the Video Wall ROM board and the main computer board. You can see that both the ROM and main computer boards were modified with additional yellow wiring.
Close up of the yellow wiring for the RAM, this was required to allow more than 64kbytes of memory on the computer board. Andy has a complete description of the MTX series RAM configurations in the Memory Investigations section of his hardware page.

Only 64k is visible to MTX BASIC and in CP/M mode, only 208k is useable, with 64k for CP/M and the remaining 144k was normally used as a RAM disk.

Close up of the ROM board, also showing additional yellow wiring modifications.
The PSU for the MTX512S2, it is the standard PSU supplied with the later model MTXs from Memotech Computers Limited and has output voltages in the expected range. (See the PSU page for details - this one is listed as "ds5").

There has been a clumsy attempt to fit a noise suppression filter in the cable between the 13A plug and the PSU. The filter was wrapped with insulating tape, with the tape removed, you can see that the filter does not have an earth connection - probably because the PSU itself does not require an earth connection and finding one was "too hard".

The PSU is "double insulated" - signified by the symbol on the label :

( Double Insulated  - IEC Appliance Class 2 )

i.e., no connection to protective earth is required

The lack of an earth connection means that the filter doesn't !

 

The Video Wall Hardware

 

As noted above, the bottom two modules in this photo are a Video Wall Controller (DDFS) and a Video Decoder / Distribution Amplifier.

I originally thought that the top module was a component of a Reflex Touch Controller which would have connected to the RS232 port of the Video Wall computer. [A Reflex Touch Controller had a keypad that enabled the operator/DJ to use “quick action” buttons to initiate any of the video sequences stored in the computer without having to use the computer keyboard.]

The remaining three modules appear to be slightly different models of Video Decoders, the bottom two look like cut down versions of the one in the photo above and the top module, labelled "Decoder/Distribution Amplifier" looks like an expanded version.

 

The Video Wall Controller  - DDFS (Distributed Digital Frame Store)

(Had a faulty PSU, now replaced with a Micro ATX PSU)

The Serial Number label from the main Video Wall Controller (DDFS).

Obviously,  not a very interesting photo in itself, but it would be nice to know what information the serial number held. Probably Video Wall serial number 1451, but what the "/2", "/6" and "SR150" meant, I don't know. Based on the information in Andy's Video Wall brochure of the MemoPix Video Wall, perhaps the SR150 was a reference to the amount of memory on the Frame Buffer cards.

The Video Wall Controller (DDFS), with the top cover removed, showing the 2 x 6" card frames for the controller and frame memory boards.

You may be able to make out the small "Dymo" label on the front upper edge, it is marked "B", suggesting that this was only a part of a larger system.

The front cover removed, exposing the DDFS internals, the large module at the right hand side is the Astec PSU, mounted on the right hand side end-plate.

The end-plate has a single 4" air inlet grill and the PSU has a single fan as shown. The left hand side end plate has a pair of fans, ensuring good air-flow through the chassis.

Inside rear view, showing the connections on the rear panel, from left to right :

  • 2x15 way ""D" connectors for the RGBS inputs from the Video Decoder

  • 2 x 34 way IDC connectors from the MTX Centronics interface

  • 12 x 9 way "D" connector panel (9 used) for the RGBS video outputs

  • The AC power connection and power switch.

Outside rear view, showing the connectors described above.

The monitor connections are marked "4", "5", "6", "10, "11", "12, "16, "17" and "18", again, suggesting that the original system included another DDFS.

The DDFS Power Supply - an Astec "SA130-3400", as the data-plate shows, compared to an FDX, this is a quite "beefy" supply, 15A@5V, 5A@+12V, 0.7A@-5V and 0.7A@-12V and will probably generate quite a bit of heat.

The PSU is shown still mounted on the right hand end-plate, here, you can see the single fan on the internal side of the PSU. More PSU photos on this page.

Looking into the DDFS with the right hand end plate and PSU removed
The end covers and PSU removed, showing the card frames holding the controller and frame store memory boards.

The two lower boards are the DDFS controller boards, above them you can see 9 (5+4) frame store boards, each driving a single monitor, so this DDFS was capable of driving 9 monitors. The ribbon cables are all connected to the sockets on the rear of the chassis as shown above.

Close up of the left hand half of the card frame
With the boards removed, the 2 x 6" card frames can be seen in more detail - they appear to be the same as the single card frames used in the FDX.

Each card file has a 34-way ribbon cable for the connection to the MTX computer and a 9-way ribbon cable for each monitor output.

With the rear cover removed, the back side of the card cage is exposed, here you can see the power distribution, including the 12VDC supply for the ventilation fans.
And just to illustrate that the DDFS controllers were not immune from the "yellow wire" treatment, here are the two controllers from my DDFS.

You can see that the variable resistors on the right hand board are of a different type to the ones on the left and have been modified to suite the board with links and yellow wires.

Close up of the "FramseStore / AD+Controller" (FSAD) in the left hand card frame.

This is an "86/02 Issue 6 board", with what appears to be a Date Code of "89 13"

Solder Side - sans yellow wires!
Close up of the "FramseStore / AD+Controller" in the right hand card frame.

This is an "86/02 Issue 5 board", with what appears to be a Date Code of "88 13"

 
Solder Side - with a multitude of yellow wires
One of the Frame Store Memory Boards (FSMB)

The PCB has the title "Frame Store Memory Board - Double Resolution" and the ROMs are marked "S150", similar to the "SR150" part of the serial number. This is an "86/07 Issue 5 board", with what appears to be a Date Code of "89 14".

Tony Brewer has spotted that this board uses 20-pin 2167 SRAM chips which are only 16kx1 (128 x 128). The PCB allows for 22-pin 7187 64kx1 SRAMs (256x256) which are fitted to the boards in Peter's DDFS.

Tony also notes that the board also allows for 7-bit video, but only 6 bits are in use. The three sets of 74HC574 + SIL resistor network + transistor form the major part of cheap but effective video DACs for red, green and blue.

Solder side of the Frame Store Memory Board - with another yellow wire.

The FSMBs are uniquely addressed by configuration programmed into the PAL in IC27, in my DDFS, the addresses are as shown :

FSMB Address

Slot Left Side Right Side
5 FS138 FS132
4 FS137 FS130
3 FS136 FS129
2 FS134 FS128
1 FS133 (empty)
0 (FSAD board) (FSAD board)
 

 

Video Decoder & Distribution Amplifier - 1

(untested)

The Serial Number label from one of the Video Decoders.

Again, not a very interesting photo, other than to suggest that this is a PAL format Video Decoder and Distribution Amplifier.

This module also has a Cameron logo on it, although it is a completely different design to the one on the DDFS. It does not actually have the Cameron name on it, but it is the same logo as is used in the Cameron Video Wall manual.

The collection of equipment here is a bit of a "mixed bag", some items are badged Cameron, some Memotech and some, not at all.

If indeed this was a single system, I wonder if it was bought piecemeal, perhaps starting with a basic system and then adding to it later?

The front panel of the Video Decoder / Distribution Amplifier.
The rear panel of the Video Decoder / Distribution Amplifier. This module is slightly different to Peter's, in that it has more inputs and outputs, it has two (one) composite video inputs as well as a single (none) RGB input , it also has eight (four) RGB outputs.
Internal view of the Video Decoder / Distribution Amplifier, the Decoder board is the same as the one in Mike and Peter's Video Wall (below), but is using an RGB input from the rear panel the Peter's doesn't.

It has a more complex Distribution Amplifier board than the more basic version in Peter's, and as noted above, has 2 x composite video inputs, rather than the single in Peter's.

The lower board is the Distribution Amplifier (Issue 2).

In my system, the two composite video inputs from the rear panel are connected to the DA, with one output connection going to the Video Decoder board, rather than the input feeding the Decoder directly, as is done in Peter's system.

Solder side of the Distribution Amplifier board.
The upper board is the Decoder, the main component is a TDA3562A, "PAL/NTSC One-Chip Decoder." This separates the RGB signals from the composite video input.

This board is also common to Mike and Peter's Video Walls, although they appear to have been used in systems with different designs.

The Decoder board input connections from the Distribution Amplifier board, the phono socket on the right is for the composite video input.

The terminals at the left appear to be taking an RGBS input and separating the combined sync signal to produce Horizontal (Hs) and Vertical (Vs) sync signals - only the Hs signal made it onto the connector - Vs has its own screw terminal.

The Decoder board output connections..

The terminal block feeds RGB and separate sync signals back to the Distribution Amplifier board in my module. In Peter's Video Wall, the RGBS Decoder board input is not used and the RGBS outputs are fed to a much more basic Distribution Amplifier board.

Solder side of the Decoder board.

 

Video Decoder & Distribution Amplifier - 2

(Working)

The Serial Number label from the second Video Decoder

Again, not a very interesting photo, other than to suggest that this is also a PAL format Video Decoder and Distribution Amplifier.

Given the previous Decoder's configuration (8 outputs) and serial number (1107/8) though, we can deduce that the number (/4) after the serial number (1254) signified that it was a 4-channel output version, like Peter's. 

The Cameron logo from this Decoder module.

This logo matches the one on the DDFS, suggesting that they were supplied by Cameron at the same time.

The front panel of the second Video Decoder / Distribution Amplifier.
The rear panel of the second Video Decoder / Distribution Amplifier. This one is different again, it has 4 RGB outputs like Peter's, and also what appears to be an RGB input like the the first Decoder, but with the plug fitted upside down.
Internal view of the Video Decoder / Distribution Amplifier, the Decoder board is the same as the one in Mike and Peter's Video Wall (below), and again, has an RGB input from the rear panel, though it is connected very strangely!

It has the more complex Distribution Amplifier board, but only a single composite video input as well as an RGB input. In another "variation", typical of Memotech hardware, the RGB input has some "unusual" wiring connecting it to the Decoder board.

In the view above, the upper board is the Distribution Amplifier (Issue 3).

This time, a single composite video input from the rear panel is connected to the DA, with the output connection going to the Video Decoder board. This DA has only 4 RGB output connections fitted to the rear panel.

In this view, the lower board is the Decoder, again, the main component is a TDA3562A.

As mentioned above, the RGB input connection from the rear panel is very odd - it is connected to 10K variable resistors visible in the upper right hand corner.

 

Video Decoder & Distribution Amplifier - 3

(untested)

The Serial Number label from the third Video Decoder

 

We know to expect a PAL format Video Decoder and Distribution Amplifier

Knowing what we do about the Decoder serial numbers, we could deduce that it will be a 4-channel output version - so, just like one of the ones above?

    - Almost, but this is Memotech hardware! - so it has got to be different . . . .

The front panel of the third Video Decoder / Distribution Amplifier.

This one has the highest serial number so far, has a "Memotech Video" label and what looks to be a product name of "PAL DEC 15". Any suggestions for what DEC might have been an acronym for?

The rear panel of the third Video Decoder / Distribution Amplifier (with the cover removed). This time, the apparently upside down RGB input connection is labelled "Decoder Adjust", that makes more sense than it being an RGB input.
View of the rear panel of the third Decoder from the inside
Internal view of the Video Decoder / Distribution Amplifier. The Decoder board is the same as the others, this time, the single Video In connection from the rear panel is connected directly to it.

It has the more complex Distribution Amplifier board , but this board does not have any external inputs and just takes in the RGB output from the Decoder board.

In the view above, the upper board is the Distribution Amplifier (Issue 3).

This time, the DA board does not receive any external video inputs and the Phono input, as well as the unused RGB outputs, do not even have connectors fitted to the board.

Like the second  , this board has some very odd corrosion type deposits on RV1 and RV2, whereas RV3, RV4 and RV5 are unaffected.

I have no idea what may have caused this, overheating perhaps? - if you know what causes this type of deposit, please let me know.

In this view, the lower board is the Decoder, again, the main component is a TDA3562A.

With the "Decoder Adjust" label on the rear panel, the wires from the 15-way "D" connector to the variable resistors on the board now makes more sense.

 

Video Decoder & Distribution Amplifier - 4

(Seen enough yet? - I don't know why there are so many of the Decoder Modules - any ideas?)

(Working)

The Serial Number label from the fourth Video Decoder

 

We know to expect a PAL format Video Decoder and Distribution Amplifier

 

This one has a much later serial number and no "/x" after the number, it is an 8-channel output module though - perhaps this had now become a Memotech "standard" product - if ever they had such a thing !

The front panel of the fourth Video Decoder / Distribution Amplifier.

This one has the highest serial number and has a "Memotech" label in a very prominent position on the front of the unit. Perhaps this is the livery that Memotech used when they started marketing the Video Wall under their own name - rather than an agent - such as Cameron?

Close up of the "Memotech" logo
The rear panel of the fourth Video Decoder / Distribution Amplifier.

This time, there is a 15-way "D" connector for "Decoder Adjust", as well as an additional connection marked as "RGB Input". The unit has 8 populated RGB output connectors, panel space for 4 more and 2 BNC Video inputs.

Unlike the rest of my Video Wall equipment, this Decoder, and the module below, are both about 2" narrower than the free standing units, and have tapped holes in the sides to facilitate mounting in a rack of some description.
View of the rear panel of the fourth Decoder from the inside

You can see that the basic transformer used on the previous versions has been removed, a "proper" power supply is installed in the square steel tube on the right hand side.

Internal view of the Video Decoder / Distribution Amplifier.

The is orientated differently from the previous examples, making for neater installation of the RGB output ribbon cables - they now pass under the board for connection to the rear panel.

The PSU removed from the steel mounting tube

 

The PSU was manufactured by "Skynet", Model Number SNP-3032

 

Input specifications: 120V-1A, 240V-0.5A, 50/60Hz.

Output specifications; V1: +5V/2A, V2: +12V/1.5A, V3: -5V/.3A.

Solder side of the PSU
The Decoder board from fourth Video Decoder / Distribution Amplifier.

You can see a large capacitor on this board that must have been added as part of the PSU modification.

A close up of another strange modification to this Decoder board, all of the Decoder boards have fixed and variable resistors soldered to some of the legs of IC2.

This one has the odd "furring" on the surface of the variable resistor. It may be something to do with the particular resistor type, as it always seems to be the black ones that display these surface "deposits".

The Distribution Amplifier from the fourth Video Decoder / Distribution Amplifier.

Although still labelled as an "Issue 3" board, all of the power supply components have been omitted since their function has been replaced by the new PSU.

 

The Reflex Controller

(Working : after the failed PSU and RAM were replaced)

The last "black box" in my Video Wall bundle. Yes, another label photo - but at least this time, I think it, or rather, they are definitely worth showing. The top label describes it as a "10 x 1 Reflex", the bottom label ("MTX2019R") is the model number and is the same as the model number printed on the rear of the case.

This is the only item in my Video Wall collection that I have not previously at least seen a photograph of. In the April 1990 issue of Lighting & Sound International", available on the Articles page,  the installation of a Memotech Video Wall as part of the sound & lighting upgrades to the Hammersmith Palais in 1990 is described. The article briefly mentions the "Reflex Touch Controller", it allowed the operator/DJ to instantly access pre-programmed Video Wall sequences using a 32 button keypad. Unfortunately, my system is missing this separate keypad.

(Originally installed by Cameron as two 5x5 Video Walls, the Palais system was upgraded by Memotech to a single 10x5 Wall, at that time, it was the largest club Video Wall in the UK.)

The front panel of the Reflex Controller.

In the same livery as the Video Decoder / Distribution Amplifier above.

The rear panel of the Reflex Controller.

With connectors for 2 x RS232 ports, Centronics interface, Video Out, as well as one marked "TC". This was for an EBU/SMPTE Time Code input.

Inside view of the Reflex Controller, showing a similar PSU as the Decoder above, it is a "Skynet", Model Number SNP-3031.

Input specifications: 120V-1A, 240V- 0.5A, 50/60Hz.

Output specifications; V1: +5V/2A, V2: +12V/1.5A, V3: -12V/0.3A.

 

Does the main PCB look vaguely familiar? It should do, it is a MTX 4000-05 Computer board, with 256k of memory, but with no Video board and a previously unseen combined RS232 and ROM board.

Component side of the MTX Computer and RS232/ROM boards.

The external Time Code input is connected to the MTX Tape Input jack and with no Video board fitted, the monitor output is connected to J3, where the MTX video board would normally have been connected.

Solder side of the Reflex Controller circuit boards
The 4000-05 Computer board with 256k of RAM

A mono video signal is being "fudged" by taking the Component video "Y" signal from J3 Pin 5 directly to the Video Out connector (see my MTX Video page).  

With the DC PSU fitted, the regulation and smoothing components are not fitted to this board - a model for the proposed MTX PSU replacement I wonder?

Solder side of the MTX 4000-05 Computer Board, as it has 256K of RAM installed, the expected yellow wires are present too.
The combined RS232 / ROM board
Solder side of the RS232 / ROM board
When I tried to power it on, I discovered that the Reflex Controller had a failed PSU and subsequently discovered that faulty RAM on the MTX computer board also prevented it from working, until I repaired it as described on this page.

The Reflex was configured to operate from ROM; on start-up, a RAM disk was created from the computer's RAM and the Video Wall operating program and sequences were copied form ROM to RAM, the software auto-started and the Reflex was put into remote mode to allow control of the Video Wall from an external device / computer. The expected prompt monitor display is shown here.

With no DDFS(s) connected when the Reflex boots, a "NO SYNC" error message is displayed on the prompt monitor, along with the "HIT (ESC)" prompt. This would have been problematic when the Video Wall was in use as the Reflex does not have a keyboard or even a keyboard connector fitted. With this error displayed, I found that it is not possible to "Take Control" of the system remotely.

During my testing though, I was able to have an MTX keyboard connected via the usual MTX computer board keyboard header.

On pressing the <ESC> key, the normal Video Wall "Main Menu" is displayed.

Function key "F8" calls up the "Special Options" Menu, from where, remote operation can be turned on using the "F1" function key.

 

My Video Wall : Likely Original System Configuration(s)

Based on the hardware that I have, the software configuration in the MTX ROMs and the limited Memotech product documentation that is still around, it is possible to make some informed guesses on how the hardware was originally configured.

 

On further investigation, I believe that the components are from two different Video Wall systems :-

  • A 4x4 Video Wall, controlled by the MTX512S2 computer (but missing 1 DDFS)

  • A 8x8 Video Wall, controlled by another computer through the Reflex controller (but missing 8 DDFSs!)

Common Components

Assertion

Supporting Evidence

Hardware Redundancy

Each DDFS had a spare FSMB installed

Each DDFS has 9 FSMBs, but the software is configured for 16 monitors.

The Video Wall Product description explains how to reconfigure the software  to use a "spare" FSMB in a DDFS to take over from a failed FSMB. Looking at it very simplistically, a FSMB failure is 4 times more likely than an FSAD failure, so fitting a "hot spare" FSMB makes sense.

Video Resolution : 6-Bit The FSMBs have 6 x 2167 RAM chips for each colour (with space for 7)
 
System 1

(possible configuration)

Assertion

Supporting Evidence

Video Wall Size : 4 x 4 (16 monitors) The Video Wall software in the MTX512S2 ROM is configured for a 4x4 wall

The DDFS is labelled "B", strongly suggesting that there was at least an "A" too

D/DA Number 2 would have been able to drive a 4 x 4 Video Wall

 

System 2

 (possible configuration)

Assertion

Supporting Evidence

Video Wall Size : 8 x 8 (64 monitors) The Video Wall software in the Reflex ROM is configured for a 8x8 wall
D/DA Numbers 1,3 & 4 would have been able to drive a 8 x 8 Video Wall
 
So, that's it - the last of my, incomplete, collection  of Video Wall hardware!

Conclusions :

  • I have a number of components from at least two different Video Walls.
  • I am missing at least 1 DDFS that would be necessary to build a Video Wall using the MTX512S2 ROMs.

It am also left with a number of questions  . . . .

  • Where was the hardware originally installed?
  • Is the equipment from two different systems, or just a collection of unrelated hardware?
  • But the big one is - can I get it working again !

 

 

mailto: Webmaster

 Terms & Conditions