Forthcoming Cambridge CfCH events: RetroComp.Sale Sun 26th Feb, RCF-Japan 1st&2nd Apr

[ortek_service]

I've recently found out about these events being held there (often hidden in their half-term special-events email update I get), that may be of interest


http://www.computinghistory.org.uk/d...February-2023/

Sellers just pay £10 for a table. And it just says buyers pay on door, and get museum admission with no advance-booking for this event - So I presume that's actually the same current £10 admission price (although £8 for over-60's) for most people!


I saw this says 'No counterfeit or copyright-infringing' items are allowed, so I guess that technically could rule out any Replica PCB's! (Although I did read elsewhere on this forum, that if only one track is slightly-different from original PCB, then that didn't count as an exact-copy of the PCB at least) But presumably non-original DIY expansion board current-designs for these would be OK, even if not completely based on retro components.


And they are also doing a Japanese systems weekend event, in April: http://www.computinghistory.org.uk/d...nd-april-2023/
- I presume it's not just Games consoles, but earlier Sord / Sharp / Tatung Einstein + various MSX etc home-computers.

[SiriusHardware]

I think it's worth flagging these things up because the last big event resulted in a few more people finding their way here, due in no small part due to Tim's presence and the interest in the things he had with him on the day.

The counterfeit / copyright infringing clause is probably specifically targeting anyone trying to flog a barrow load of pirated software but you raise an interesting question about replica PCBs.

A couple of friends of mine had a knack of picking computers which never really became mainstream, so they had a Yamaha CX5(?) MSX computer with the MIDI / synth gizmo incorporated into it, and they also had a Tatung Einstein, the only one I ever saw or used. Both were nice machines, but under-supported by software in the UK so my friends eventually admitted defeat and bought an Atari ST.

The MSX concept in particular was a very laudable attempt to make all software usable on 'any' computer no matter which manufacturer you bought your computer from so it's a shame it wasn't more successful in the UK. I never knew anyone else who did have an MSX here in the UK.

[ortek_service]

Yes, although it seemed most at the last RCF were already into Retro Computers, there were probably quite a few who weren't that aware of things like the MK14 or current enthusiasm for using & developing these. So may be a more active Retro computer than some other models that have rather disappeared into being historical only. Although many Commodores are still quite active, with new hardware being developed & sold commercially (plus there being a fairly-new magazine for the Amiga (Addict)).

And I did discover the Newbear 77-68 there, that somehow I'd not ever known about (even though I later found I actually had an original main board from). So the RCF's are very-useful in discovering what others were currently up to.

I'm not so sure if this sale (the first one I'm aware of they've held) will be as good for that, if it's mainly people selling-off spare Beeb's & Spectrums etc. But I suppose there's nothing stopping people from having an MK14 on display, with details of where to get further info, even if not actually selling them (If it's all mainly cash-only, then might be a bit limiting to be able to sell more-expensive ones at the time, although it does avoid hassle of P&P if selling on online sites).

I've not yet decided whether to go, to see what sort of things people have even if not really going to buy any (well not complete computers, as not too many I'd want to buy that weren't cheap and in need of repair) or wait to see if they hold another one and hope there's some picture / reports from others as to what was there.


I did recently finally get round to repairing the Tatung Einstein I'd got given 25years ago, after discovering Chris also had (a working) one so could test the rare video IC, before tracking-down the several faults to three rather lower-cost & more available logic / Z80 Peripheral IC's I mostly had.
So it was good to finally be able to use one of these, that I recalled being launched but never being very-successful here, and were shortly-followed by enhanced models. They seemed to be an attempt to take on the Beeb, with lots of rather-similar I/O-Ports & BBC (Z80) BASIC. But with the built-in disk-drive maybe just meant the base cost was just too-high for many.
Plus never seemed to be much software released for it - probably due to small user base in a catch-22 situation.


Yes, although apparently very-successful in Japan, MSX never really caught-on in the UK where they were probably a bit too late amongst all the other platforms. And apart from seeing these featured in computer magazines, never saw one until I picked-up a used Toshiba one many years ago.
Although I do recall the Yamaha one was quite popular with Semi-Pro musicians, due to its enhanced sound capabilities (Although Atari ST's, with MIDI, were probably eventually much more widely-used for sequencing synths etc)

Of course the first attempt at standardising computers, CP/M, was also not that successful - especially in the home-computer market. Probably because it seemed every different model had an incompatible disk-format, and were maybe a bit too early with high-price and no graphics, for the (substantially games driven) home computer boom.
And maybe PC's would never have eventually become the norm in homes, if cheaper-clones of the v.expensive IBM's hadn't existed - Plus more computer-games being developed, then Internet availability to homes (although that has also allowed them to be widely-used for business-use from home).

[SiriusHardware]

One of the exceptions to the 'rule' that the Einstein never had any decent software released for it was 'Elite', the space trading / shoot'em up so if you want something to demo the machine with it might be worth looking for that.

CP/M, I agree it was lack of a common disc standard which doomed it.

[ortek_service]

Originally Posted by SiriusHardware

One of the exceptions to the 'rule' that the Einstein never had any decent software released for it was 'Elite', the space trading / shoot'em up so if you want something to demo the machine with it might be worth looking for that.

Thanks for info. I have Elite on its original Beeb platform, from back in the day (Think I also had a Spectrum version, that I recall wasn't as good And I seem to recall it really needed a disk drive, to load more levels etc. from).
(Its author, David Braben OBE, was also a co-founder of the RPi foundation and recall seeing him on TV news when RPi first launched).

I only really have the originally-supplied Tatung demo programs, plus a original MON80 Monitor-Assembler etc. tool.
So I will have to track down the Einstein-version, to see how it compares with Beeb one (In theory it should be similar, as the Einstein is quite-like a BBC Model B(+), but Z80-based. plus 64K RAM as standard that the early BBC B+ went from 32K up to (but extra 32K was only 20K video shadow-RAM + 12K workspace, so really needed a 6502 Co-Pro to have close-to 64K User RAM).
Hopefully the conversion was done fairly-well, as not all were if they were done by others.

I presume it runs on the standard Einstein TC01, and I doubt there was much that needed the 256K of the later Einstein 256 - Particularly as the extra 192K RAM on that was only for VRAM / RAMDisk and wasn't actually available for running programs from: http://www.tatungeinstein.co.uk/front/256.htm
Although > 64K Program RAM always seemed a bit unnecessary for an 8bit 4MHz Z80A - but it didn't stop the SAM Coupe, Spectrum 128/+2 etc. trying (However I don't think there was ever rally much software that used it).

[DrStrangelove]

Ah, the Einstein TC01: I have two, one has two 3" drives and two 5.25" drives, the other has a single 3" drive and a 40 track 5.25" drive.

I think I wrote* a transfer programme so I could read 3" diskettes from the Amstrad 256.

*Wrote it or found it somewhere. Transferred some Forth programmes off an Amstrad diskette for my then boss.

[SiriusHardware]

(Think I also had a Spectrum version, that I recall wasn't as good And I seem to recall it really needed a disk drive, to load more levels etc. from).

I suspect the Einstein version was a port of the Spectrum or more probably the Amstrad CPC version, perhaps the disc-based version for the CPC6128.

I can't play 'Elite' or any 'flight' type game with anything less than an analogue stick and unfortunately the Spectrum and Einstein both favoured switched sticks, so I found those versions 'unflyable'. I liked the original BBC B 'wireframe' version best of all.

[TonyDuell]

I have a Tatung Einstein. The seem to be some BBC-micro related jokes in the design, not least that the system bus connector is called the 'Tatung Pipe', obvously a pun on the Acorn Tube.

The Einstein's Joystick ports are most certainly analogue. There's an ADC0844 chip on the board, 2 channels go to each joystick socket.

The serial port is a 5 pin quincuncial DIN plug, but wired 'correctly'. Unlike on the BBC micro, if you turn the plug over in the Einstein it gives a useful pair of signal swaps -- TxD with RxD and RTS with CTS. I think the serial port used the 8251 chip and the DSR/DTR signals are available on marked solder pads on the PCB

There's a Z80A -PIO, one side of which is the printer port, the other side the user port.

The video chip is actually a standard Texas Instruments part (obviously the PAL version, the NTSC one -- 9918 I think -- is moderarely more common). It has its own 16K of video RAM. It gives YUV video outputs, I am told Tatung made a monitor for the machine to accept this. There's an RGB decoder on the PCB, you can set links to have its output on the monitor socket. Oddly the definitions for the user-defined keys are stored in an unused part of the video RAM and have to be accessed through the video chip.

Sound is an AY-3-8910. The 8 bit ports on this chip form the keyboard interface.

There is a fairly small ROM which can be paged out to have 64K of RAM on the Z80A.The ROM contains a disk bootstrap and a reasonable machine code monitor. BASIC, etc, is loaded from disk.

Finally, the power supply is a normal SMPSU which will be very familiar to anyone who worked on Decca/Tatung TV sets.

[SiriusHardware]

The Einstein's Joystick ports are most certainly analogue.

I was sure it used switched sticks - clearly a false memory on my part. In that case Einstein 'Elite' may actually be playable.

[TonyDuell]

It is of course trivial to interface a switched joystick to an analogue input, you just have the switches select one of a number of voltage levels. Somewhere I have a little commercial interface which lets you use an Atari-type switched joystick with a Dragon or TRS-80 Color Computer. It's just a few transistors and resistors inside.

I think I have (or if not, I've seen) a switch-type joystick with a DA15 plug for the BBC analogue input socket. The joystick switches were connected to a resistor chain and thus controlled the voltage to the analogue input pins.

I never had any games for the Einstein, not a joystick for it. It's entirely possble there were switch-type joysticks for said machine, or an interface for them. But analogue joysticks would be possible too.

[SiriusHardware]

I have an interface that I built to enable me to connect a switched stick to a PC 'game' (analogue joytick) input because while the PC uses analogue sticks, switched sticks are actually better for 'platformers' which typically require fast, precise input. As far as the analogue PC input is concerned the switched stick just looks like an analogue stick held fully right or left and fully up or down.

However, this would be useless for 'flying' games including Elite because the fully variable / proportional characteristic of analogue sticks is essential for those sorts of games.

[SiriusHardware]

Heading back towards the subject of computers of Japanese origin, I didn't even realise Sord were Japanese until someone here (probably Tony) mentioned it - I had always thought they were of U.S. origin.

The only other Japanese computer from that era that I personally encountered and used a little bit was a Sharp MZ-80K. It was a nice looking machine with a proper keyboard and a built in CRT monitor so it had a PET-like aesthetic but I seem to remember that even BASIC was an application which had to be loaded from tape.

An amateur radio club I was involved with had an 'old computers' night and among the machines brought in was a Sharp MZ-700 - again a nice machine but like the others, rather under supported by software in the UK.

[ortek_service]

My Einstein TC01 only originally had the single 3" S/S 40T? FDD A-Drive Fitted. But I've made up another longer cable to reach to the vacant B-drive side that Chris's already had, also adding 34w 2-row header IDC-sockets as used on 3.5" FDD's, as well as another 34w card-edge IDC connector. Annoying, I discovered these are on opposite-side + also the other way up between 3" FDD's & 3.5" FDD's, so I needed a bit of origami on the cable for it to line-up with the drives kept the correct way-up.

And I was able to run a 3.5" D/S 80T 720KB FDD OK, to copy some 3" ones to, as although I've quite a few 3" disks, most seem to be Amstrad CPC formatted and not many are known-blank at present - I also spliced into the drive / side select lines, adding switches, so I could swap A / B drives over, to boot from 3.5" / map to 2nd side of the 3.5" FDD to avoid wasting that space as well as tracks 41-80 (which unlike 3", you can't flip the disk over!)

You can just about get a 3.5" FDD to fit into the original 3" FDD bracket. And the Disk will just about clear front of case, if you remove the B-drive blanking panel. Ultimately, I'd also like to add a GOTEK, as well as original 3" + a 3.5", with switches to allow any combination of drive-letter mapping. But I need to work out best way to mechanically fit all this, without any cutting / drilling to original case / front panels etc. I don't think I'll bother with 5.25" FDD's, as not easy to fit one inside (And there is always the ext FDD connector).

I've had to use standard PC FDD Power-splitters / adaptors, as I haven't found exactly what the 0.2"? pitch power connector is that there's 2 of on the main board (look like they may have been used in CRT TV's for deflection coil connectors etc)


I seem to recall that Acorn made the 'Tube' a registered trademark, so presumed that's why Tatung called their expansion connector the Pipe - But not sure if there were ever any alternative processors, even though there was buffers all round the Z80A that could isolate it from the bus.

I think the Z80-PIO on mine had slightly-failed, but strangely caused the (more-recently developed by someone) Diagnostic-Test Utility (Needs original MOS one working) ROM I downloaded to fail at an earlier AY-3-8910 IC test, that would pass if the Z80-PIO was absent / swapped for another. But I can't recall if it stopped normal boot.

Yes, the VIDC was a standard TMS9918, but now quite rare and seem to be around £40 to get hold of one that hopefully is OK. Fortunately it was socketed (Only IC, other than ROM sockets, on mine), so was able to get Chris to try it in his (which had a few others socketed). Although I ended-up socketing all the large IC's, as well as CPU RAM's, trying to find why Diag-Test ROM RAM Test froze mid-way, seemingly an errant IRQ, until I found an LS157 Row/Column Mux wasn't switching properly, seemingly corrupting program's RAM.

Luckily the large SMPSU module all seems to have lasted OK with no faults developing - Not even any dodgy RIFA caps, to develop cracks in clear-epoxy case.

The large / deep TC01 case was designed to take a monitor. I think the links on mine are set to original YUV, rather than RGB, as had checked these when I first found it appeared dead with no video output. But the later Einstein 256 was rather less deep, so would need a shelf for one.

I think the Einstein's 15w D Analogue / Joystick port is compatible with the Beeb's - Except it probably lacks the light-pen input, that it seems was added
to the Einstein 256 on VAMP I/F: https://www.cpcwiki.eu/index.php/Tatung_Einstein

I do recall seeing a magazine project for a adjustable-time? ramping-voltage analogue-port interface for a digital joystick.
But the Altai ones I have for Dragon / Tandy are rather-more basic.

[ortek_service]

Originally Posted by SiriusHardware

Heading back towards the subject of computers of Japanese origin, I didn't even realise Sord were Japanese until someone here (probably Tony) mentioned it - I had always thought they were of U.S. origin.

The only other Japanese computer from that era that I personally encountered and used a little bit was a Sharp MZ-80K. It was a nice looking machine with a proper keyboard and a built in CRT monitor so it had a PET-like aesthetic but I seem to remember that even BASIC was an application which had to be loaded from tape.

An amateur radio club I was involved with had an 'old computers' night and among the machines brought in was a Sharp MZ-700 - again a nice machine but like the others, rather under supported by software in the UK.

I had originally considered getting a Sord M5, after seeing them advertised in the early-80's, as they did look a quite nice design. I do recall the unusual BASIC on a cartridge (Like earlier, equally obscure Exidy-sorcerer that had its own 35T FDD's).

I never quite got a Sharp MZ-80K, but I think I have some original service manuals etc and I did meet someone who ran a Sharp User Group, at a local 'all-formats' computer show in the 90's. But I have since acquired a rather-smaller MZ-70x with built-in cassette, that I need to try sometime.

I did get a brand-new in box 4-pen Alps mech wide 'Till-roll' plotter for the Sharp, as cheap surplus from J&N Bull Electrical in the 80's. And had rewired the original special-parallel lead connection to a standard 36w centronics connector, so I could use it on my Spectrum with a DIY Z80 PIO Interface / the Beeb I later sold my Spectrum to get.
So I will now have to look at restoring that back again, as luckily I'd kept the original cable.

[DavidMS]

Originally Posted by SiriusHardware

I think it's worth flagging these things up because the last big event resulted in a few more people finding their way here, due in no small part due to Tim's presence and the interest in the things he had with him on the day.

Yes I was one of those people, hoping to get there myself on the 26th

[SiriusHardware]

Just reading back here it seems there is an unexpectedly high incidence of Tatung Einstein owners here in our little group, maybe enough to justify a general purpose 'Tatung Einstein' thread for more detailed discussions about that machine.

[Mark1960]

Originally Posted by ortek_service

Yes, the VIDC was a standard TMS9918, but now quite rare and seem to be around £40 to get hold of one that hopefully is OK. Fortunately it was socketed (Only IC, other than ROM sockets, on mine), so was able to get Chris to try it in his (which had a few others socketed). Although I ended-up socketing all the large IC's, as well as CPU RAM's, trying to find why Diag-Test ROM RAM Test froze mid-way, seemingly an errant IRQ, until I found an LS157 Row/Column Mux wasn't switching properly, seemingly corrupting program's RAM.

Did you mean the original TMS9918 or the later TMS9918A?

I was able to get a few TMS9918As that seem to be genuine from ebay a few years ago. I was trying to get them working with more recent 4464 dram by delaying the controls and buffering the data, but without a scope at the time I couldn’t get the timing quite right to avoid memory corruption when the timing changed when the chips warmed up. I switched to the TMS9118 sourced from Utsource and this was more reliable as they were intended to run with 4416 dram.

I had started with an original TMS9918 that I bought by mistake from either RS or Farnell back in the eighties thinking it was the same as the A version. This earlier version doesn’t have all the graphic modes of the A version and also the dram timing may be a bit off, schematics that I found using this version seem to have capacitors on Control lines to trim the timing. This version without the A doesn’t seem to have been so well documented as the later A version.

I would expect any UK computer or console would actually be using the TMS9929A which was the PAL version, with additional circuit to provide composite output.

[Mark1960]

Originally Posted by ortek_service

I had originally considered getting a Sord M5, after seeing them advertised in the early-80's, as they did look a quite nice design. I do recall the unusual BASIC on a cartridge (Like earlier, equally obscure Exidy-sorcerer that had its own 35T FDD's).

I have a CGL M5 which was a version of the Sord M5 sold in the UK. Also the G-Basic cartridge which was a big improvement over the version originally supplied and made it possible to create quite good animation of the sprites and audio. The G-Basic cartridge also added extra ram. I haven’t powered it up since moving to Canada due to the 240v power supply brick and concern about power supply timing for the dram if I try to make a replacement. I also still have a few cassette games and the cassette recorder which no longer works since my son fitted the batteries reversed, mechanics still work so one day I’ll see if I can get a replacement IC for this. I disassembled the M5 and removed the internal RFI shields but still have all the bits except the screws to hold it together.

[TonyDuell]

I've just checked the hardware manual. The Einstein uses a TMS9129, which seems to be a TMS9929 variant to use 4416-type RAMs. The address is split 8-6 rather than 7-7 bits I think.

[ortek_service]

I'd just quoted the (TMS)9918 number Tony had mention earlier - which he thought was correct (and I knew it was a TMS9xxx number).
However, I've just checked and the my Einstein at least actually has a TMS9129NL fitted (along with 2x TMS4416 DRAM's for 16KB VRAM and some external RGB/YUB discrete-transistor (de)matrixing).

I'm not sure how the TMS9129 differs from the TMS9929, but have found this:
https://www.msx.org/forum/msx-talk/h...l-vs-tms9129nl

So it was probably actually the TMS9129 I'd searched for, and found it could be had for £25 many years ago, but now around £40 - Probably from a far-east source, that may not always be a genuine one.
- I discovered that many AY-3-8910 Sound IC's used in this are being sold as re-marked Yamaha YM2149's: https://www.bytedelight.com/?p=6303


It seems the Einstein 256 used the V9938, as used in MSX2 computers. With some screenshots of it in action, here: http://www.tatungeinstein.co.uk/front/256revealed.htm
- Although these probably aren't anything that the TC01's TMS9129 couldn't do?