The most disappointing hardware for the ZX Spectrum?

[Hedge1970]

Can’t really comment much as back in the day I only had the Speccy and a tape drive. The Speccy broke so I guess that was the weakest link.

In this modern era it has to be the Microdrive. I was very excited to get one and bought endless cassettes off eBay, but then the reliability kicked in as I started using them to store game snap shots generated on my Multiface. I lost programming stuff and game saves . Thankfully vDriveZX was invented and all is good - thank you Charlie.

As to the ZX Printer I have the Alpha com and it still works flawlessly, has a ready supply of paper and is a god send when writing large programs on the Speccy. Emulators leave me a bit cold, I do use them (mainly for MC stuff) but if possible I always use real hardware.

[1024MAK]

An 8K RAM and an 8K of ROM containing a boot loader, a few fonts, and some optimised sound and graphics routines strikes me as the most effective arrangement, as it then wouldn't need paging. A bigger share like 1K ROM and the rest RAM would be even better, but even something like a 4K/12K split would mean the cost of two RAM chips PLUS a ROM at the time.

If a mixed arrangement is wanted, 8k of RAM (SRAM) and 8k of ROM (a 2764 EPROM) is easy. Other splits increase the chip count, and hence the expense.

Mark

[clebin]

I dunno if this is an oddball vote but I would say any kind of mouse! Some systems were more or less built to work with a mouse (Amiga, Acorn) and some just can't. What are you going to do with it? Art Studio? It's about as good as the light pen really. You only have a few things that recognise it, and those are slow and shitty too.

Have to disagree with you there. I had a Genius mouse that came in a bundle with The Artist II and I loved it. Had I had an Amiga to compare it to, I probably would've thought it slow & shitty, but as my only experience with a mouse I thought it was a dream.

Actually, the one thing I could compare it with was my Trojan lightpen and it was leaps & bounds better than that. Even the lightpen I had fun with, but I could tell the software was a bit ropey. I particularly remember how slow the flood-fill was - if you left a little gap in the bit you were trying to fill, you could wait 10 minutes for to finish filling the whole screen, line-by-line, until you could undo (at least it had an undo!)

Looking back, my black & white hand-scanner on the Amiga was a bit crap, but somehow I got plenty mileage out of it. I guess I'm turning into one of those old codgers who tell you how they used to spend hours playing with only a cup-and-ball...

[spider]

The idea of some kind of erm "shadow" RAM for the lower 16K once something else was up and running is a good one however costs I expect curtailed this idea.

I must agree (sorry) I do not like the + or toastyrack keyboard at all, when they age they tend to miss the odd 'extended' key , cursor keys especially can produce numbers instead as shift is not acknowledged. I can't honestly remember what a QL keyboard feels like, the last time I think I tried one was when they were on display (new) in WHSmiths!


Would be interested to know what was wrong with the Disciple interface, I always thought they were meant to be pretty good aside the cost.


My own: The Inteface 1 and Interface 2. The former for not having more developed networking software (Econet) but I do realise that is likely a ROM issue as its known as mentioned more was meant for the original ROM. And the latter, I think was missing some edge connector signals meaning it had to be last in line.

[Juan F. Ramirez]

I've always been curious to see if anyone could get one of these back in the day...



Cool, isn't it?

[Joefish]

I've always been curious to see if anyone could get one of these back in the day...



Cool, isn't it?

Judging by the drawing, it seems to be an identical to the gun supplied with the Grandstand series of TV games; we had the 'target shooting' barrel extender, but not the rifle stock.



It featured a lens over a single simple photoresistor. It could tell a light spot on the screen from a dark background, but would respond nowhere near fast enough to be able to do line-scan counting. I imagine the games were not much more sophisticated than the ones on that pong-like console, with a simple single bright object moving around as a target on a dark background. Or I guess one-at-a-time pop-up targets. Maybe it could do a quick flicker when you pulled the trigger to tell one object from another, but my guess is still single targets.

It would need some sort of interface to supply it with +5V and 0V to drive the detector circuit, and return two single bits of data for the trigger and the high/low state from the light sensor.

[1024MAK]

Interface 2. <snip> I think was missing some edge connector signals meaning it had to be last in line.

It only has a ZX81 sized rear edge-connector with only the signals needed for a ZX Printer…

Mark

[XTM]

I've always been curious to see if anyone could get one of these back in the day...



Cool, isn't it?

I know it's a slightly offtopic, but I can't resist googling places when there is an address, so it is safe to assume that ill-fated rifle was produced in this old building which apparently has been empty for over a decade, which is a bit sad ...

[PeteProdge]

I dunno if this is an oddball vote but I would say any kind of mouse! Some systems were more or less built to work with a mouse (Amiga, Acorn) and some just can't. What are you going to do with it? Art Studio? It's about as good as the light pen really. You only have a few things that recognise it, and those are slow and shitty too.

No.

In hindsight I really really should have got the mouse. I'm sure I tried out a mouse at a trade show. Naively, I didn't think it was the future of computing, but yeah, it truly was.

(I should add that in my Spectrum years, I did try to do some 'serious' computing, but there was barely anything that could be done. When I went over to the Amiga, I did loads of sensible work, got into desktop publishing, graphic design, writing, etc.)

[XTM]

The mouse wasn't the future of graphics work when it comes to drawing though, that really was the graphics tablet, so the light pen was at least a step in the right direction.

[MatGubbins]

Looking back, my black & white hand-scanner on the Amiga was a bit crap, but somehow I got plenty mileage out of it. I guess I'm turning into one of those old codgers who tell you how they used to spend hours playing with only a cup-and-ball...

Gotta agree with you there, I had the Power Computing B&W hand scanner, looked great but it took a lot of skill, timing and patience to get the results wanted. I went as far as making a Lego housing unit, guide rails with motor to make it smooth running - with a bit of sellotape to hold the scan button down. It worked but not as good as a real A4 flatbed scanner that I managed to get hold of much later - the mighty Epson GT-6500

The Spectrum interfaces that didn't have a through-port are the ones that I hated... Currah speech unit, Kempston interface, Interface III and probably a few more.

I borrowed a lot of interfaces from a mate, he had money to buy them and let me borrow a few when they were not used or got bored of them.
I do remember the Currah speech unit when playing Lunar Jetman, Interface III for copying games to microdrive (and tape!), Ram Music Machine for sampling noises and creating a music demo (sadly lost to time), Dktronics slow-mo joystick interface that was great to see how graphics were drawn to the screen, Com-Com joystick interface and the Dktronics lightpen.

The most disappointing from that lot it was probably the light pen interface. Yeah, it looked great for what it was, but of no real use. Melbourne Draw was streets ahead of that light pen.

I owned a Multiface 128, and later (around 1990) got the Interface one with microdrive expansion kit from a boot-fair for a tenner - all working with a handful of carts. Much later a K-55 interface!

[clebin]

I had the Power Computing B&W hand scanner

Same!

I went as far as making a Lego housing unit, guide rails with motor to make it smooth running

You're on your own there.

[equinox]

I dunno if this is an oddball vote but I would say any kind of mouse! Some systems were more or less built to work with a mouse (Amiga, Acorn) and some just can't. What are you going to do with it? Art Studio? It's about as good as the light pen really. You only have a few things that recognise it, and those are slow and shitty too.

No.
In hindsight I really really should have got the mouse. I'm sure I tried out a mouse at a trade show. Naively, I didn't think it was the future of computing, but yeah, it truly was.

I'm not saying mice aren't an important part of computing. I'm saying they sucked on a Spectrum...

(I should add that in my Spectrum years, I did try to do some 'serious' computing, but there was barely anything that could be done. When I went over to the Amiga, I did loads of sensible work, got into desktop publishing, graphic design, writing, etc.)

Did loads of Sensible Soccer more like.

HAHAH GOT 'EEEEM

[clebin]

I'm not saying mice aren't an important part of computing. I'm saying they sucked on a Spectrum...

Did you have one? EDIT:

[presh]

I've always been curious to see if anyone could get one of these back in the day...



Cool, isn't it?

I remember seeing this on The Spectrum Show:
https://youtu.be/oBtaqGaGrmU?t=227

I remember our family getting the "James Bond pack" one Christmas with the Sinclair light gun, the games did the horrible "screen flash" thing when you pulled the trigger... the Stack one doesn't seem to do that, which is a huge plus. Not sure how good it actually was though, doesn't help that all the games for it look godawful!

[Journeyman]

Nobody mentioned microdrives so far?

They were supposed by sir Clive to be a standard add-on to Spectrum which would be much cheapier than disk drives and yet load data in seconds.

They were a big failure as turned out to be unreliable, you could easily lose all your data only after using your microdrive several times.

Well, the cartridges were obviously the weak link in the system, but I've now got a vDriveZX which emulates it using SD cards. The thing is actually brilliant and makes using the Speccy MUCH better. I've dumped a load of games onto cartridge images, including 128K stuff like Where Time Stood Still.

The code for the Microdrive operating routines is far from brilliant because commands like CAT don't give you much info on the files, but there's programs to get around that. The concept was really good, although yes, there were flaws in execution.

[Journeyman]

Just a question for those who got to own a ZX Printer... was it worth having it?

The person I got one of my Spectrum+, and my IF1 & Microdrive from made use of it printing out nice file catalogues for all their cartridges and things. In some ways it's a cheap and nasty gimmick, but the printouts are just as readable now as when they were made, which is an advantage it has over similar thermal printers.
Obviously it's only useful if you want to print out program listings and things. If you bought one to do desktop publishing you'd be very disappointed

Oh, I don't know - I made a zine with an Alphacom 32!

[Journeyman]

(I should add that in my Spectrum years, I did try to do some 'serious' computing, but there was barely anything that could be done. When I went over to the Amiga, I did loads of sensible work, got into desktop publishing, graphic design, writing, etc.)

Likewise. The stuff you needed to bolt on to a Speccy to make it particularly useful was really expensive. The screen resolution and attribute clash issues were real problems too. I really wanted to use my Speccy to do useful things when I was a kid, but within a couple of years of launch, it really was only a cheap games machine, and doing anything creative or productive with it was a real challenge.

[JollyRoger]

You left out the lacklustre QL keyboard (which the ZX Spectrum+ keyboard was developed from), the non-standard video output, the bugs in the ROM and of course, the dongle! Why a dongle we still don’t know. The root cause being that the operating system (QDOS) and SuperBASIC combined were too large to fit in the allocated 32k bytes of ROM space. Hence an extra 16k byte ROM chip was needed. It was possible to include this extra chip inside the QL on later production machines, and this was done. I don’t know why it was not done on the first production machines. Nearly all of which were recalled by Sinclair, where these first production boards were scrapped.

True, the further development of the microdrive system increased their reliability. But commercial software companies could not duplicate microdrive cartridges. They had to rely on Sinclair…

I did not like the QL keyboard (still not keen on it), and hence I didn’t (and still don’t) like the ZX Spectrum+ keyboard.

Mark

Don't forget the non-standard ports on the UK version. A mind boggling blunder considering that this was a machine purportedly aimed at the business market, where adherence to industry standards would be crucial. The QL was also saddled with the Motorola 68008 CPU which has an 8 bit memory bus: crippling its performance and potential. David Karlin who was the QL's chief designer defended this choice as a budget conscious decision in order to keep the price down but as with many of Sinclair's cost and corner cutting measures, it was a false economy that hurt the machine's prospects in the long run.

From "The Quantum Leap - to where?"

In the event, in terms of both design and component costs, the difference is minimal. Choosing a limited chip rather than the full-specification 68000 version (now used by the Apple Macintosh, the Atari 5205T and other modern machines) is an example of Sinclair Research's incapacity to get it right when it matters. It ended up paying more or less the same price for a processor that does less, more slowly, than the correct choice. In any case, the price differences were a matter of a few dollars, and not of major significance to production costs. However the QL saga illustrates the capacity of Sinclair Research to make not only single wrong choices and assessments, but a whole sequence of them.

As to ZX Spectrum hardware disappointments, the inclusion of MIDI on the 128K+ model was impressive - particularly for the time but Sinclair went with a non-standard implementation for the port instead of the established 5 pin DIN standard which would've allowed musicians to immediately connect it to their gear rather than having to solder a cable or track one down.

[777]

yeah, i had a dk tronics light pen, it was awful. it used to flash the whole screen and was really hard to draw anything with. a useless piece of garbage.

[1024MAK]

A lot of the non-standard connectors on the QL and the ZX Spectrum 128k (toastrack) are actually a British Telecom standard. Unfortunately no one else in the computing market used them, so yes, that was a big mistake.

For joysticks and serial ports, 9 pin D connectors should have been used (the Sinclair Interface 1 used a 9 pin D connector for its serial port, although the pin out was unique to Sinclair) and the Sinclair Interface 2 used two 9 pin D connectors for its joystick ports, both of which were Atari joystick compatible). For MIDI, 5 pin DIN connectors should have been used.

The so-say advantage of the 68008 was not just the price of the microprocessor alone (although being in a 48 pin DIP/DIL package, it was supposed to be cheaper than the 68000 in the larger 64 pin pin DIP/DIL package, but that assumes large scale production of the 68008) but also because by using a 8 bit data bus, ROM / EPROM chips, SRAM and I/O chips designed for use with 8 bit microprocessors could be used.

If only eight DRAM chips were used (each providing one data bit), again that would have saved money (however the QL actually uses 16 DRAM chips to get 128k, so these could have been rearranged for a 16 bit data bus).

Also, a 8 bit data bus is easier to route on a double sided PCB. If a 68000 had been used, a three or four layer PCB may have been needed, which would have increased costs.

Light pens / light guns for the ZX Spectrum were always doomed to be disappointing. The ULA has absolutely no provision for a light pen / light gun unlike the 6845 CRTC used in some other computers, or the custom video chips used by other manufacturers. These chips have provision to latch the current video beam counter values when the button on the light pen / light gun is pressed. Hence the microprocessor can then read these values, and then easily work out which pixel(s) were being shown when the button was pressed.

On the ZX Spectrum, EVERYTHING has be be done in software by the Z80A. And as the Z80A has no access whatsoever to the video beam counter values, other tricks and techniques have to be used. But these are a poor substitute… An example of where software can never replicate hardware.

Mark

[Turtle_Quality]

Nobody mentioned microdrives so far?
They were supposed by sir Clive to be a standard add-on to Spectrum which would be much cheapier than disk drives and yet load data in seconds.
They were a big failure as turned out to be unreliable, you could easily lose all your data only after using your microdrive several times.

Yeah, I have to agree. Microdrive + "Interface 1" were the worst computing peripherals I ever bought.

Not for me, I was really happy to get high speed storage (relative to tape) at a low price. Using my compressor I could get a few games per cart, and it made testing my own code so much easier to be able to quickly save updates and reload when crashes occur. Luckily I had no issues with reliability either on Spectrum or QL, though admittedly I only used them a couple of years until floppy disks became an affordable upgrade.

[catmeows]

I wonder what would happen if effort put into development of microdrive was spent on developing cheap 5 inch drive.

[Matt_B]

The problem was that even a 'cheap' disk drive was around £200 in 1982/3, plus you'd need another £100 or so for the disk controller chip; Commodore put a CPU and some RAM in theirs instead because it worked out cheaper! Either way, a disk drive that worked out at 2-3 times the price of a 48K Spectrum was going to be a tough sell.

Making something that was cheaper but offered most of the same features was the right idea, and Microdrives could have been more of a success if they'd had reliable media from the outset. I got a couple of years out of mine before it started chewing cartridges, for what it's worth.

[catmeows]

I read somewhere that Apple was able to manufacture Disk II at cost $150, using Shuggart drives. Later they switched to cheaper Alps drives and pushed manufacturing cost to $100. I don't know how many pounds would it be in 1982/83. Less than 100 ?
Disk II is quite a hack but that is exactly what I would expect from Sinclair, a hack.