Sinclair ZX80 / ZX81 / Z88 Forums

1024MAK wrote: ↑Fri Feb 24, 2023 10:52 am Timex Sinclair never released a TS2000 model (they had problems with complying with the regulations), so I can’t imagine that advert running for long. Timex Sinclair instead released the TS2068 in November 1983.

Oh, I didn't realize that was vaporware. I just thought it was referring to the TS2068.

1024MAK wrote: ↑Fri Feb 24, 2023 10:52 am As I said earlier, the cassette tape circuitry in the ULAs for the ZX Spectrum issue one and issue two boards is different to that in the issue 3 and onward boards.

Do you think a different ULA could cause ZX Slowloader to actually work in the Spectrum (i.e. without the extra circuit)? I ask because I have Charlie Ingley's ULA replacement and could try that. I have a 2nd broken 16K Spectrum that I haven't been able to debug and someone donated the ULA to see if that fixed it.

It's still a shame I can't locate even a picture of the device they used with the software. Probably was pretty simple and could be re-created, but always nice to at least document interesting software/hardware like that.

I’m not sure if Charlie Ingley's ULA replacement would help, I think it is based on the later ULAs. As later ULAs were backwards compatible with earlier issue boards. Although not backwards compatible with software written for the ULAs originally used on issue one and two boards.

There are brief details of the different ULAs and which boards they were normally used on here.
A 5C112E may not work in later issue boards, especially if it’s an issue 4A or later.

The only way to know if ZX Slowloader would work as you suggest, would be to try it. Without knowing more facts, we are just speculating…

Mark

That explains why it didn't work on my 48plus which has a 6C001E-7 ULA fitted.

I found the map with the slowloader code printed out.

First of all, the code is from a dutch guy called Ivo Breeden.

A google on his name and slowloader gave exactly 0 matches.

I could spent time analyzing how the loader works.
I only have the loading part, the converterpart must be taken from another slowloader.

I downloaded my own 81TOspeccie tool and in that tool it is mentioned that the SLOWLOADER from Ivo Breeden is from 1985.

I would suggest to download this loader and see if you can load and convert a program to my emulator.
If this works then you know that the loader will work and you can combine this loader with the converter from the other tool.

So I got ZX Slowloader to work on my 48K Spectrum with the vLA82 replacement ULA from Charlie Ingley. So indeed, the version of the ULA seems to matter. With the ULA clone, it was able to pick up the ZX81 audio at any level, but interestingly, only read it in when I moved it to about 80% (usually I load at around 95%).

Here's a picture of my original ULA to verify it's indeed an earlier version. It says 45th week of 1983, so I'm guessing that's pretty early on:

Absolutely wonderful that you have been able to get the program working and have loaded a ZX81 program

The 6C001E-6 like the one in your picture, was mainly used on issue 3 boards.

Later versions, as far as the cassette circuitry/function were the same.

It’s actually earlier ULA types, as used on mainly issue one and issue two boards that have differences in the cassette circuitry/function.

You would have to chat with Charlie to find out more about how he implemented the cassette circuitry/function on the vLA82.

Mark

I got a reply from Charlie:

The vLA82 analogue circuitry is fairly close to the later Sinclair ULAs but ended up slightly more sensitive. I'll send you the schematic...

Speaking of sensitivity, I also tried my 4th generation ULA (out of 5) with different volume levels (between 70% and 100% at 5% increments) and at no time did ZX Slowloader ever recognize a program was being loaded (i.e. no _ was seen). With vLA82, I was getting the _ every time, it just took me finding the 80% mark to get it to load without error.

I'm not one to read schematics, so let me know if anyone wants to take a look.

I’ve sent you a PM…

Mark

It could be the program was developed for a different zx spectrum issue.
Issue 1 and 2 have a different polarity and sensibility in the EAR input than issue 3 (and 4)