ZXmore or the ZX80CORE follow-up

[PokeMon]

I find it a bit strange that you chose the ZX-80 layout for the board connectors, given many more ZX-81s were produced (1 500 000 or so) than there were ZX-80s (100 000), but it's of course your choice.

I'll just have to drill more holes in the Indescomp box, I suppose...

This has historical reasons as I developed a ZX80 replica before (ZX80CORE) which is 100% hardware compatible to the ZX80 case.
Now the development went further with new functions. I doubt it would match into a ZX81 case this way. And the connectors of ZX80 and ZX81 are 100% compatible except that the ZX80 didn't have a ROMCS on the connector (left free). It was jumperable at the previous version but there is no known conflict with hardware so I decided to put it fix on the edge connector. That way it is more ZX81 than ZX80. Anyway all hardware of ZX80 and ZX81 are more or less compatible (except the ROMCS issue) and could be used both. On the other hand this was not much hardware published for the ZX80 - the only one I know is a ram pack (3k).

To be 100% sure it will load with Fast Load Programs I would propose to do a test. I don't know Fast Load Monitor - is it just software or hardware ?

This is very kind and considerate of you, and it's 100% software... However, I only have it on tape, so I can't send you a binary (not to mention the manual is written in French)... This said, if the audio circuit bandwidth of the ZXmore is of equal or better quality than the ZX-81's, and given the routines are the same, it should work just fine.

Maybe it is possible for you to digitize it while recording it with Audacity (free software on WIN - not sure what to use on Linux) and send some data in MP3 format with email. Then I could try to load it in my ZXmore. The audio characteristic is similar except that the output level is increased from ZX81 level of 5mV level to 50mV level. It would be not a big deal to increase it more to 500mV when changing one resistor and capacitor.

The audio file must contain a two stage loader as it is not possible to load data at higher speeds by default. I used this in my ZX-IDE development tool but there are some problems when loading with this high speed of up to 1 kByte per second. Normal speed is about 38 Byte/s. Something inbetween would maybe better, maybe my implementation is not best choice and maybe poorly designed.

[dinosaur]

Maybe it is possible for you to digitize it while recording it

Err... Do you suggest me to connect the EAR output of the tape recorder to the MIC input of my sound card ?... Won't this require a level adaptation to avoid destroying the sound card input ?... My guess it that would indeed be needed, but then I must make an attenuator that will preserve the signal bandwidth (a simple voltage divider with resistors won't do it).

with Audacity (free software on WIN - not sure what to use on Linux)

I won't be surprised if Audacity started as a Linux software... Yes, it always existed under Linux and is, of course, free.

and send some data in MP3 format with email.

MP3 would be a very bad choice... It's a lossy format. WAV is the way to go to preserve the full signal.

Then I could try to load it in my ZXmore.

That would be very kind of you. I just managed to re-read my old tapes (had to readjust the taper recorder head azimuth and to find the proper sound level for the output...).

The audio characteristic is similar except that the output level is increased from ZX81 level of 5mV level to 50mV level. It would be not a big deal to increase it more to 500mV when changing one resistor and capacitor.

The output level from the ZX to the tape recorder is not much of a problem, in my experience (especially since most tape recorders got an automatic level adjustment via a dynamic compressor), but rather the output level from the tape recorder to the ZX (a tiny bit too loud or a tad bit too low, and the loading fails).

Note also that, after all, all what actually matters, is that I could re-read the FLM tapes with the ZXmore (I could then save the programs via the USB and be done with the old tapes).

The audio file must contain a two stage loader as it is not possible to load data at higher speeds by default.

Yes, the FLM loads itself in two stages: a first fast loader "boot" program that loads at the ZX-81 standard speed (250 bauds) and then the program itself, loading at full speed.

[IanB]

Do you suggest me to connect the EAR output of the tape recorder to the MIC input of my sound card ?

Your sound card should have a line level input which will accomodate a line level signal.

[Andy Rea]

So i modified my sound-card tied the /RD to +5v so it will never try to put data on the bus

no use, ZXmore still does not boot

i will have to program a plain ZX81 rom and try manual instance selection for maximum compatibility, but how is now the question ?

regards Andy

[PokeMon]

Hi Andy,

you could try simply to use the rom switch in position 1 which use the ZX81 rom directly with no ZXmore software / feature / addon's.
Just to see if it is a hardware or software issue.

Karl

[dinosaur]

Do you suggest me to connect the EAR output of the tape recorder to the MIC input of my sound card ?

Your sound card should have a line level input which will accomodate a line level signal.

Found it !... In Linux advanced mixer, it was configured as an output (HD audio chips allow this kind of things), thus why I didn't see it, at first.

I made a couple of *.wav files (one with the FLM software itself, and another with a FLM-saved program, to test it with). Let me know your email address in PM, and I'll send you the files together with the "recipe" for testing them.

Another question about the ZXmore: is it possible to configure it so to let 4Kb "holes" between the ROM top address and the RAM bottom (i.e. in the 8-12K and/or 12-16K) range ?... Some addon cards use these for their own ROM (got an old EPROM programmer that uses this trick, and so did the Memotech HRG module).

[PokeMon]

Another question about the ZXmore: is it possible to configure it so to let 4Kb "holes" between the ROM top address and the RAM bottom (i.e. in the 8-12K and/or 12-16K) range ?... Some addon cards use these for their own ROM (got an old EPROM programmer that uses this trick, and so did the Memotech HRG module).

No - this isn't implemented due to an easy configuration management.
But it is possible for external interfaces to disable ROMCS or RAMCS full or partly (in an address area) to achieve this feature.
The same technique is necessary for the ZX81 for example as there is never a hole in address space but mirrored address areas.
They have to be switched off, too, for proper function.
So in general all hardware should be able to run with the ZXmore.
PN follows.

Still waiting on feedback from Andy testing his soundcard in switch position 1 or 2.
By the way - it would be possible to remove IC22/IC24 - the two registers used to control RAM/ROM and other features from ZXmore software for another test. This way it would work in manual switched position (ROM switch) only.

[Andy Rea]

Hi Karl,

some strange goings on.
Firstly i had a bad solder on the memory chip A10, i think was just the cause of random crashes.

for sound-card problems i first tried the switch in position 1, it powered up ok first time, but then after power cycling it fails to start. i noted that the LED would light all random colours at each power cycle. so i remove ic22 problem goes away and it starts fine at each power cycle. so looking at the diagram i see that ic22 sets the active ROM bank ( bu only able to pull lines low due to resistors ) and it in turn is latched by a signal from the CPLD/

so i put ic22 back in and switch back to position 0 power up its ok.. huh I'm confused... poweer cycle... broken... what is going on..

and then i realize the problem is me not allowing enough time for the power rails to discharge enough for a proper reset .

so um yeah i can get it to power up ok now, so further testin is needed.


Regards Andy

[PokeMon]

Hmmm - so it works in position 1 but not in position 0 ?
So ZXmore software do not startup or maybe try again with soldered A10 ?

[Andy Rea]

Sorry Karl, it works in both position 1 and 0 if enough time is left between power cycles


Regards andy