ZXblast - ZX81 memory extension plus USB
Re: ZXblast - ZX81 memory extension plus USB
Are you thinking what I'm thinking, Andy??
Re: ZXblast - ZX81 memory extension plus USB
Well this is something... whilst i have no doubt that the technique will work the question remains as to how likely (if at all) damage will occur to the cpu, many different types of cpu are in use in our beloved little black boxes.
regards Andy
what's that Smell.... smells like fresh flux and solder fumes...
Re: ZXblast - ZX81 memory extension plus USB
Then again CPU is pretty easy to replace if it's socketed ..
- 1024MAK
- Posts: 5153
- Joined: Mon Sep 26, 2011 10:56 am
- Location: Looking forward to summer in Somerset, UK...
Re: ZXblast - ZX81 memory extension plus USB
Then again, if you are overly concerned, you could always cut the A15 track near the CPU and wire a 220 ohm resistor in series, or bend the pin out and fit said resistor between CPU pin and the board...
Mark
Mark
ZX81 Variations
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp
Standby alert
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb
Looking forward to summer later in the year.
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp
Standby alert
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb
Looking forward to summer later in the year.
Re: ZXblast - ZX81 memory extension plus USB
That sounds like a great idea.
Regards Andy
what's that Smell.... smells like fresh flux and solder fumes...
Re: ZXblast - ZX81 memory extension plus USB
I think, that cutting the track of A15 near the ULA and adding a resistor and 2 diodes is also a great idea(*)
Siggi
(*) Usually called "M1 NOT circuit"
My ZX81 web-server: online since 2007, running since dec. 2020 using ZeddyNet hardware
http://zx81.ddns.net/ZxTeaM
http://zx81.ddns.net/ZxTeaM
Re: ZXblast - ZX81 memory extension plus USB
Well I am absolutely sure that this is no problem to the CPU. I think you can short A15 for a longer time with a multimeter and will not draw more than 10mA (if at all). So a resistor is useless. A continues short is not recommended - so this is a temporary short only.
As you see in the data sheets a high voltage of 2.4V is guaranteed for the CMOS version with maximum 1.6mA (0.25mA for the NMOS version). So in practice a short won't draw more currency than maybe 10mA. In fact I can measure about 10mA with a multimeter.
This is working in the same way as any pulsed LED which can be overdriven in currency for short moments. This depends mostly on the rate of hi-low ratio and frequency used. An example:
A half clock cycle is 158ns. This can be repeated maximum every 4 clock cycles (1230ns) and give a load of 1/8 or 12.5%. But 4 clock cycles is the shortest instruction while up to 16 clock cycles are also used like LD HL,(xxxx) or similar. So you can expect the ratio to be much lower in the range of 6-8%. In a maximum currency of 10mA this is comparable to a continues load of 0.7mA or so. Nothing spectacular.
Remember that the CPU is specified to deliver currency for a capacitive load of up to 100pF for the address bus. This is in the first 20ns with a currency of 10mA totally out of range but dynamically a normal situation. Guess what happen if 10 address lines changes will draw much more currency or for longer times. Yes, there are different output drivers but all output drivers have to be supplied internally with currency. So internal structures have to handle about 100mA currency for quite a lot of moments. Every capacitor draws currency - this is the life of all electronic devices. So a short is not comparable to other shorts. It depends always on the situation.
The load is only active if code is executed in address range $8000-$BFFF. If M1NOT is not used at all, there is no special load to A15. And people can setup ZXblast not to do so. In my eyes this is harmless. If you want to change the world you need courage and trust.
Decide now yourself if you want to be a coward or a hero.
May the force be with you !
As you see in the data sheets a high voltage of 2.4V is guaranteed for the CMOS version with maximum 1.6mA (0.25mA for the NMOS version). So in practice a short won't draw more currency than maybe 10mA. In fact I can measure about 10mA with a multimeter.
This is working in the same way as any pulsed LED which can be overdriven in currency for short moments. This depends mostly on the rate of hi-low ratio and frequency used. An example:
A half clock cycle is 158ns. This can be repeated maximum every 4 clock cycles (1230ns) and give a load of 1/8 or 12.5%. But 4 clock cycles is the shortest instruction while up to 16 clock cycles are also used like LD HL,(xxxx) or similar. So you can expect the ratio to be much lower in the range of 6-8%. In a maximum currency of 10mA this is comparable to a continues load of 0.7mA or so. Nothing spectacular.
Remember that the CPU is specified to deliver currency for a capacitive load of up to 100pF for the address bus. This is in the first 20ns with a currency of 10mA totally out of range but dynamically a normal situation. Guess what happen if 10 address lines changes will draw much more currency or for longer times. Yes, there are different output drivers but all output drivers have to be supplied internally with currency. So internal structures have to handle about 100mA currency for quite a lot of moments. Every capacitor draws currency - this is the life of all electronic devices. So a short is not comparable to other shorts. It depends always on the situation.
The load is only active if code is executed in address range $8000-$BFFF. If M1NOT is not used at all, there is no special load to A15. And people can setup ZXblast not to do so. In my eyes this is harmless. If you want to change the world you need courage and trust.
Decide now yourself if you want to be a coward or a hero.
May the force be with you !
Re: ZXblast - ZX81 memory extension plus USB
I have read this entire thread, and done a fruitless search, also i have tried to make sense of the german ZXBLAST thread, all without an answer.
So if anybody knows the answer then great...
The question i have 128K Ram zxblast running version 1.1D and despite all my attempts i can not get RAM at 8 - 16K to work.
if i select ram then 8-16K area appears to return 255 at every address
if i select rom or off then i can read what ever i have loaded into the Flash memory
Regards Andy
So if anybody knows the answer then great...
The question i have 128K Ram zxblast running version 1.1D and despite all my attempts i can not get RAM at 8 - 16K to work.
if i select ram then 8-16K area appears to return 255 at every address
if i select rom or off then i can read what ever i have loaded into the Flash memory
Regards Andy
what's that Smell.... smells like fresh flux and solder fumes...
Re: ZXblast - ZX81 memory extension plus USB
I have had problems with the 8-16k area as well. I seem to remember PokeMon has mentioned this in the German forum (maybe in a private post). It has been many months since the last firmware due to PokeMon having his hands full in his job.
I have not have any luck loading data to RAM in this area, due to the bug.
I have however had luck with:
I have not have any luck loading data to RAM in this area, due to the bug.
I have however had luck with:
- Loading to ROM in that area e.g. UDG chars
- Just setting it to RAM, and software that loads in 16-32k relocates itself to area 8-16k