That was the reason I post edited (immediatly) the post to say "( on these memory chips)". I have had no problems with it on static rams, but some dynamic rams have special second uses on some of the pins and so this wouldn't be possible.
To do this trick on something that is not totally self contained ( ie programmed or read externally ) would be a simple form of encryption.
If you have gone through all the heartache of doing the layout, then just carry on.
The quick and dirty pin swapping was the only reason the "Baby" boards worked out so well. To do that without resorting to sneaky tricks would require professional vias etc.
I prefer the design and building, after that I often get restless and move onto the next project. I always have half a dozen things on the wish list.
Many thanks for your link. It is the first time I read a tutorial about designing a board and I find it quite useful. By the way, I learned we should use "thou" instead of "mil" because mil can easily be confused with millimetre.
And there was me trying to use "mils" because I thought that "thous" had gone into the past. My father always uses "thous" (ex Rolls-Royce aero engines).
It is interesting to see just how much of the stuff I do is classed as professional, and how little I do that is not (according to this artical). More luck than judgement, me thinks!
It is almost 2 weeks since I wrote something on the forum but I don't do electronics all the day long.
I started with a new design for the Memory board, taking into account the fact that I could mix up the data lines on one hand and the address lines on the other. I also wanted to make the board and the schematic consistent. This was a bit of a pain because the pins are not in the same order in the schematic and on the board and one is not allowed to draw the board first (at least with Eagle).
At the end, I realized that I could draw the board on half an European size (8 x 10 cm), so I thought it would be a waste to use a full size. This drove me to redesign the propeller board so that it could receive 2 half size boards. By the way, I restarted from the beginning to make consistent both the board and the schematic.
These are the resulting pdf files of the Propeller Module and the Memory Module.
The board of the Basic Propeller Module is finished (see joined picture).
I had a few trouble: at the beginning, there was a short made od solder between two tracks. I also had a bad solder on one of the VGA pins. But now, the VGA picture is perfectly stable.
Unfortunately, this is not the case for the keyboard/mouse. I don't detect any signal from them. Well, the first cause was that the pinout of my double PS/2 socket was wrong. I corrected the library (see attached file) and the board with a few wires. After correcting the board, I expected a functioning keyboard. I tried Keyboard.spin (in the Propeller Tool directory), as well as kbmtest.spin and PS2 HIDEngine Demo from the OBEX. For this last program, I get the basic screen (see attached picture) but nothing changes when hitting the keyboard. However, the LEDs of the keyboard flash every 2 seconds.
I verified that I have 5V on pin 4, GND on pin 3, 100 ohm between pin 1 and P26 and 100 ohm between pin 5 and P27, 10 K between pin 1, pin 5 and +3.3V. Really I don't know what is still to be tested.
I am at work, and will be tomorrow too (15 hour days), so I have no access to any Props to try and duplicate your setup.
The KBD shouldn't flash on and off and unfortunately it has been a while since I stuck a mouse up a Prop. The screen shown is obviously a mouse driver, I'll have crack at it in a couple of days.
With this screen, the leds of the keybord are not exactly flashing. It is the reverse: the leds are on and every 2 seconds they go off very briefly.
If I try "Keyboard.spin" from the Propeller Tool directory, I get a black screen and the LEDs of the keyboard do the same: the reverse of a 2 seconds flashing.
There must be something to do I ignore, an initialization or something else. I admit that I know very little in Spin language. For example, I am a bit surprised not to find at the beginning of the program a place to declare which pin will be used for the clock or for the data of the keyboard, a sort of "constant declaration".
This is the beginning of "Keyboard.spin":
''***************************************
''* PS/2 Keyboard Driver v1.0.1 *
''* Author: Chip Gracey *
''* Copyright (c) 2004 Parallax, Inc. *
''* See end of file for terms of use. *
''***************************************
{
REVISION HISTORY
v1.0.1 - Updated 6/15/2006 to work with Propeller Tool 0.96}
VAR
long cog
long par_tail 'key buffer tail read/write (19 contiguous longs)
long par_head 'key buffer head read-only
long par_present 'keyboard present read-only
long par_states[8] 'key states (256 bits) read-only
long par_keys[8] 'key buffer (16 words) read-only (also used to pass initial parameters)
PUB start(dpin, cpin) : okay
'' Start keyboard driver - starts a cog
'' returns false if no cog available
''
'' dpin = data signal on PS/2 jack
'' cpin = clock signal on PS/2 jack
''
'' use 100-ohm resistors between pins and jack
'' use 10K-ohm resistors to pull jack-side signals to VDD
'' connect jack-power to 5V, jack-gnd to VSS
''
'' all lock-keys will be enabled, NumLock will be initially 'on',
'' and auto-repeat will be set to 15cps with a delay of .5s
We are already in the PUB declarations. No trace of pin assignment. Is it because everybody is supposed to use the same pins for the same functions ? On the other hand, I use the same pins as the DEMO Board. So my keyboard should work fine.
I even tryed a IBM keyboard. Only the LED "Num Lock" was on (not flashing) but nothing happened when I typed the keys.
By the way, this is the modified layout after correcting the PS2 library.
The pins used in the KBD driver will be there somewhere, it would normally be in the calling routine or as a CONST right back at the begining. The KBD is not somehow locked to any pins, it is defined purely by software, I have pushed it over to the I2C pair via a 4053 switch after boot up, without any problems.
I can mimic your use of P16-P18 (rather than the usual P12-P15) for video quite easily on my "Lunchbox" so that will be done sometime tomorrow, hopefully.
At first, quick, look the mouse/ KBD conns look to be ok. I will have to load your code into my "lunchbox" demoboard look a like and fire it up fpr real.
Are you using the video or the VGA output ( your code should tell me, but I have several bits of "caring" yet to do ). Hey-Ho.
I am quite happy to tell you that you can stop searching the solution of my problem.
This afternoon, a friend of mine came with his computer out of order. I naturally took the keyboard I was using with the Propeller and . It was not working on his PC. I still had another spare one which I could use to repair his PC. He has just left and I immediately tried this new keyboard with the propeller. It works !!!
Digging hard into my drawers, I could find an old rolling mouse with only two buttons. It also works !!!
Wouwh. It was quite hard to find the error: I had to try 3 keyboards and 3 mice before finding the right ones. I was quite sure my printed board was wrong. I was also about to replace the Propeller (I have two of them). Anyway, it is really sympathetic to find guys ready to help another one. Thanks a lot to you. :yeah:
I originally got the KBD that I am typing this on for the Prop projects, but it seemed to have troubles. So I just kept on with the large ones. Fortunately The one small KBD I had that I wanted to use on the Z80 EMUs would work. Thats the one on "Clunky", the one that looks a bit like a C64.
(I still have a "list", caring just doesn't seem to have an end)
In fact, it already started quite a few weeks ago but is increasing as I go along.
What is the next goal?
The first goal is to check the SD card, not yet with CP/M as I need the memory module for that, but with another driver. For example, I think Kye has a driver. I need to read a bit about it. I am waiting for a 2 Gb Sandisk SD Card from Hong-Kong (much cheaper than what I can buy in Belgium).
The second goal is to make the PCB of the memory module. I have the connectors, the 138 and the 374's but I am still waiting the SRAM from Hong-Kong. It should come within a few days.
When the memory module will be ready, I shall certainly need your help to format, copy, configure, tune everything in order to run CP/M.
But I don't intend to stop there. I own 4 Z80 (CPU), 2 PIO and 2 SIO. I don't plan to build the whole N8VEM system but it would be interesting to interface the Prop and the Z80, expanding the possibilities of both.
I've got the feeling that the fun will still increase in the future.
That was the path that I was treading, I wanted to get a Z80 bolted to a Prop and get a Nascom running. Then Heater, Clusso, Dr_A and Pullmoll came along and made it possible, and a lot faster, to keep it all within the Prop.
Whilst the DracBlade is slower than the TriBlade for this it does neatly reduce the Prop count. Then I got cocky and thought I would truy and cut down on the ancillary chip count. This was the idea behind trying to get some anchient EDO DRAM going (allowing for the latches to be got rid off). I have some XC95xx chips that I could use instead and just keep the SRAM.
Lack of time, and lack of brain power, has left all that undone (so far). Jazzed has SDRAM pretty well going now, but I don't think that there were that many Z80 aplications that required 128MB.
Yes, it is perhaps difficult to say this, but I've run into some minor problems with MPM crashing every hour or two. CP/M is fine, but for MP/M, I need a solid platform that never crashes.
So today I got a large breadboard and some wires, and I'm going to try wiring up a hybrid Z80 and propeller circuit. In some ways maybe this is going backwards, but I can see some advantages too eg 1) we can push the speed up to 6, 8 maybe 10Mhz, and maybe 33Mhz with the Z180 (compared with the equivalent of about 1Mhz on the MPM emulation) And 2), we could devote all the hub memory on the propeller to video memory, and that means better graphics, and full colour text (pullmoll's code).
Re getting CP/M working, first step is Kye's SD card code. You can load KyeDOS and that will test the SD card is working. Then you can put the disk images on the SD card for CP/M, and load up the CP/M spin emulation. No need to reformat the sd card as the newer pullmoll code (and Kye code) works fine with fat32.
Only yesterday I wrestled with the urge (HERS) to clean out more "junk" from the garage. I just couldn't bring myself to chuck out an old PCB which has a Z180, 32KB ram, 128KB EEROM and a sort of 8255. I also have a couple of CMOS Z80 etc kicking about. The last time I looked the Z180s were getting hard to get hold of.
We are so spoilt in having chips now, that have virtually everything built in.
You wouldn't happen to have the schematic for that board by any chance? It is just that Andrew Lynch is working on a Z180 design but has got stuck with a few pins that he is not sure what to connect to. A working board would be very handy as a reference.
It seems that I left a problem to rush into another one.:hop:
I started with "SD2.0_FATDemo" of Kye, the version which can run without a RTC because I don't have one. After adapting the baud rate and the SD Card pins, I inserted a Emtec SD Card 1 Gb in the socket and downloaded the program to the Propeller. I managed to have a beautiful list of all the commands with "help". This was promising!
When I tried the command "mount 0", the answer was:
Running command: mount 0
Disk I/O Error.
I started verifying the electrical connections between the socket , the Prop and the power supply. No problem.
I verified that the layout was consistent with the Dracblade schematic and with the DS Card pinout. No problem.
The SD Card was OK because it is the Card of my camera and I can read the pictures I took yesterday.
Suddenly, I had an idea. I asked the properties of the Card and I saw the format was "raw". I was quite happy to find out why it was not working. I formatted it in FAT32 and had a new try :
Disk I/O Error.
I continued to read the thread of Kye about SD Card and came across the post #44 of Dr_Acula: http://forums.parallax.com/showpost.php?p=912882&postcount=44
Who proposes a modified version called KyeDos, with VGA and keyboard instead of the serial terminal..
But my hope was of short duration: the only thing I had was a blue monitor with a white blinking cursor in the upper left corner. The line "**** SD card operating system by Kwabena W. Agyeman ****" and the following ones never appeared. Reading the program, I saw " fat.mountPartition(0,0) ' mount the sd card" just before the line I was expecting. I guess this program line failed in my case so the program hangs.
So, what to do next ? I am expecting another card to come in the near future (sandisk 2 Gb). I hope it will not take me 3 different cards to find the one which works, like with the keyboard ! In the meantime , I will continue reading the thread. It is very interesting in itself but I dont expect finding a newer version of Kye's program because the last version is probably the one in the first post and the program has showed many times it works. The failure must come from the card or from my circuit.
So if you think of something I could check while I am waiting for the new card it would be very appreciated.
Yes, you are correct. If KyeDOS displays that one line, and nothing more, then the SD card is not mounting. It goes into an endless program loop trying to mount the card (maybe I should change the code so it prints something to say it is doing this).
A few things to try:
1) Try reformatting the card. You can do this in Windows, and use FAT32 (ie the default). It probably does not matter if you use quick format instead of the long format - I usually use quick format. I've had to do this many times. Pull out an SD card half way through a write, and it probably will need a reformat.
2) I have had one board design that did not work so well with the sd card. The only difference was the SD card was about 1cm further from the propeller, and the autorouter didn't put the power traces in the same place. For good measure, I've put a 33uF tantalum right across the SD card power supply pins, and also a 0.1uF bypass capacitor.
3) This one is unlikely, but recently, somehow I zapped one of the propeller pins that drives the SD card. I have no idea how that happened. So I fixed that with a new prop chip.
4) I was testing this just now and I found another fault. It was a micro SD in a standard SD adaptor, and the adaptor was faulty. Lucky I had another one to test it with, otherwise I would have thrown away a working sd card.
Also I tried downloading my own code from post #44 in that thread and it is not quite complete. See below - attached is the latest version
@Toby, Look into my eyes... Buy The Chips, Buy The Chips...
This forum is fantastic.
When I am too tired and go to bed, Dr_Dracula wakes up and continues thinking about the problem !
I downloaded the last version of KyeDos
I tried the point 1: reformatting the Card --> no change
Point 4 is out of question because I don't have any micro card or adapter.
Before trying my second propeller or soldering capacitors, I remembered that I had another card: a 2 Gb Kingmax.
I received the BS62LV4006PIP55 Low Power CMOS 512K SRAM IC and the memory module is finished (see attached pictures).
I am presently struggling to find out how to test it and run CP/M.
Dr_Acula, you said me a few posts ago :
Re getting CP/M working, first step is Kye's SD card code. You can load KyeDOS and that will test the SD card is working. Then you can put the disk images on the SD card for CP/M, and load up the CP/M spin emulation.
I downloaded from your site CPM_MPM-archive-19-Sept-2010.zip. This file contains 30 files, 29 of which being .spin files and one being cpm.binary
I changed the name of cpm.binary to cpm.bin, put it on the sd card, ran KyeDos and tried to run cpm.bin --> I got garbage on the screen (see fourth picture).
Is there a place where I can find the disk images your spoke about or should I compile some or all the spin files to binary and put them on the sd card before running CP/M ?
CPM is using the external ram, so if there is just one problem with the ram, then it won't run. So if there is a problem, we might need to do some low level ram testing.
Comments
To do this trick on something that is not totally self contained ( ie programmed or read externally ) would be a simple form of encryption.
If you have gone through all the heartache of doing the layout, then just carry on.
Drawing the layout is more a game than a heartache or a headache. And I am interested to see how the new layout is simpler than the old one.
http://www.alternatezone.com/electronics/pcbdesign.htm
In my opinion it is an inspiring document...
Massimo
I prefer the design and building, after that I often get restless and move onto the next project. I always have half a dozen things on the wish list.
Many thanks for your link. It is the first time I read a tutorial about designing a board and I find it quite useful. By the way, I learned we should use "thou" instead of "mil" because mil can easily be confused with millimetre.
It is interesting to see just how much of the stuff I do is classed as professional, and how little I do that is not (according to this artical). More luck than judgement, me thinks!
It is almost 2 weeks since I wrote something on the forum but I don't do electronics all the day long.
I started with a new design for the Memory board, taking into account the fact that I could mix up the data lines on one hand and the address lines on the other. I also wanted to make the board and the schematic consistent. This was a bit of a pain because the pins are not in the same order in the schematic and on the board and one is not allowed to draw the board first (at least with Eagle).
At the end, I realized that I could draw the board on half an European size (8 x 10 cm), so I thought it would be a waste to use a full size. This drove me to redesign the propeller board so that it could receive 2 half size boards. By the way, I restarted from the beginning to make consistent both the board and the schematic.
These are the resulting pdf files of the Propeller Module and the Memory Module.
The next stage should be etching both boards.
The board of the Basic Propeller Module is finished (see joined picture).
I had a few trouble: at the beginning, there was a short made od solder between two tracks. I also had a bad solder on one of the VGA pins. But now, the VGA picture is perfectly stable.
Unfortunately, this is not the case for the keyboard/mouse. I don't detect any signal from them. Well, the first cause was that the pinout of my double PS/2 socket was wrong. I corrected the library (see attached file) and the board with a few wires. After correcting the board, I expected a functioning keyboard. I tried Keyboard.spin (in the Propeller Tool directory), as well as kbmtest.spin and PS2 HIDEngine Demo from the OBEX. For this last program, I get the basic screen (see attached picture) but nothing changes when hitting the keyboard. However, the LEDs of the keyboard flash every 2 seconds.
I verified that I have 5V on pin 4, GND on pin 3, 100 ohm between pin 1 and P26 and 100 ohm between pin 5 and P27, 10 K between pin 1, pin 5 and +3.3V. Really I don't know what is still to be tested.
Any idea ? Thanks.
The KBD shouldn't flash on and off and unfortunately it has been a while since I stuck a mouse up a Prop. The screen shown is obviously a mouse driver, I'll have crack at it in a couple of days.
The screen comes from "#PS2_HIDDemo.spin":
Parallax Propeller Chip Project Archive
───────────────────────────────────────
Project : "#PS2_HIDDemo"
Archived : Sunday, August 8, 2010 at 6:02:53 PM
Tool : Propeller Tool version 1.2.7
#PS2_HIDDemo.spin
│
├──VGA64_TMPEngine.spin
│
├──PS2_HIDEngine.spin
│
└──ASCII0_STREngine.spin
If I try "Keyboard.spin" from the Propeller Tool directory, I get a black screen and the LEDs of the keyboard do the same: the reverse of a 2 seconds flashing.
There must be something to do I ignore, an initialization or something else. I admit that I know very little in Spin language. For example, I am a bit surprised not to find at the beginning of the program a place to declare which pin will be used for the clock or for the data of the keyboard, a sort of "constant declaration".
This is the beginning of "Keyboard.spin":
''* PS/2 Keyboard Driver v1.0.1 *
''* Author: Chip Gracey *
''* Copyright (c) 2004 Parallax, Inc. *
''* See end of file for terms of use. *
''***************************************
{
REVISION HISTORY
v1.0.1 - Updated 6/15/2006 to work with Propeller Tool 0.96}
VAR
long cog
long par_tail 'key buffer tail read/write (19 contiguous longs)
long par_head 'key buffer head read-only
long par_present 'keyboard present read-only
long par_states[8] 'key states (256 bits) read-only
long par_keys[8] 'key buffer (16 words) read-only (also used to pass initial parameters)
PUB start(dpin, cpin) : okay
'' Start keyboard driver - starts a cog
'' returns false if no cog available
''
'' dpin = data signal on PS/2 jack
'' cpin = clock signal on PS/2 jack
''
'' use 100-ohm resistors between pins and jack
'' use 10K-ohm resistors to pull jack-side signals to VDD
'' connect jack-power to 5V, jack-gnd to VSS
''
'' all lock-keys will be enabled, NumLock will be initially 'on',
'' and auto-repeat will be set to 15cps with a delay of .5s
okay := startx(dpin, cpin, %0_000_100, %01_01000)
PUB startx(dpin, cpin, locks, auto) : okay
........................
........................
........................
I even tryed a IBM keyboard. Only the LED "Num Lock" was on (not flashing) but nothing happened when I typed the keys.
By the way, this is the modified layout after correcting the PS2 library.
The pins used in the KBD driver will be there somewhere, it would normally be in the calling routine or as a CONST right back at the begining. The KBD is not somehow locked to any pins, it is defined purely by software, I have pushed it over to the I2C pair via a 4053 switch after boot up, without any problems.
I can mimic your use of P16-P18 (rather than the usual P12-P15) for video quite easily on my "Lunchbox" so that will be done sometime tomorrow, hopefully.
Are you using the video or the VGA output ( your code should tell me, but I have several bits of "caring" yet to do ). Hey-Ho.
I am quite happy to tell you that you can stop searching the solution of my problem.
This afternoon, a friend of mine came with his computer out of order. I naturally took the keyboard I was using with the Propeller and . It was not working on his PC. I still had another spare one which I could use to repair his PC. He has just left and I immediately tried this new keyboard with the propeller. It works !!!
Digging hard into my drawers, I could find an old rolling mouse with only two buttons. It also works !!!
Wouwh. It was quite hard to find the error: I had to try 3 keyboards and 3 mice before finding the right ones. I was quite sure my printed board was wrong. I was also about to replace the Propeller (I have two of them). Anyway, it is really sympathetic to find guys ready to help another one. Thanks a lot to you. :yeah:
I originally got the KBD that I am typing this on for the Prop projects, but it seemed to have troubles. So I just kept on with the large ones. Fortunately The one small KBD I had that I wanted to use on the Z80 EMUs would work. Thats the one on "Clunky", the one that looks a bit like a C64.
(I still have a "list", caring just doesn't seem to have an end)
What is the next goal?
In fact, it already started quite a few weeks ago but is increasing as I go along.
The first goal is to check the SD card, not yet with CP/M as I need the memory module for that, but with another driver. For example, I think Kye has a driver. I need to read a bit about it. I am waiting for a 2 Gb Sandisk SD Card from Hong-Kong (much cheaper than what I can buy in Belgium).
The second goal is to make the PCB of the memory module. I have the connectors, the 138 and the 374's but I am still waiting the SRAM from Hong-Kong. It should come within a few days.
When the memory module will be ready, I shall certainly need your help to format, copy, configure, tune everything in order to run CP/M.
But I don't intend to stop there. I own 4 Z80 (CPU), 2 PIO and 2 SIO. I don't plan to build the whole N8VEM system but it would be interesting to interface the Prop and the Z80, expanding the possibilities of both.
I've got the feeling that the fun will still increase in the future.
Whilst the DracBlade is slower than the TriBlade for this it does neatly reduce the Prop count. Then I got cocky and thought I would truy and cut down on the ancillary chip count. This was the idea behind trying to get some anchient EDO DRAM going (allowing for the latches to be got rid off). I have some XC95xx chips that I could use instead and just keep the SRAM.
Lack of time, and lack of brain power, has left all that undone (so far). Jazzed has SDRAM pretty well going now, but I don't think that there were that many Z80 aplications that required 128MB.
So today I got a large breadboard and some wires, and I'm going to try wiring up a hybrid Z80 and propeller circuit. In some ways maybe this is going backwards, but I can see some advantages too eg 1) we can push the speed up to 6, 8 maybe 10Mhz, and maybe 33Mhz with the Z180 (compared with the equivalent of about 1Mhz on the MPM emulation) And 2), we could devote all the hub memory on the propeller to video memory, and that means better graphics, and full colour text (pullmoll's code).
Re getting CP/M working, first step is Kye's SD card code. You can load KyeDOS and that will test the SD card is working. Then you can put the disk images on the SD card for CP/M, and load up the CP/M spin emulation. No need to reformat the sd card as the newer pullmoll code (and Kye code) works fine with fat32.
We are so spoilt in having chips now, that have virtually everything built in.
I have been musing on the RS site, there are Z180s in PLCC68 @ 33MHz. Hmmm
Note to self : DON'T DO IT ALAN!
It seems that I left a problem to rush into another one.:hop:
I started with "SD2.0_FATDemo" of Kye, the version which can run without a RTC because I don't have one. After adapting the baud rate and the SD Card pins, I inserted a Emtec SD Card 1 Gb in the socket and downloaded the program to the Propeller. I managed to have a beautiful list of all the commands with "help". This was promising!
When I tried the command "mount 0", the answer was:
Running command: mount 0
Disk I/O Error.
I started verifying the electrical connections between the socket , the Prop and the power supply. No problem.
I verified that the layout was consistent with the Dracblade schematic and with the DS Card pinout. No problem.
The SD Card was OK because it is the Card of my camera and I can read the pictures I took yesterday.
Suddenly, I had an idea. I asked the properties of the Card and I saw the format was "raw". I was quite happy to find out why it was not working. I formatted it in FAT32 and had a new try :
Disk I/O Error.
I continued to read the thread of Kye about SD Card and came across the post #44 of Dr_Acula:
http://forums.parallax.com/showpost.php?p=912882&postcount=44
Who proposes a modified version called KyeDos, with VGA and keyboard instead of the serial terminal..
But my hope was of short duration: the only thing I had was a blue monitor with a white blinking cursor in the upper left corner. The line "**** SD card operating system by Kwabena W. Agyeman ****" and the following ones never appeared. Reading the program, I saw " fat.mountPartition(0,0) ' mount the sd card" just before the line I was expecting. I guess this program line failed in my case so the program hangs.
So, what to do next ? I am expecting another card to come in the near future (sandisk 2 Gb). I hope it will not take me 3 different cards to find the one which works, like with the keyboard ! In the meantime , I will continue reading the thread. It is very interesting in itself but I dont expect finding a newer version of Kye's program because the last version is probably the one in the first post and the program has showed many times it works. The failure must come from the card or from my circuit.
So if you think of something I could check while I am waiting for the new card it would be very appreciated.
A few things to try:
1) Try reformatting the card. You can do this in Windows, and use FAT32 (ie the default). It probably does not matter if you use quick format instead of the long format - I usually use quick format. I've had to do this many times. Pull out an SD card half way through a write, and it probably will need a reformat.
2) I have had one board design that did not work so well with the sd card. The only difference was the SD card was about 1cm further from the propeller, and the autorouter didn't put the power traces in the same place. For good measure, I've put a 33uF tantalum right across the SD card power supply pins, and also a 0.1uF bypass capacitor.
3) This one is unlikely, but recently, somehow I zapped one of the propeller pins that drives the SD card. I have no idea how that happened. So I fixed that with a new prop chip.
4) I was testing this just now and I found another fault. It was a micro SD in a standard SD adaptor, and the adaptor was faulty. Lucky I had another one to test it with, otherwise I would have thrown away a working sd card.
Also I tried downloading my own code from post #44 in that thread and it is not quite complete. See below - attached is the latest version
@Toby, Look into my eyes... Buy The Chips, Buy The Chips...
When I am too tired and go to bed, Dr_Dracula wakes up and continues thinking about the problem !
I downloaded the last version of KyeDos
I tried the point 1: reformatting the Card --> no change
Point 4 is out of question because I don't have any micro card or adapter.
Before trying my second propeller or soldering capacitors, I remembered that I had another card: a 2 Gb Kingmax.
I let you admire the result.
It's Hallowe'en, he's busy.
Some cards just never seem to work, Sandisks are very good.
@Toby, yes I have been working a lot at night recently. The Transylvanian Blood Bank needs lots of donations.
I'm still chuckling about what Erco is doing http://forums.parallax.com/showthread.php?t=126739
I received the BS62LV4006PIP55 Low Power CMOS 512K SRAM IC and the memory module is finished (see attached pictures).
I am presently struggling to find out how to test it and run CP/M.
Dr_Acula, you said me a few posts ago : I downloaded from your site CPM_MPM-archive-19-Sept-2010.zip. This file contains 30 files, 29 of which being .spin files and one being cpm.binary
I changed the name of cpm.binary to cpm.bin, put it on the sd card, ran KyeDos and tried to run cpm.bin --> I got garbage on the screen (see fourth picture).
Is there a place where I can find the disk images your spoke about or should I compile some or all the spin files to binary and put them on the sd card before running CP/M ?
Your lights would be greatly appreciated.
CPM is using the external ram, so if there is just one problem with the ram, then it won't run. So if there is a problem, we might need to do some low level ram testing.