EPROM instead of PLA
Ville Jouppi
(or can I just read it from a working PLA?)
I seem to remember there being talk that this can be done.
--
CBM, PEZ, and TI-calc nut, Scout, Glider pilot, Helpdesk-person
Email: vjo...@sci.fi, URL: http://www.sci.fi/~vjouppi/
GSM: +358-40-5679999, IRCNet: Jope
"I see", said the blind man as he picked up his hammer and saw.
Spiro Trikaliotis
Ville Jouppi wrote:
>Ok, where could I find the type of the eprom chip and the rom code?
>(or can I just read it from a working PLA?)
I haven't tried it, but the following should apply:
First, we have to find a suitable EPROM type.
N.C. 1 FE Vcc 28 Vcc
A13 2 I7 I8 27 A12
A14 3 I6 I9 26 BA
A15 4 I5 I10 25 -AEC
-VA14 5 I4 I11 24 R/-W
-CHAREN 6 I3 I12 23 -EXROM
-HIRAM 7 I2 I13 22 -GAME
-LORAM 8 I1 I14 21 -VA13
-CAS 9 I0 I15 20 -VA12
-ROMH 10 F7 OE 19 GND
-ROML 11 F6 F0 18 -CASRAM
-I/O 12 F5 F1 17 -BASIC
GR/-W 13 F4 F2 16 -KERNAL
GND 14 GND F3 15 -CHAROM
The first thing we see: There is no equivalent EPROM type
(AFAIK), so we need an adaptor.
Lets look at the 27512 (which is 64Kx8, so it has the required
capacity):
Its pinout is as follows:
(FE) 1 A15 Vcc 28 (Vcc)
(I7) 2 A12 A14 27 (I8)
(I6) 3 A7 A13 26 (I9)
(I5) 4 A6 A8 25 (I10)
(I4) 5 A5 A9 24 (I11)
(I3) 6 A4 A11 23 (I12)
(I2) 7 A3 E2 22 (I13)
(I1) 8 A2 A10 21 (I14)
(I0) 9 A1 E1 20 (I15)
(F7) 10 A0 D7 19 (OE)
(F6) 11 D0 D6 18 (F0)
(F5) 12 D1 D5 17 (F1)
(F4) 13 D2 D4 16 (F2)
(GND) 14 GND D3 15 (F3)
We see that the pinouts are similar (but not identical), so we need an
adaptor. A suggestion is:
PLA-pin <-> EPROM-pin
10 <-> 19
20 <-> 10
22 <-> 1
14 <-> 20
14 <-> 22
pins of PLA to stay not connected: 1, 19
pins to be connected one-to-one:
2, 3, 4, 5, 6, 7, 8, 9, 11, 12, 13, 14,
15, 16, 17, 18, 21, 23, 24, 25, 26, 27, 28.
Remark that pin 14 of the PLA is connected to pins 14, 20,
and 22 of the EPROM!
With this adaptor, you should be able to read the PLA in an ordinary
EPROM reader/writer which can read 27512 types.
Note: Some EPROM readers can read PLAs, too, but that's not a good
choice since the numberings of the PLA and the EPROM do not stay
the same with the above suggested adaptor, so you would have to
write a program to map the appropriate pins. Reading the PLA with
the adapter would allow you to use the read information without
intervention.
Note2: It should also be possible to get the memory contents from
the abel listing on
ftp.funet.fi/pub/cbm/firmware/computers/c64/pla.txt,
but this would require a program which interprets the equations
and produces the output. It would even possible to get
I hope someone will check for errors!
Spiro.
PS: When you've done it successfully, please let me now!
PPS: I think you should use the fastest 27512 you can find; AFAIR,
the original PLA has 50ns access time which is relatively fast.
PPPS: Again, I HAVE NOT TRIED IT, but I think it should work.
Nevertheless, I'm not responsible for any damages this might
cause. If you don't feel fit enough to check my statements
for errors, don't do it anyway!
PPPPS: Oh yes, I'm not sure, but I think some newer C64s have other
PLAs, so this would probably not apply to you. Check if your
PLA is a 82S100 type (mine is, on a version A revision B board).
(note: On some C64, the PLA is marked with the CBM part no.,
not as 82S100).
Marko Mäkelä
Hello Spiro,
Spiro> Hi Ville (I hope that's your surname?),
I'm sorry to disappoint you, but it's his first name. :-)
Spiro> PPS: I think you should use the fastest 27512 you can find;
Spiro> AFAIR, the original PLA has 50ns access time which is
Spiro> relatively fast.
That's 80 nanoseconds, if I remember correctly. I haven't had much
experience with EPROMs lately, but a few (5-10) years ago, the access
times typically were between 120 and 150 nanoseconds.
Spiro> PPPPS: Oh yes, I'm not sure, but I think some newer C64s have
Spiro> other PLAs, so this would probably not apply to you.
Yes, the BN/E board has one "monster chip" that has the functionality
of the PLA and some logic glue. A more integrated version of this
chip also encompasses the colour memory.
Spiro> (note: On some C64, the PLA is marked with the CBM
Spiro> part no., not as 82S100).
Isn't that so on most C64s?
Marko
Spiro Trikaliotis
Marko Mäkelä wrote in message <7lhfhnn...@siphon.tcs.hut.fi>:
>>>>>> "Spiro" == Spiro Trikaliotis <Trik...@gmx.de> writes:
>Spiro> Hi Ville (I hope that's your surname?),
>I'm sorry to disappoint you, but it's his first name. :-)
Damned, yes, I meant the first name...
Sorry Ville, I didn't want to be inpolite...
>experience with EPROMs lately, but a few (5-10) years ago, the access
>times typically were between 120 and 150 nanoseconds.
Well, I believe I've seen (Standar-) EPROMs with 80 or 90 ns 'till
now, so it might work.
>Spiro> (note: On some C64, the PLA is marked with the CBM
>Spiro> part no., not as 82S100).
>Isn't that so on most C64s?
I don't know, I only see that mine states 82S100 (the part no. isn't
mentioned anywhere on the chip).
Spiro.
Ville Jouppi
<Trik...@gmx.de> wrote:
>Hi Marko,
>
>Marko Mäkelä wrote in message <7lhfhnn...@siphon.tcs.hut.fi>:
>
>>>>>>> "Spiro" == Spiro Trikaliotis <Trik...@gmx.de> writes:
>
>>Spiro> Hi Ville (I hope that's your surname?),
>
>>I'm sorry to disappoint you, but it's his first name. :-)
>
>Damned, yes, I meant the first name...
>Sorry Ville, I didn't want to be inpolite...
Heh, no problems.. :-)
>>experience with EPROMs lately, but a few (5-10) years ago, the access
>>times typically were between 120 and 150 nanoseconds.
>
>Well, I believe I've seen (Standar-) EPROMs with 80 or 90 ns 'till
>now, so it might work.
Hmm, those faster ones will be difficult to track down - eproms are
getting scarce these days.
>>Spiro> (note: On some C64, the PLA is marked with the CBM
>>Spiro> part no., not as 82S100).
>
>>Isn't that so on most C64s?
>
>I don't know, I only see that mine states 82S100 (the part no. isn't
>mentioned anywhere on the chip).
82S100 is the type of the chip, I've mostly seen those CBM chips in my
C-64s. (the 82S100s were self blown replacements for fried ones)
I'll check back on this when I have more time, and if I succeed, I'll
distribute the rom code and personality module wiring diagram.
Spiro Trikaliotis
Ville Jouppi wrote in message <9I1XOA7cDXNEgD...@4ax.com>...
>On Mon, 13 Dec 1999 17:15:57 +0100, "Spiro Trikaliotis"
><Trik...@gmx.de> wrote:
>>>experience with EPROMs lately, but a few (5-10) years ago, the access
>>>times typically were between 120 and 150 nanoseconds.
>>
>>Well, I believe I've seen (Standar-) EPROMs with 80 or 90 ns 'till
>>now, so it might work.
>
>Hmm, those faster ones will be difficult to track down - eproms are
>getting scarce these days.
Well, might be. I think I must have some lying around here, but they
are not very much. I believe there are 27256 and 27512 with 120 ns,
AFAIR.
>>>Spiro> (note: On some C64, the PLA is marked with the CBM
>>>Spiro> part no., not as 82S100).
>>
>>>Isn't that so on most C64s?
>>
>>I don't know, I only see that mine states 82S100 (the part no. isn't
>>mentioned anywhere on the chip).
>
>82S100 is the type of the chip, I've mostly seen those CBM chips in my
>C-64s. (the 82S100s were self blown replacements for fried ones)
Yes, I know that 82s100 is the type and the part no. is CBM specific.
I have my C=64 since December 1983, and the only replaced chip is one
CIA, so I'm sure it was factory shipped with a chip which reads 82s100
(and w/o CBM part no.).
>I'll check back on this when I have more time, and if I succeed, I'll
>distribute the rom code and personality module wiring diagram.
Yes, that would be nice.
Spiro.
Teemu Keskinarkaus
<Trik...@gmx.de> wrote:
>Hi Ville,
>
>Ville Jouppi wrote in message <9I1XOA7cDXNEgD...@4ax.com>...
>
>>On Mon, 13 Dec 1999 17:15:57 +0100, "Spiro Trikaliotis"
>><Trik...@gmx.de> wrote:
>
>
>>>>experience with EPROMs lately, but a few (5-10) years ago, the access
>>>>times typically were between 120 and 150 nanoseconds.
>>>
>>>Well, I believe I've seen (Standar-) EPROMs with 80 or 90 ns 'till
>>>now, so it might work.
>>
>>Hmm, those faster ones will be difficult to track down - eproms are
>>getting scarce these days.
>
>Well, might be. I think I must have some lying around here, but they
>are not very much. I believe there are 27256 and 27512 with 120 ns,
>AFAIR.
I've followed this conversation and I was just wondering why don't use
flashrom instead of EPROM?? Flashrom's are fast enought(I've seen 70ns
version and I think there could be even faster). It's easier to
program them because you don't need external eraser.
This is just a suggestion. :)
TK
Spiro Trikaliotis
Teemu Keskinarkaus wrote in message <385880db....@news.ramk.fi>:
>I've followed this conversation and I was just wondering why don't use
>flashrom instead of EPROM?? Flashrom's are fast enought(I've seen 70ns
>version and I think there could be even faster). It's easier to
>program them because you don't need external eraser.
>This is just a suggestion. :)
Well, IMHO it's a good suggestion, even because there are pin-to-pin
compatible types of Flash ROMs (BTW, EEPROMs would do it, too), and
to be honest, I never even thought about this.
But Flash ROMs or EEPROMs have one big disadvantage: It's easier
to find an EPROM write than a Flash or EEPROM writes, especially
for the C64, and they should also be cheaper. But if someone has
the ability to find a writer, he should be able to use it, too.
Thanks a lot for this suggestion!
Spiro.
Teemu Keskinarkaus
<Trik...@gmx.de> wrote:
>But Flash ROMs or EEPROMs have one big disadvantage: It's easier
>to find an EPROM write than a Flash or EEPROM writes, especially
>for the C64, and they should also be cheaper. But if someone has
>the ability to find a writer, he should be able to use it, too.
It's easy to do flasher if you know anything about electronics. It's
true that EPROM's are little cheaper but then you need eprom eraser so
are they anymore cheaper?? Of course if you already have eraser then
they are cheaper. And if you have good eprommer then you maybe(?) can
do software to use eprommer to program flashroms.
And if you use single voltage flashes you could do flasher for C64 if
you want. :) It's really easy to do. I have done it, not for C64 but
for Motorola 68HC11.
>Thanks a lot for this suggestion!
You're welcome.
TK
Spiro Trikaliotis
Teemu Keskinarkaus wrote in <3858ea29....@news.ramk.fi>:
>On Thu, 16 Dec 1999 10:18:46 +0100, "Spiro Trikaliotis"
><Trik...@gmx.de> wrote:
>It's easy to do flasher if you know anything about electronics. It's
>true that EPROM's are little cheaper but then you need eprom eraser so
>are they anymore cheaper?? Of course if you already have eraser then
>they are cheaper. And if you have good eprommer then you maybe(?) can
>do software to use eprommer to program flashroms.
Well, yes, it's not very complicated to build a flasher, esp. the
5 Volt types (like i.e. the ATMEL types), but it has to be done.
An EPROM writer can be found much more easily in my experience, and
you don't have to build something on your own.
Look, I don't want to tell you your idea were not good!
BTW, you don't really need an eraser, at least if you don't use EPROMs
very frequently, because you can use sun-light for erasing the EPROM.
>And if you use single voltage flashes you could do flasher for C64 if
>you want. :) It's really easy to do. I have done it, not for C64 but
>for Motorola 68HC11.
Well, I feel you want to show your schematics to me and the others...
Spiro.
PS: I don't really feel familiar with flashes; probably, because back
in my C=64 times, there weren't any.
Bman
>>>Spiro> (note: On some C64, the PLA is marked with the CBM
>>>Spiro> part no., not as 82S100).
>>
>>>Isn't that so on most C64s?
>>
>>I don't know, I only see that mine states 82S100 (the part no. isn't
>>mentioned anywhere on the chip).
>
>82S100 is the type of the chip, I've mostly seen those CBM chips in my
>C-64s. (the 82S100s were self blown replacements for fried ones)
Maybe I am missing something here, but are you sure you can replace
the PLA with a PROM of any sort at all? Maybe an EPLD, but I don't
think a PROM (includes E, EE and/or Flash types) will work.
I thought the PLA was a programmed logic array - and if so, this
suggests to me that it's not a chunk of ROM code, but rather a whole
lotta digital gates (AND, OR, NOT, XOR, etc) connected together to
perform logic functions... EPROMs can't simulate that... or am I
wrong?
Now, an EPLD would be the way to go if it cannot be done with a PROM
of some sort... but it might be more expensive, and harder to find a
programmer for it.
Does anyone have the inner workings of the PLA chip(s) available?
Byron
White Flame (aka David Holz)
news:3859610...@news.niagara.com...
> Maybe I am missing something here, but are you sure you can replace
> the PLA with a PROM of any sort at all? Maybe an EPLD, but I don't
> think a PROM (includes E, EE and/or Flash types) will work.
>
> I thought the PLA was a programmed logic array - and if so, this
> suggests to me that it's not a chunk of ROM code, but rather a whole
> lotta digital gates (AND, OR, NOT, XOR, etc) connected together to
> perform logic functions... EPROMs can't simulate that... or am I
> wrong?
Take every possible state of input lines and calculate the respective state of
the output lines. This basically makes a lookup table, which you can burn into
a ROM.
--
White Flame (aka David Holz)
http://fly.to/theflame
Spiro Trikaliotis
"Bman" (Byron) wrote in message <3859610...@news.niagara.com>:
>I thought the PLA was a programmed logic array - and if so, this
>suggests to me that it's not a chunk of ROM code, but rather a whole
>lotta digital gates (AND, OR, NOT, XOR, etc) connected together to
>perform logic functions... EPROMs can't simulate that... or am I
>wrong?
[...]
>Does anyone have the inner workings of the PLA chip(s) available?
Yes, you're wrong, unless there is a specific timing required, an
EPROM can fully replace a PLA.
A PLA works as follows:
- It takes its inputs (there are 16 in our case) and the negative
of its inputs, I call them I0..I15 and CI0..CI15 for the
complements.
- It generates minterms, i.e., combinations
I0 AND I1 AND I2 AND ... AND I15,
CI0 AND I1 AND I2 AND ... AND I15,
I0 AND CI1 AND I2 AND ... AND I15,
CI0 AND CI1 AND I2 AND ... AND I15,
...
CI0 AND CI1 AND CI2 AND ... AND CI15.
- These minterms can be ORed with others; which will be ORed
depends upon its programming. The results are its outputs,
in our case 8 outputs O0..O7.
A ROM is a device which can take any inputs (its address lines)
and generate outputs out of them (the data lines). It can be
freely programmed, and thus is more powerfull than a PLA.
So, why are PLAs used?
Because they don't need as much logic, they are cheaper and
often faster than ROMs. Look that we need a 64 KByte ROM
to replace the PLA inside the C64.
The timing is anuser issue. Depending on how critical the C64
is, the ROM might be too slow (it's my biggest concern), but
it also might be too fast - the C64 depends upon -CASRAM being
delayed after -CAS for the address multiplexers U13 and U25 to
get their jobs done in time!
Spiro.
Teemu Keskinarkaus
<Trik...@gmx.de> wrote:
>Well, yes, it's not very complicated to build a flasher, esp. the
>5 Volt types (like i.e. the ATMEL types), but it has to be done.
>An EPROM writer can be found much more easily in my experience, and
>you don't have to build something on your own.
True.
>Look, I don't want to tell you your idea were not good!
I know that. I just try to get you convinced about possibilies of
flashroms. :)
>BTW, you don't really need an eraser, at least if you don't use EPROMs
>very frequently, because you can use sun-light for erasing the EPROM.
This is true too. But this time of year sun is not shining here at
all! So you should send eprom's to somewhere where sun is shining to
get eproms erased. :)
>>And if you use single voltage flashes you could do flasher for C64 if
>>you want. :) It's really easy to do. I have done it, not for C64 but
>>for Motorola 68HC11.
>Well, I feel you want to show your schematics to me and the others...
It's part of bigger project. It is system where you can download
software from PC via RS-232 to 68HC11 based computer's flashrom and
then run it from there.
Basicly Flashes are simply to use. You write specified bytes to
specified address to get flash chip in write mode and then write one
byte at the time or one sector(256B or 512B) It depends what kind of
Flash you have. I have both kind of flashes. On some flashes you have
to erase chip(or sector) before write and some flashes don't need to
be erase before rewrite.
If you have eprommer it is possible to make it program flashes just
writing new software. This of course need that you can freely set
address in eprommer. Some eprommer have automatic address increment
logic.
>PS: I don't really feel familiar with flashes; probably, because back
> in my C=64 times, there weren't any.
It is relatively easy to get familiar with flashes. Download one
datasheet of some flash and check it out. You should get the idea
quite fast. Least I did. :)
TK
Jeroen Vlasveld
Spiro Trikaliotis wrote:
> Hi Teemu,
>
> Teemu Keskinarkaus wrote in <3858ea29....@news.ramk.fi>:
>
> >On Thu, 16 Dec 1999 10:18:46 +0100, "Spiro Trikaliotis"
> ><Trik...@gmx.de> wrote:
>
> >It's easy to do flasher if you know anything about electronics. It's
> >true that EPROM's are little cheaper but then you need eprom eraser so
> >are they anymore cheaper?? Of course if you already have eraser then
> >they are cheaper. And if you have good eprommer then you maybe(?) can
> >do software to use eprommer to program flashroms.
>
> Well, yes, it's not very complicated to build a flasher, esp. the
> 5 Volt types (like i.e. the ATMEL types), but it has to be done.
> An EPROM writer can be found much more easily in my experience, and
> you don't have to build something on your own.
>
> Look, I don't want to tell you your idea were not good!
>
> BTW, you don't really need an eraser, at least if you don't use EPROMs
> very frequently, because you can use sun-light for erasing the EPROM.
>
> >And if you use single voltage flashes you could do flasher for C64 if
> >you want. :) It's really easy to do. I have done it, not for C64 but
> >for Motorola 68HC11.
>
> Well, I feel you want to show your schematics to me and the others...
>
> Spiro.
>
> PS: I don't really feel familiar with flashes; probably, because back
> in my C=64 times, there weren't any.
Hi,
just wanted to say that there is a schematics for a flash device for the
6502 on the web.
have a look at:
http://www.6502.org/howto/flash/28f.htm
Jeroen.
Jo'ogn
Spiro Trikaliotis wrote:
> BTW, you don't really need an eraser, at least if you don't use EPROMs
> very frequently, because you can use sun-light for erasing the EPROM.
have you tested this !? i doubt that you can erase an EPROM with sun-light.
at least quick and properly
anyways i have missed the start of this threat and was wondering if eproms
are really too slow for the C64 !?
a CLK cycle has 1000ns - ok for delicate hardware-timing 120ns ARE a lot.
sounds if somebody has experience with (non working) eproms ?!
BTW - i reprogrammed the C't Atari ST prommer to write NVRAMs for me.
only tricky bit: don't set 12.5/21V "burn-voltage" to the RAM (;
regards
--
======================================================================
\v/ Anaconda Space Center NGC0417
° "Welcome to the Pleasuredome !"
======================================================================
"We apologize for the inconvenience" http://www.joogn.de
Pablo
> just wanted to say that there is a schematics for a flash device
> for the
> 6502 on the web.
> have a look at:
> http://www.6502.org/howto/flash/28f.htm
> Jeroen.
Hi, there is another project of this kind in MJK's page at:
http://www.student.informatik.tu-darmstadt.de/~mjk/c64.html
* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!
Jonas Bevren
On Thu, 16 Dec 1999, Bman wrote:
> Date: Thu, 16 Dec 1999 22:05:45 GMT
> From: Bman <bm...@niagara.com>
> Newsgroups: comp.sys.cbm
> Subject: Re: EPROM instead of PLA
>
>
> I thought the PLA was a programmed logic array - and if so, this
> suggests to me that it's not a chunk of ROM code, but rather a whole
> lotta digital gates (AND, OR, NOT, XOR, etc) connected together to
> perform logic functions... EPROMs can't simulate that... or am I
> wrong?
You're right.. a PLA _is_ a boatload of interconnected gates in one chip.
> Now, an EPLD would be the way to go if it cannot be done with a PROM
> of some sort... but it might be more expensive, and harder to find a
> programmer for it.
>
> Does anyone have the inner workings of the PLA chip(s) available?
Well, Byron, you dont fully understand hardware. The computer doesnt care
what's inside the device that sits in U17 on the c64. The contents are
irrelevant, so long as it responds like the original PLA. The PLA has a
limited number of possible input combinations, and a limited number of
outputs. I dont know how far back you've read this thread, so I'll do a
short recap.
The PLA has 16 inputs, and 8 outpus, as well as an output enable singal
(that's tied active permanently). Now, we could go through the list of
possibilities on the PLA's inputs, and record each possibilities' output
settings on a table. Then we have a list of 65,536 8-wire states. Clearly,
we could toss this into any sort of memory device that's fast enough to
respond reasonably to the c64's input. A lot of oder eproms are 150ns
devices.. sadly, they're too slow. Some more modern chips can move much
faster, and since I've seen mention of the 82s100n having a response time of
80ns, these chips should do.
One very important thing to remember: It's not what's inside that counts..
it's the way it acts.
-jb
Spiro Trikaliotis
Jo'ogn wrote in message <385A5D84...@tcpip-gmbh.de>:
>Spiro Trikaliotis wrote:
>> BTW, you don't really need an eraser, at least if you don't use EPROMs
>> very frequently, because you can use sun-light for erasing the EPROM.
>have you tested this !? i doubt that you can erase an EPROM with sun-light.
>at least quick and properly
Yes, it works, but you need to place it for _very_ long into the sun.
But if you have enough EPROMs handy, it is no problem.
>anyways i have missed the start of this threat and was wondering if eproms
>are really too slow for the C64 !?
Well, you remember that we want to replace the PLA which decodes
the CS-lines with an EPROM. It has _possibly_ delicate timing requirements,
although I don't really know, so I thought the best way were to use an
EPROM (or EEPROM or FLASH or ...) with an access time as near to the original
PLA as possible.
Especially the timing with external cartridges that try to replace the KERNAL
or other parts seems to be very critical, because that type of card has to
change the PLA's configuration after it has given some output - as it seems to
me, the PLA's reaction to this change must occur very fast, so there might be
problems with a slow PLA substitution.
Spiro.
PS: Thanks to all references for FLASH programmers.
Leinonen Mika
>>flashrom instead of EPROM?? Flashrom's are fast enought(I've seen 70ns
>But Flash ROMs or EEPROMs have one big disadvantage: It's easier
>to find an EPROM write than a Flash or EEPROM writes, especially
>for the C64, and they should also be cheaper. But if someone has
>the ability to find a writer, he should be able to use it, too.
There's an AT 29C256 28-pin 32 KB Flash programmer at
http://www.students.tut.fi/%7El150540/c64fprog.html
But there might not be fast enough 29c256's to be used
as a replacement for the 82S100 PLA.
www.epanorama.net has many links to DIY programmers.
--
mika.leinonen"at"cc.tut.fi
http://www.students.tut.fi/%7El150540/
Ed
that glitches don't occur on the outputs as the inputs change
state - something that eproms will apparently do. Depending
on the application this may cause unforseen side-effects.
Ed
Leinonen Mika <l15...@lehtori.cc.tut.fi> wrote in message
news:84dhnt$ekl$1...@baker.cc.tut.fi...
Mike Gregory
Here's a definitive answer, apologies for the length.
I tried it with a 27C010A-10 which was the fastest eprom I had to
hand.
It started the normal C=64 blue screen OK. I was able to load and
run a program from disk. One program had an intro which had music plus
a mid-screen scroller (not through the border). Both the music and
scrolling seemed OK. A cartridge worked fine.
The only noticeable glitch was that the character at row 1 / col 1 (ie
the first screen position) was not light blue. It was predominantly
black or yellow. It also flickered when a key was pressed.
I assumed that this was a timing problem and was indicating that the
eprom was borderline fast enough.
I then tried a 27C010-150 (same size, slower, different brand).
This worked as described above and the char at position 1 was OK????
Maybe the problem with the 100ns eprom will disappear if I try another
eprom.
I can leave the setup as is for a week or so, so if anyone can think
of any useful tests that might be run.........
Mike G
On Mon, 13 Dec 1999 13:57:09 +0100, "Spiro Trikaliotis"
<Trik...@gmx.de> wrote:
>Hi Ville (I hope that's your surname?),
>
> (note: On some C64, the PLA is marked with the CBM part no.,
> not as 82S100).
>
Spiro Trikaliotis
Mike Gregory schrieb in Nachricht <387ed64...@news.syd.ihug.com.au>...
>I tried it with a 27C010A-10 which was the fastest eprom I had to
>hand.
[...]
>the first screen position) was not light blue. It was predominantly
>black or yellow. It also flickered when a key was pressed.
>I assumed that this was a timing problem and was indicating that the
>eprom was borderline fast enough.
Well, I think this might be a timing problem, too. On the other hand,
are you sure the EPROM is correct written? Possibly, there's an error
in it. Have you tried to re-burn it?
>I then tried a 27C010-150 (same size, slower, different brand).
>This worked as described above and the char at position 1 was OK????
Hm. I think of the statement of "Ed" <ed...@alphalink.com.au> from
today. Probably, the fastest EPROM is not the best since it will
try to respond to every change of the inputs. A slower EPROM might
not respond so quickly, e.g., might not respond at all to some wrong
inputs (I'm sorry, I don't know how it's called in english; I mean
the effect a ripple counter has), so it might be better.
But I'm not sure at all (and the only data sheets for EPROMs is an
old data book by Toshiba from 1982, so the timings might not be
very accurate.
But I'll propose you should try to re-burn the 100 ns EPROM, and if
the error remains, try another one (I hope you have one handy).
Spiro.
BTW: I had another look into the data book from 1982. I can see there
a TMM2364P mask-programmable ROM which is 28 pins, not 24 pins as is in
the C64, and is compatible to a 2764. The 24 pin 8k x 8 EPROM is called
TMM2366P by Toshiba. I'm surprised since I always thought at least the
EPROM numberings were the same along the different manufacturers.
If only CBM had used a 28 pin 2364, things would have been much easier.
Mike Gregory
I have now tried a second 100ns eprom (same brand) and get the same
glitch of a color change and a flickering char in the first screen
position. In case you ask, I used the 150ns eprom to burn the 100ns
eprom.
Something strange happens. If I POKE55297,14 then the char color at
55296 goes to light blue and the flickering stops. The drawback is
that I am then unable to change the color at 55296????? That is
POKE55296,x does not change anything.
Mike G
Mike Gregory
<Trik...@gmx.de> wrote:
>Well, I think this might be a timing problem, too. On the other hand,
>are you sure the EPROM is correct written? Possibly, there's an error
>in it. Have you tried to re-burn it?
I used the two adaptors to verify the eprom against the original PLA
data. That is, I converted the eprom to a PLA and the converted that
back to an eprom, so I am sure that the burning was OK. I always read
a freshly burnt eprom and compare it against the original file also.
At this stage I believe that the eprom may just be too "noisy" if
that's possible. I will try it later today using another 100ns eprom.
>If only CBM had used a 28 pin 2364, things would have been much easier.
The CBM roms are of course pin compatible with the 68764 series
eproms. However these are much more expensive than the 27 series and
less common. In this case though the PLA is different again, so it
would not have mattered what CBM had used for the roms.
Mike G
David Evans
Mike Gregory <mike...@tig.com.au> wrote:
>
>At this stage I believe that the eprom may just be too "noisy" if
>that's possible. I will try it later today using another 100ns eprom.
>
Did you try sticking some filter caps on the output lines? I've been bitten
by transition spikes with EPROMs before. After all they usually sit on busses
where the data lines are sampled a reasonable time after the address is set
up.
--
David Evans (NeXTMail/MIME OK) dfe...@bbcr.uwaterloo.ca
Computer/Synth Junkie http://bbcr.uwaterloo.ca/~dfevans/
University of Waterloo "Default is the value selected by the composer
Ontario, Canada overridden by your command." - Roland TR-707 Manual