Computer made with a zilog chip

On 7/30/2016 5:54 AM, Vassilis Spiliopoulos wrote:
>
> I am going to make a computer or at least attempt to using some
>  CPU chips and breadboards. The initial idea is to link each
>  output pin such as halt to a led light.And the address buses too
>  and link the inputs to switches.It will be a very basic CPU unit.
>  What are your thoughts on this?

This is not really the right group for that question.  You might try 
comp.arch.embedded for a better response and being on topic with the group.

-- 

Rick C

On Saturday, July 30, 2016 at 2:54:44 AM UTC-7, Vassilis Spiliopoulos wrote:
> I am going to make a computer or at least attempt to using some
>  CPU chips and breadboards. The initial idea is to link each
>  output pin such as halt to a led light.And the address buses too
>  and link the inputs to switches.It will be a very basic CPU unit.
>  What are your thoughts on this? 


I did one a few years ago, following the logic, and ROM data, of the SDK-85

I bought some strange board through eBay with an 80C85, RAM, and some other chips.

Put the 80C85, 74LS373 latch (for the address bus), 27xxx EPROM, RAM, 1488, and 1489 on a breadboard.

The 80C85 has a built-in bit bang serial port, which the SDK-85 ROM knows about.

After I got tired of 110 baud, I changed the timing loop for 9600.   

You do need to be able to program EPROMs though, or, I suppose Flash memory.

The hard part was figuring out which CPU lines to ground or Vcc to get it to run.

herrmannsfeldt@gmail.com writes:
> [...]
> You do need to be able to program EPROMs though, or, I suppose Flash
> memory.

How I miss the days of the 2732 UVEPROM and my Data I/O programmer (not
to mention 8085 assembly language)! That's when men were men and
programmers were programmers!
-- 
Randy Yates, DSP/Embedded Firmware Developer
Digital Signal Labs
http://www.digitalsignallabs.com

On 8/2/2016 9:24 AM, Randy Yates wrote:
> herrmannsfeldt@gmail.com writes:
>> [...]
>> You do need to be able to program EPROMs though, or, I suppose Flash
>> memory.
>
> How I miss the days of the 2732 UVEPROM and my Data I/O programmer (not
> to mention 8085 assembly language)! That's when men were men and
> programmers were programmers!

Not old enough to have used 1702A parts, eh?

-- 

Rick C

rickman <gnuarm@gmail.com> writes:

> On 8/2/2016 9:24 AM, Randy Yates wrote:
>> herrmannsfeldt@gmail.com writes:
>>> [...]
>>> You do need to be able to program EPROMs though, or, I suppose Flash
>>> memory.
>>
>> How I miss the days of the 2732 UVEPROM and my Data I/O programmer (not
>> to mention 8085 assembly language)! That's when men were men and
>> programmers were programmers!
>
> Not old enough to have used 1702A parts, eh?

Rick,

I guess not, I don't remember that one! I think prior to the 2732 there
were the 2708 and 2716, right? I remember being amazed at how fast
memory sizes were increasing.
-- 
Randy Yates, DSP/Embedded Firmware Developer
Digital Signal Labs
http://www.digitalsignallabs.com

On 02/08/16 14:48, rickman wrote:
> On 8/2/2016 9:24 AM, Randy Yates wrote:
>> herrmannsfeldt@gmail.com writes:
>>> [...]
>>> You do need to be able to program EPROMs though, or, I suppose Flash
>>> memory.
>>
>> How I miss the days of the 2732 UVEPROM and my Data I/O programmer (not
>> to mention 8085 assembly language)! That's when men were men and
>> programmers were programmers!
>
> Not old enough to have used 1702A parts, eh?

Some things are best left in the dim recesses of memory,
until you want to show kids what a "chip" looks like.

Randy Yates  <yates@digitalsignallabs.com> wrote:

>How I miss the days of the 2732 UVEPROM and my Data I/O programmer (not
>to mention 8085 assembly language)! That's when men were men and
>programmers were programmers!

I recall having a UV lamp that would reliably erase the ROM in about
30 minutes.  But for testing a quick bug fix, it might be expedient to
give it just 15 or so minutes -- if it works, great, if not put the ROM
back in the eraser while looking for more bugs in the code.

Steve

On Tuesday, August 2, 2016 at 6:48:28 AM UTC-7, rickman wrote:

(snip)

> Not old enough to have used 1702A parts, eh?

I remember them, but never used them.

Both 1702A and 2708 used unusual voltages and timing on the data lines.

Starting with the 2716, programming is much simpler. 

On 8/2/2016 11:25 AM, Randy Yates wrote:
> rickman <gnuarm@gmail.com> writes:
>
>> On 8/2/2016 9:24 AM, Randy Yates wrote:
>>> herrmannsfeldt@gmail.com writes:
>>>> [...]
>>>> You do need to be able to program EPROMs though, or, I suppose Flash
>>>> memory.
>>>
>>> How I miss the days of the 2732 UVEPROM and my Data I/O programmer (not
>>> to mention 8085 assembly language)! That's when men were men and
>>> programmers were programmers!
>>
>> Not old enough to have used 1702A parts, eh?
>
> Rick,
>
> I guess not, I don't remember that one! I think prior to the 2732 there
> were the 2708 and 2716, right? I remember being amazed at how fast
> memory sizes were increasing.

I got into electronics design at the time 1702A (256 bytes) were the 
only game in town, but were soon replaced by the 2708, then the 2716 and 
then 2732.  I recall I was very impressed with how fast that happened... 
Moore's law in action.

The 1702A was worse than just small.  It required insane high voltage 
pulses on the address and/or data lines which had to be reversed to do 
the programming.  I think the other EPROMs were amazingly simple to 
program in comparison.

-- 

Rick C

On Tuesday, August 2, 2016 at 10:56:43 PM UTC-7, rickman wrote:
> On 8/2/2016 11:25 AM, Randy Yates wrote:

(snip)

> > I guess not, I don't remember that one! I think prior to the 2732 there
> > were the 2708 and 2716, right? I remember being amazed at how fast
> > memory sizes were increasing.
 
> I got into electronics design at the time 1702A (256 bytes) were the 
> only game in town, but were soon replaced by the 2708, then the 2716 and 
> then 2732.  I recall I was very impressed with how fast that happened... 
> Moore's law in action.
 
> The 1702A was worse than just small.  It required insane high voltage 
> pulses on the address and/or data lines which had to be reversed to do 
> the programming.  I think the other EPROMs were amazingly simple to 
> program in comparison.

I never tried to design or build a 1702A or 2708 programmer, but yes,
they are high voltages, but even more, the timing had to be right.

They are fast pulses, overshoot and you might destroy the device.

The 1702A is PMOS, the 2708 NMOS, which affects the polarity.

For the 2716, you supply 25V or 21V, and pulse a TTL level signal.
Much easier. 

I built a programmer from a BYTE article, which has a serial port
with UART, a few latches, and a 555 to generate the timing pulse.