🚀 go-pugleaf

Gareth's Downstairs Computer
<headstone255.but.n...@yahoo.com> wrote:
> On 16/02/2018 23:04, Charlie Gibbs wrote:
>> On 2018-02-16, Gareth's Downstairs Computer
>> <headstone255.but.n...@yahoo.com> wrote:
>> 
>>> On 16/02/2018 16:36, Charlie Gibbs wrote:
>>> 
>>>> On 2018-02-16, Huge <Huge@nowhere.much.invalid> wrote:
>>>> 
>>>>> On 2018-02-15, Peter Flass <peter...@yahoo.com> wrote:
>>>>> 
>>>>> [22 lines snipped]
>>>>> 
>>>>>> The nostalgia is for the sights, sounds, and smells those days. I can still
>>>>>> recall leaning over the 1132 printer waiting for my printout to start.
>>>>>> We're also nostalgic for out first acquaintance with the wonders of
>>>>>> computers. Now every kid over 8 knows more than we did 50 years ago.
>>>>> 
>>>>> I consider that unlikely.
>>>> 
>>>> Agreed.  I can't accept that the creators of the Internet don't know
>>>> as much as the script kiddie who abuses it.
>>>> 
>>>> Too many doting parents think their kid is a computer genius
>>>> simply because he knows how to click a mouse.
>>> 
>>> I once contrated in a company where the meme that kids were
>>> ace on computers was held. The son of one of the directors had
>>> produced their first computerised product, all in machine code,
>>> and all typed in manually into a "Softy" PROM programmer.
>> 
>> Well, don't keep us in suspense.  Did it work?  Did it work well?
>> 
> 
> Yes, it did work and sold well. Upgrading though was such a pain that
> they recoded in assembler *WITHOUT ANY COMMENTS* because, based on their
> experience with hexadecimal machine code, assembler was "self documenting"!

I think most of us started out writing impenetrable, commentless code. When
you're doing a student project it's A) small, B) completed quickly enough
that you usually don't forget how it's working C) keypunched/typed by you
D) run once and thrown away.  In my experience that's the biggest step from
student programming assignments to production work-the code may be complex
and had to be understood by lots of people, perhaps years later.

> 
> I was contracted to work on a range of PID controllers for fermentation
> vessels (Temperature, pH, Dissolved Oxygen, etc) and when I said that
> they should let it be done in C because of all the maths involved they
> declined, so I spent much time developing 32 bit arithmetic in 8 bit
> assembler, which took longer to do (and was thus more expensive) than
> had C been permitted.

So use assembler, but call routines from the C library ;-)

> 
> The division routine was faultless, but took several days to debug
> because I was doing easy to follow divisions such as 300,000,000 (hex)
> divided by 3 and using an early hex-capable pocket calculator with
> only 7 segment displays, was misreading the B in a hex answer as a 6.
> 
> Live and learn!

Amen! Still learning.




-- 
Pete