Project with "IF" function and timer

On Mon, 15 Jun 2015, rickman wrote:

> On 6/15/2015 1:13 PM, Ed Prochak wrote:
>> On Sunday, June 14, 2015 at 5:50:34 PM UTC-4, rickman wrote:
>> []
>>> 
>>> I don't think he is trying to learn really.  I expect this is an
>>> assignment and he is trying to get something to work as quickly as
>>> possible.  It's not a difficult assignment really, but he needs to
>>> *learn* about the issues involved rather than just copying code from the
>>> Internet.

>> yes, he seems to be doing on the job learning, poorly. That seems to be the 
>> modern mentality:  "I don't need to understand how it works. I'll just 
>> throw code at it until something happens."
>> 
>> I hope he follows your advice.
>
> I remember being in a hurry myself when I was in school.  I thought I was 
> learning everything I needed and the actual practice of the craft was just an 
> "exercise".  lol.  I'm still learning the practice of the craft.

Yes, there are still a few of us old timers out here who not only
programmed "right down on the iron" in assembly language, we may have
set bootstrap loaders with toggle switches(!!!) and even did a little
programming in straight octal/hexadecimal, not even assembly language.
It seems that these days to too many young whippersnappers, CPUs are
just black boxes about which they have little understanding how they
really work under the covers. I wonder even about embedded systems in
which people code in C? Do they really understand what is going on in
the hardware layer? Do they need to?

-- 
Paul Bartlett

On Tue, 16 Jun 2015 13:59:56 -0400, Paul Bartlett <bartlett@panix.com>
wrote:

>On Mon, 15 Jun 2015, rickman wrote:
>
>> On 6/15/2015 1:13 PM, Ed Prochak wrote:
>>> On Sunday, June 14, 2015 at 5:50:34 PM UTC-4, rickman wrote:
>>> []
>>>> 
>>>> I don't think he is trying to learn really.  I expect this is an
>>>> assignment and he is trying to get something to work as quickly as
>>>> possible.  It's not a difficult assignment really, but he needs to
>>>> *learn* about the issues involved rather than just copying code from the
>>>> Internet.
>
>>> yes, he seems to be doing on the job learning, poorly. That seems to be the 
>>> modern mentality:  "I don't need to understand how it works. I'll just 
>>> throw code at it until something happens."
>>> 
>>> I hope he follows your advice.
>>
>> I remember being in a hurry myself when I was in school.  I thought I was 
>> learning everything I needed and the actual practice of the craft was just an 
>> "exercise".  lol.  I'm still learning the practice of the craft.
>
>Yes, there are still a few of us old timers out here who not only
>programmed "right down on the iron" in assembly language, we may have
>set bootstrap loaders with toggle switches(!!!) and even did a little
>programming in straight octal/hexadecimal, not even assembly language.

Some architectures like PDP-11 and 8080 are naturally octal (three bit
groups) while others, such as IBM360/370 as well as VAX-11 series are
essentially hexadecimal (4 bit groups). 

Doing front panel programming with toggle switches was not a big deal.
On some architectures calculating the branch relative offset required
pen and paper, otherwise programming with toggle switches or 8/16
position rotary swathes was quite easy.

I still do not understand why intel swathes from octal ( 2 bit opcode
and 3 bit destination address and  3 bit source address) to
hexadecimal on 8080 and 8085.
  

On 2015-06-16, upsidedown@downunder.com <upsidedown@downunder.com> wrote:

> I still do not understand why intel swathes from octal ( 2 bit opcode
> and 3 bit destination address and  3 bit source address) to
> hexadecimal on 8080 and 8085.

IIRC, Heathkit's assembler used octal instead of hex by default for
8080/8085/Z80.  Probably due to the previous work with PDP-11 kits.

-- 
Grant Edwards               grant.b.edwards        Yow! And then we could sit
                                  at               on the hoods of cars at
                              gmail.com            stop lights!

upsidedown@downunder.com wrote:

(snip)

> Some architectures like PDP-11 and 8080 are naturally octal (three bit
> groups) while others, such as IBM360/370 as well as VAX-11 series are
> essentially hexadecimal (4 bit groups). 
 
> Doing front panel programming with toggle switches was not a big deal.
> On some architectures calculating the branch relative offset required
> pen and paper, otherwise programming with toggle switches or 8/16
> position rotary swathes was quite easy.

The Altair 8800 front panel has the LEDs grouped in threes.

More than one I have known has sticky labels to divide them into
groups of four, instead.  Even if it is more natural as octal,
if you are more used to hex, it is easier that way.

Also, the PDP-11 with 16 its, and the 8080 with eight, don't map
to whole numbers of octal digits. That complicates working with
them in octal.
 
> I still do not understand why intel swathes from octal ( 2 bit opcode
> and 3 bit destination address and  3 bit source address) to
> hexadecimal on 8080 and 8085.

-- glen
-- glen

On 6/16/2015 1:59 PM, Paul Bartlett wrote:
> On Mon, 15 Jun 2015, rickman wrote:
>
>> On 6/15/2015 1:13 PM, Ed Prochak wrote:
>>> On Sunday, June 14, 2015 at 5:50:34 PM UTC-4, rickman wrote:
>>> []
>>>>
>>>> I don't think he is trying to learn really.  I expect this is an
>>>> assignment and he is trying to get something to work as quickly as
>>>> possible.  It's not a difficult assignment really, but he needs to
>>>> *learn* about the issues involved rather than just copying code from
>>>> the
>>>> Internet.
>
>>> yes, he seems to be doing on the job learning, poorly. That seems to
>>> be the modern mentality:  "I don't need to understand how it works.
>>> I'll just throw code at it until something happens."
>>>
>>> I hope he follows your advice.
>>
>> I remember being in a hurry myself when I was in school.  I thought I
>> was learning everything I needed and the actual practice of the craft
>> was just an "exercise".  lol.  I'm still learning the practice of the
>> craft.
>
> Yes, there are still a few of us old timers out here who not only
> programmed "right down on the iron" in assembly language, we may have
> set bootstrap loaders with toggle switches(!!!) and even did a little
> programming in straight octal/hexadecimal, not even assembly language.
> It seems that these days to too many young whippersnappers, CPUs are
> just black boxes about which they have little understanding how they
> really work under the covers. I wonder even about embedded systems in
> which people code in C? Do they really understand what is going on in
> the hardware layer? Do they need to?

How about microcode?  Anyone else done that?

-- 

Rick

On 6/16/2015 4:59 PM, glen herrmannsfeldt wrote:
> upsidedown@downunder.com wrote:
>
> (snip)
>
>> Some architectures like PDP-11 and 8080 are naturally octal (three bit
>> groups) while others, such as IBM360/370 as well as VAX-11 series are
>> essentially hexadecimal (4 bit groups).
>
>> Doing front panel programming with toggle switches was not a big deal.
>> On some architectures calculating the branch relative offset required
>> pen and paper, otherwise programming with toggle switches or 8/16
>> position rotary swathes was quite easy.
>
> The Altair 8800 front panel has the LEDs grouped in threes.
>
> More than one I have known has sticky labels to divide them into
> groups of four, instead.  Even if it is more natural as octal,
> if you are more used to hex, it is easier that way.
>
> Also, the PDP-11 with 16 its, and the 8080 with eight, don't map
> to whole numbers of octal digits. That complicates working with
> them in octal.

It does?  I have seen many a listing in octal and it works wonderfully 
with the LSI-11.  Who cares about the left digit being just one bit?

-- 

Rick

On 2015-06-16, rickman <gnuarm@gmail.com> wrote:
> On 6/16/2015 4:59 PM, glen herrmannsfeldt wrote:
>
>> Also, the PDP-11 with 16 its, and the 8080 with eight, don't map to
>> whole numbers of octal digits. That complicates working with them in
>> octal.
>
> It does?

Not that I recall.

> I have seen many a listing in octal and it works wonderfully with the
> LSI-11.  Who cares about the left digit being just one bit?

I remember seeing Z80 listings that used split octal notation for
addresses and other 16-bit values: 000_000 to 377_377.  I thought that
a bit odd, but never worked with that notation enough to comment on
it's utility.

-- 
Grant Edwards               grant.b.edwards        Yow! I left my WALLET in
                                  at               the BATHROOM!!
                              gmail.com            

On 6/16/2015 5:57 PM, Grant Edwards wrote:
> On 2015-06-16, rickman <gnuarm@gmail.com> wrote:
>> On 6/16/2015 4:59 PM, glen herrmannsfeldt wrote:
>>
>>> Also, the PDP-11 with 16 its, and the 8080 with eight, don't map to
>>> whole numbers of octal digits. That complicates working with them in
>>> octal.
>>
>> It does?
>
> Not that I recall.
>
>> I have seen many a listing in octal and it works wonderfully with the
>> LSI-11.  Who cares about the left digit being just one bit?
>
> I remember seeing Z80 listings that used split octal notation for
> addresses and other 16-bit values: 000_000 to 377_377.  I thought that
> a bit odd, but never worked with that notation enough to comment on
> it's utility.

If you've never used a flat bladed screw driver I expect you wouldn't be 
very good with that either.

-- 

Rick

On Tue, 16 Jun 2015 20:59:51 +0000 (UTC), glen herrmannsfeldt
<gah@ugcs.caltech.edu> wrote:

>upsidedown@downunder.com wrote:
>
>(snip)
>
>> Some architectures like PDP-11 and 8080 are naturally octal (three bit
>> groups) while others, such as IBM360/370 as well as VAX-11 series are
>> essentially hexadecimal (4 bit groups). 
> 
>> Doing front panel programming with toggle switches was not a big deal.
>> On some architectures calculating the branch relative offset required
>> pen and paper, otherwise programming with toggle switches or 8/16
>> position rotary swathes was quite easy.
>
>The Altair 8800 front panel has the LEDs grouped in threes.
>
>More than one I have known has sticky labels to divide them into
>groups of four, instead.  Even if it is more natural as octal,
>if you are more used to hex, it is easier that way.
>
>Also, the PDP-11 with 16 its, and the 8080 with eight, don't map
>to whole numbers of octal digits. That complicates working with
>them in octal.

If you just enter a program from a listing (generated by assembler or
manually assembled) with toggle switches, it is quite irrelevant if
octal or hex is used. Of course, if you have a hex keypad, using hex
will save a lot of keystrokes.

My original point was about writing a program into memory through
front panel toggle switches without any paper listing. In this case it
is essential that the mode (octal/hex) matches the machine
architecture. Using hex on a PDP-11/8080/Z80 requires remember the
whole 8 or 16 bit opcode. For PDP-11 dual operand instructions, you
only needed to remember a 3 bit opcode, then use the byte/word bit
(most significant bit) the addressing mode (3 bits), (base)register (3
bits) for the first operand and 3+3 bits for the other operand.
 

On Tue, 16 Jun 2015 13:59:56 -0400, Paul Bartlett <bartlett@panix.com>
wrote:

>Yes, there are still a few of us old timers out here who not only
>programmed "right down on the iron" in assembly language, we may have
>set bootstrap loaders with toggle switches(!!!) and even did a little

Sometimes also called key-in loader -)