23 Jan
2008
23 Jan
'08
5:31 p.m.
Date: Tue, 22 Jan 2008 19:33:15 -0800
From: dwight elvey <dkelvey at hotmail.com>
Here goes with using your idea to double use the a
register. Except
If I start it with 2, I can use the carry instead of the sign bit. What
fun!
Div10:
ld a,#2
ld de,#-640d
divloop:
add hl,de
jr c, div1
sbc hl,de
div1:
rl a
jr nc,divloop
add hl,hl
ret
Ah, but you forgot the add hl,hl to shift the dividend after each
iteration. Put it at the top of the loop so it doesn't interfere
with the operation of the "rl a/jr nc" pair and scale your divisor by
an additional power of 2 and you should have it.
By now, we've probably bored the hardware guys to tears and DEC
people are tearing out their hair, so we should probably drop the
subject.
My apologies, but I've always found bit-twiddling fun. It seems that
every platform has some peculiar characteristic that can be exploited
to good advantage.
Cheers,
Chuck
23 Jan
23 Jan
5:55 p.m.
On 23/01/2008 17:31, Chuck Guzis wrote:
By now, we've probably bored the hardware guys to
tears and DEC
people are tearing out their hair, so we should probably drop the
subject.
Well, I collect DEC kit and build/mend hardware, and I found it
interesting. But I learned on Z80s :-)
My apologies, but I've always found bit-twiddling
fun. It seems that
every platform has some peculiar characteristic that can be exploited
to good advantage.
Me too -- I've just been attempting that with PIC code for the first time.
--
Pete Peter Turnbull
Network Manager
University of York
6:58 p.m.
New subject: div by 10 on Z80
Chuck Guzis wrote:
By now, we've probably bored the hardware guys to tears and DEC
people are tearing out their hair, so we should probably drop the
subject.
My apologies, but I've always found bit-twiddling fun. It seems that
every platform has some peculiar characteristic that can be exploited
to good advantage.
I'm still following this and watching the improvements. (I'm still
looking at what docs I have here to ascertain exactly which way the Z80 vs 6800
borrow works, but some of it's ambiguous (just for my own sake, I still do some
assembler for both of them on occasion)).
Of course, from a clarity-is-all-important modern software design perspective,
we should all be shot.
If I were an economist, I would predict that at this rate, in another few weeks
the routine will be down to 1 byte.
7:39 p.m.
New subject: div by 10 on Z80
Brent Hilpert wrote:
Of course, from a clarity-is-all-important modern software design perspective,
we should all be shot.
If I were an economist, I would predict that at this rate, in another few weeks
the routine will be down to 1 byte.
It could real soon ... We have yet to study the
*secret Z80* instructions.
BTW was this for disk sector calculations again? What OS?
24 Jan
24 Jan
5:30 a.m.
New subject: div by 10 on Z80
Date: Wed, 23 Jan 2008 12:39:58 -0700
From: bfranchuk at jetnet.ab.ca
To:
Subject: Re: div by 10 on Z80
Brent Hilpert wrote:
Hi
It is for System88 that was used by Polymorphic. It is 10 hard sectored
dsdd disk. I've just figured out the headers. Something that is only practical
with a scope that has delayed sweep ( which I had to fix for this project ).
The code for the disk controller runs on the controller board and uses
a Z80 processor ( the main processor is a 8080 running on the S100 ).
There are two windows in the Z80 memory that are double accessed
by both the S100 side and the Z80 side. One window at 1000H for
buffering while the other is only 16 bytes used for command, parameters
and status.
The code that I have is for 256 byte sectors on 32 hard sectored
8 inch disk.
I'm trying to modify the code to read the 512 byte sectors of the
5.25 disk. I've already modified the PLL, crystal and a oneshot for timing.
There is a slight difference in the headers, being that the 8 inch
does track, track\, sector, sector\ and the 5.25 does sector, sector\
track, track\.
Since the controller runs on its own EPROM, I need to add code to the
EPROM to deal with things like dividing by 10. I have some space
to add this but I also have to make some other changes to deal
with things like the larger sector size. That is why I'm interested in
size. I have about 40 bytes left.
What is funny is that I beleive that one of the disk I'm trying to
read has the source and binary for the complete ROM code for the 5.25.
A chicken and egg problem now.
The disk are from the Polymorphic archive and have a lot of valuable
source, manuals and binaries. As I stated earlier, there is even source
for their version of BASIC ( not MS ).
Dwight
_________________________________________________________________
Need to know the score, the latest news, or you need your Hotmail?-get your
"fix".
http://www.msnmobilefix.com/Default.aspx
Of course, from a clarity-is-all-important modern software design perspective,
we should all be shot.
If I were an economist, I would predict that at this rate, in another few weeks
the routine will be down to 1 byte.
It could real soon ... We have yet to study
the *secret Z80* instructions.
BTW was this for disk sector calculations again? What OS?
6:09 a.m.
New subject: div by 10 on Z80
On Thursday 24 January 2008 00:30, dwight elvey wrote:
There are two windows in the Z80 memory that are
double accessed
by both the S100 side and the Z80 side. One window at 1000H for
buffering while the other is only 16 bytes used for command, parameters
and status.
I'd be real interested in more details about how they do that.
(Snip)
The disk are from the Polymorphic archive and have a
lot of valuable
source, manuals and binaries. As I stated earlier, there is even source
for their version of BASIC ( not MS ).
Not a big fan of BASIC, but a non-m$ one would be nice, I suppose.
Speaking of which, I remember some Kaypro stuff coming with more than one
kind, but I don't recall ever running across any docs for those -- anybody
know of any?
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, ?a critter that can
be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
6:45 a.m.
New subject: div by 10 on Z80
On Thu, Jan 24, 2008 at 01:09:31AM -0500, Roy J. Tellason wrote:
(Snip)
Is there a TinyBASIC for 8088? I know Tom Pittman's is/was available
for 6502 and 1802 (and still is, directly from his website, IIRC).
Unless you are doing work on a small memory machine (minimally equipped
1802 or 6502 or INS8073), I don't know that TinyBASIC is an improvement
over M$ BASIC, but on a machine with 2K-4K, it's really useful.
-ethan
--
Ethan Dicks, A-333-S Current South Pole Weather at 24-Jan-2008 at 06:40 Z
South Pole Station
PSC 468 Box 400 Temp -29.7 F (-34.3 C) Windchill -47.3 F (-44.0 C)
APO AP 96598 Wind 5.2 kts Grid 70 Barometer 686.2 mb (10395 ft)
Ethan.Dicks at usap.gov http://penguincentral.com/penguincentral.html
The disk are from the Polymorphic archive and
have a lot of valuable
source, manuals and binaries. As I stated earlier, there is even source
for their version of BASIC ( not MS ).
Not a big fan of BASIC, but a non-m$ one would be nice, I suppose.
4:40 p.m.
New subject: div by 10 on Z80
Date: Thu, 24 Jan 2008 06:45:05 +0000
From: ethan.dicks at usap.gov
To: cctalk at classiccmp.org
Subject: Re: div by 10 on Z80
On Thu, Jan 24, 2008 at 01:09:31AM -0500, Roy J. Tellason wrote:
Hi Ethan
The version of TB that was on my Poly88 has the ability to have
extended operations. I used it to add PEEK, POKE and SAVE.
With this, one could extend to just about anything.
The BASIC on the disk is a full function BASIC. There was a 8K
and a 12K version. I was told that they were considered as
much better than MS. They were used a lot for business applications.
Dwight
_________________________________________________________________
Need to know the score, the latest news, or you need your Hotmail?-get your
"fix".
http://www.msnmobilefix.com/Default.aspx
(Snip)
Is there a TinyBASIC for 8088? I know Tom Pittman's is/was available
for 6502 and 1802 (and still is, directly from his website, IIRC).
Unless you are doing work on a small memory machine (minimally equipped
1802 or 6502 or INS8073), I don't know that TinyBASIC is an improvement
over M$ BASIC, but on a machine with 2K-4K, it's really useful.
-ethan The disk are from the Polymorphic archive and
have a lot of valuable
source, manuals and binaries. As I stated earlier, there is even source
for their version of BASIC ( not MS ).
Not a big fan of BASIC, but a non-m$ one would be nice, I suppose.
4:51 p.m.
New subject: div by 10 on Z80
Date: Thu, 24 Jan 2008 01:09:31 -0500> From:
rtellason at verizon.net> To: cctalk at classiccmp.org> Subject: Re: div by 10 on
Z80> > On Thursday 24 January 2008 00:30, dwight elvey wrote:> > There are two
windows in the Z80 memory that are double accessed> > by both the S100 side and the
Z80 side. One window at 1000H for> > buffering while the other is only 16 bytes used
for command, parameters> > and status.> > I'd be real interested in more
details about how they do that.
Hi
Not much to it. There is some state of the Z80 cycle when one
can access memory from someplace else. On the S100 side,
it can be synced with wait states. The 8080 is the slower
side.
I have the schematic but it is not scanned.
Dwight
_________________________________________________________________
Climb to the top of the charts!?Play the word scramble challenge with star power.
http://club.live.com/star_shuffle.aspx?icid=starshuffle_wlmailtextlink_jan
26 Jan
26 Jan
6:15 a.m.
New subject: div by 10 on Z80
From: dkelvey at hotmail.com
To: cctalk at classiccmp.org
Date: Thu, 24 Jan 2008 08:51:42 -0800
Subject: RE: div by 10 on Z80
Date: Thu, 24 Jan 2008 01:09:31 -0500
From: rtellason at verizon.net
To: cctalk at classiccmp.org
Subject: Re: div by 10 on Z80
---snip---
I went back and looked at the schematics. It uses the BUSREQ lines
on the Z80. The circuit is a little too complicated to describe. There
is some gating and because the RAMs on the controller ar DRAMs, things
are a little more complicated than would be needed for statics.
I still believe the fact that the Z80 is suppose to be running faster
than the 8080, it should responded with a BUSAK that handshakes
to the S100 by the PRDY line.
Dwight
_________________________________________________________________
Helping your favorite cause is as easy as instant messaging.?You IM, we give.
http://im.live.com/Messenger/IM/Home/?source=text_hotmail_join
On Thursday 24 January 2008 00:30, dwight elvey
wrote:>> There are two windows in the Z80 memory that are double accessed
by both the S100 side and the Z80 side. One window at 1000H for
buffering while the other is only 16 bytes used for command, parameters
and status.
I'd be real interested in more details about how they do that.
24 Jan
24 Jan
1:08 a.m.
Chuck Guzis wrote:
My apologies, but I've always found bit-twiddling
fun. It seems that
every platform has some peculiar characteristic that can be exploited
to good advantage.
I found this interesting too. Besides, it is no less germane to
old
computers than details of TTL operation.
Some platforms require total submission to peculiarity. I have been
playing with a PDP-8 over the last year. It can only be programmed
efficiently when you accept it as it is. It just does not do modern
programming paradigms.
It is entertaining to spend the evening trying to eliminate the last
unnecessary cycle from a program...
-chuck (a different chuck)
1:48 a.m.
Charles H Dickman wrote:
Some platforms require total submission to
peculiarity. I have been
playing with a PDP-8 over the last year. It can only be programmed
efficiently when you accept it as it is. It just does not do modern
programming paradigms.
It is entertaining to spend the evening trying to eliminate the last
unnecessary cycle from a program...
Well it does make for Software Updates to come out a Tad Slower
than most systems. Look at the ADVENTURE update, almost 20
years later. :)
-chuck (a different chuck)
PS: Does anybody have
the PDP-8 Adventure link,(updated)
and the link for a GOOD DOS terminal program.
Thanks
6:01 a.m.
From: cclist at sydex.com
To: cctalk at classiccmp.org
Date: Wed, 23 Jan 2008 09:31:50 -0800
Subject: Subject: RE: Subject: Re: div by 10 on Z80
Hi Chuck
See my other post. I realized the missing 'add hl,hl'. As you
can see, I moved this to the front of the loop and increased
the value of DE to -1280. This allowed the use of the bit
in the a register!!
Date: Tue, 22 Jan 2008 19:33:15 -0800
From: dwight elvey
Here goes with using your idea to double use the
a register. Except
If I start it with 2, I can use the carry instead of the sign bit. What
fun!
Div10:
ld a,#2
ld de,#-640d
divloop:
add hl,de
jr c, div1
sbc hl,de
div1:
rl a
jr nc,divloop
add hl,hl
ret
Ah, but you forgot the add hl,hl to shift the dividend after each
iteration. Put it at the top of the loop so it doesn't interfere
with the operation of the "rl a/jr nc" pair and scale your divisor by
an additional power of 2 and you should have it.
By now, we've probably bored the hardware guys to tears and DEC
people are tearing out their hair, so we should probably drop the
subject.
I think there is still some interest in coding tricks. Even if they
are using DEC machines.
Dwight
_________________________________________________________________
Shed those extra pounds with MSN and The Biggest Loser!
http://biggestloser.msn.com/
6181
days inactive
6184
days old
12 comments
8 participants
participants (8)
-
bfranchuk＠jetnet.ab.ca -
cclist＠sydex.com -
chd_1＠nktelco.net -
dkelvey＠hotmail.com -
ethan.dicks＠usap.gov -
hilpert＠cs.ubc.ca -
pete＠dunnington.plus.com -
rtellason＠verizon.net