Re: 18 bit CPU; was: Speed now & then

2018-04-15 Thread Chuck Guzis via cctalk
On 04/15/2018 04:47 PM, Jay Jaeger via cctalk wrote: > Yes, the Univac 1100 series were one’s complement (had brief experience as a > student with 1108 and 1110 from 1969 to 1975) I thought I'd originally said as much. 9 bit characters with a range of +/-255 as well as +/- zero. Here are some

Re: 18 bit CPU; was: Speed now & then

2018-04-15 Thread Jay Jaeger via cctalk
> On Apr 15, 2018, at 09:44, Bill Gunshannon via cctalk > wrote: > > > >> On 04/15/2018 02:28 AM, r.stricklin via cctalk wrote: >>> On Apr 14, 2018, at 4:00 PM, Bill Gunshannon via cctech wrote: >>> >>> I'm familiar with Univac's having worked on the 1100 many moons

Re: 18 bit CPU; was: Speed now & then

2018-04-15 Thread Bill Gunshannon via cctalk
On 04/15/2018 02:28 AM, r.stricklin via cctalk wrote: > On Apr 14, 2018, at 4:00 PM, Bill Gunshannon via cctech wrote: > >> I'm familiar with Univac's having worked on the 1100 many moons ago, >> But look at the line above my comment: >> "you assume that a char is 8 bits, with a signed char

Re: 18 bit CPU; was: Speed now & then

2018-04-15 Thread r.stricklin via cctalk
On Apr 14, 2018, at 4:00 PM, Bill Gunshannon via cctech wrote: > I'm familiar with Univac's having worked on the 1100 many moons ago, > But look at the line above my comment: > "you assume that a char is 8 bits, with a signed char having a range > of +/-255". > > An 8 bit signed char has the

Re: 18 bit CPU; was: Speed now & then

2018-04-14 Thread Toby Thain via cctalk
On 2018-04-14 7:00 PM, Bill Gunshannon via cctalk wrote: > ... > I'm familiar with Univac's having worked on the 1100 many moons ago, > But look at the line above my comment: >  "you assume that a char is 8 bits, with a signed char having a range > of +/-255". > > An 8 bit signed char has the

Re: 18 bit CPU; was: Speed now & then

2018-04-14 Thread Bill Gunshannon via cctalk
On 04/12/2018 09:55 AM, Toby Thain via cctalk wrote: > On 2018-04-12 7:48 AM, Bill Gunshannon via cctalk wrote: >> >> On 04/12/2018 02:45 AM, Chuck Guzis via cctalk wrote: >>> On 04/11/2018 06:38 PM, Chuck Guzis via cctalk wrote: On 04/11/2018 02:48 PM, ben via cctalk wrote: > I

Re: 18 bit CPU; was: Speed now & then

2018-04-12 Thread ben via cctalk
On 4/12/2018 6:51 PM, Chuck Guzis via cctalk wrote: On 04/12/2018 03:16 PM, ben via cctalk wrote: Well I have one better with my cpu, a char is -128 to 384. Care to elaborate on why such a lopsided range? --Chuck OK, you forced me to go back to 8 bit unsigned bytes. Not really, but having

Re: 18 bit CPU; was: Speed now & then

2018-04-12 Thread ben via cctalk
On 4/12/2018 7:55 AM, Toby Thain via cctalk wrote: On 2018-04-12 7:48 AM, Bill Gunshannon via cctalk wrote: On 04/12/2018 02:45 AM, Chuck Guzis via cctalk wrote: On 04/11/2018 06:38 PM, Chuck Guzis via cctalk wrote: On 04/11/2018 02:48 PM, ben via cctalk wrote: I have a nice 18 bit cpu

Re: 18 bit CPU; was: Speed now & then

2018-04-12 Thread Charles Anthony via cctalk
On Thu, Apr 12, 2018 at 9:18 AM, Chuck Guzis via cctalk < cctalk@classiccmp.org> wrote: > On 04/12/2018 06:23 AM, Diane Bruce wrote: > > \ > I could see lots of problems doing it any other way, just from the > viewpoint of 'C'. A character on the Cyber 70/170 series is either 6 > bits or 12

sizes and negative ints (Was: 18 bit CPU; was: Speed now & then

2018-04-12 Thread Fred Cisin via cctalk
So, it should be pretty straightforward unless you assume that a char is 8 bits, with a signed char having a range of +/-255. Signed 8 bits would be  -128 to +127. Well, he had previously mentioned NINE bits per character. Besides, 9 bits and 18 bits improves the convenience of using octal

Re: 18 bit CPU; was: Speed now & then

2018-04-12 Thread Chuck Guzis via cctalk
On 04/12/2018 06:23 AM, Diane Bruce wrote: > Amusingly years ago I worked for Computing Devices Canada that used some > CDC computers. I was told through a very reliable source that they > got Unix ported to the Cyber by SoftQuad based in Toronto. They were > well known as a 'troff house' at

Re: 18 bit CPU; was: Speed now & then

2018-04-12 Thread Toby Thain via cctalk
On 2018-04-12 7:48 AM, Bill Gunshannon via cctalk wrote: > > > On 04/12/2018 02:45 AM, Chuck Guzis via cctalk wrote: >> On 04/11/2018 06:38 PM, Chuck Guzis via cctalk wrote: >>> On 04/11/2018 02:48 PM, ben via cctalk wrote: >>> I have a nice 18 bit cpu here, with only a few hardware bugs.

Re: 18 bit CPU; was: Speed now & then

2018-04-12 Thread Diane Bruce via cctalk
On Thu, Apr 12, 2018 at 07:16:58AM -0500, Mark Linimon via cctalk wrote: > On Wed, Apr 11, 2018 at 11:45:07PM -0700, Chuck Guzis via cctalk wrote: > > Looks pretty much like standard C until you get into the minutiae, such > > as "A character constant is 1 to 4 characters" and page 4-4 "Data

Re: 18 bit CPU; was: Speed now & then

2018-04-12 Thread Mark Linimon via cctalk
On Wed, Apr 11, 2018 at 11:45:07PM -0700, Chuck Guzis via cctalk wrote: > Looks pretty much like standard C until you get into the minutiae, such > as "A character constant is 1 to 4 characters" and page 4-4 "Data Types" > (9 bit characters and 36 bit ints and 18 bit short ints). > > So, it

Re: 18 bit CPU; was: Speed now & then

2018-04-12 Thread Bill Gunshannon via cctalk
On 04/12/2018 02:45 AM, Chuck Guzis via cctalk wrote: > On 04/11/2018 06:38 PM, Chuck Guzis via cctalk wrote: >> On 04/11/2018 02:48 PM, ben via cctalk wrote: >> >>> I have a nice 18 bit cpu here, with only a few hardware bugs. >>> Hmm would it work better if I change that around ideas. >>> >>>

Re: 18 bit CPU; was: Speed now & then

2018-04-12 Thread Chuck Guzis via cctalk
On 04/11/2018 06:38 PM, Chuck Guzis via cctalk wrote: > On 04/11/2018 02:48 PM, ben via cctalk wrote: > >> I have a nice 18 bit cpu here, with only a few hardware bugs. >> Hmm would it work better if I change that around ideas. >> >> Care to point to a nice 18 bit version of unix or C. >> BTW The