Allison's paltry collection [Archive] - The Vintage Computer Forums

View Full Version : Allison's paltry collection

Allison

March 18th, 2004, 06:45 PM

Allison

Been on the usenet for about 18 years, COMP.OS.CPM.

Collection of operational hardware:

PDP-8f
Intersil sampler (6100 chipset)
6120 based board, homebrew
2 Decmate-IIIs OS/278

2 LSI-11/03 rx02
3 PDP11/23 various hardware configs
BA11 with 11/23 +tu58
homebrew design using T-11 (the 40pin PDP11)
Rt-11, XXDP-11 and unix V6

Microvax-II (ba23 based)
Microvax-II/GPX (Ba123 based, SCSI disks)
3 Microvax2000 all with RD53 or 54 drives, one with ultrix
2 Microvax3100/m76/gpx
3 Microvax3100/server (not M10e)
VMSv5.4-4, V7.2 netbsd, Ultrix 4.2

Altair8800(pre-A) Built jan 1975 SN200!
2 Northstar horizon, CP/M, NS*dos (one I built in '77)
CCS-2200 CP/M2.2
Compupro full boat with 8085/8088 card and MPX-1 (CCPM)
1 Vt180 complete
2 Vt180 CP/M board built up as standalone one modded for 6mhz
1 Vt185 Thats a Vt125 with CP/M board.
AmproLB+ Cmos modded and running with 45mb 3.5" SCSI
3 Visual technolgies 1050, CPM-3 two with outboard 10mb SCSI disk
SB180 with SCSI adaptor, adptec scsi bridge and 20mb CPM2.2
Kaypro 4/84 w/handyman and Advent turborom+personality card
Osborne 1
Epson PX-8 with 120k ram wedge and 300bd modem wedge
NS* Advantage
Hurkon Z80 Multibus system CP/M2.2
Netronics 8085

Motorola 6800D1 SBC with TBX
National SC/MP board
National Nibble basic [sc/mpII] SBC
NEC TK80A 8080a SBC with protocard
KIM-1
Trackstar 128 (dual 6502 for ISA PC improvement)
Technico superstarter system with assembler roms
Cosmac ELF orgional (built back in '78)
DEC ADVICE VAX chipset on an SBC for in circuit emulation.
IMSAI IMP48 (8035 based SBC)
Homebrew 8039 based SBC (switches and leds pannel!)
Prompt-48 804x development tool/system
NEC EVAkit-48 8048/9 development system/board
2 TI99A (with disks and SW)
8 DEC VT100 printerbuffer cards modded as SBCs (8085 based).

Commodore 128, I keep forgetting that one..

H19 terminal
Vt100/125 terminal
Vt1200 Xterminal
3 Vt320 terminal White, amber and green!
Vt340 Color Terminal

PCs of interest, generally I dont bother but these are interesting.
Leading edge Model D (either V20 or Intel Inboard386)
Tandy 1000 (V20)
2 Vi15g boards VESA with 5x86/133

Oh yes and a severe shortange of storage. I may have forgotten a few
along the way.

Thomas Hillebrandt

March 19th, 2004, 12:21 AM

Niiiiiiiiiice... 8)

Welcome to the Forum, on behalf of all of us, I guess...

The more the merrier :lol:

CP/M User

March 19th, 2004, 12:38 AM

Allison wrote:

> Been on the usenet for about 18
> years, COMP.OS.CPM.

I thought that was you! ;-)

I'm the joke of COMP.OS.CPM !

> Collection of operational hardware:

My heads just spinning looking at that
list!

Cheers,
CP/M User.

Micom 2000

May 19th, 2004, 12:48 AM

Allison

Been on the usenet for about 18 years, COMP.OS.CPM.

Collection of operational hardware:

SNIP

PCs of interest, generally I dont bother but these are interesting.
Leading edge Model D (either V20 or Intel Inboard386)
Tandy 1000 (V20)
2 Vi15g boards VESA with 5x86/133

Oh yes and a severe shortange of storage. I may have forgotten a few
along the way.

Classiccmp is moving to the vintage computer forum.
Good to see you're still functioning even if DEC/Compaq is not.
BTW I have a disk copy of the Intel Inboard setup I scooped from
the Intel site eons ago. And my IBM PC with it is still functioning.
Fired it up today. The Win 1.01 isn't working but I noticed the HD has
a copy of Uniform on it. Don't know which version.

Lawrence

Allison

April 1st, 2005, 07:32 PM

Further update on the paltry collection....

recent addition of a complete Altair 8800B-T.

That is a 8800 turnkey business system complete with 64k of 88-16 static ram, and Altair disk controller and matching drives in the desk. The only foreign board is a Tarbell SD controller. The SD Tarbel is for CPM and the other is used for MITS based stuff. Runs too. I got it from the original owner that used it for business since new around early '79.

Allison

Erik

April 1st, 2005, 08:27 PM

Hello Allison and welcome to the VC Forum!

That's hardly a paltry collection, by the way. I'm impressed! :)

I especially like your latest edition. I hope to find one of those someday and then, sometime thereafter, find the space to play with it!

Enjoy!

Erik

Allison

April 2nd, 2005, 08:13 AM

there are a few things here.

The biggest challenge is space to actually operate the equipment.

Some like my Amprolb+ and the NS* Horizon and the Kaypro4/84 are
always within reach and are part of my "toolkit" systems that I use for real work.

However what I really enjoy is hotrodding systems and my first NS* Horizon, Compupro crate and a VT180 board set up for standalone (no vt100 requried) are for experiments. One such experiment is the VT180 board running at 6mhz. In stock form they are 4mhz z80 64k CP/M engines.

Allison

Chris2005

April 11th, 2005, 11:42 AM

Allison

April 11th, 2005, 01:12 PM

just a few comments...
Allison...I take it the Ampro LB is the one with the 8088. I have the model with the NEC V40, which is more or less an 80186. I think Terry Yager has the one with the actuall '186. They did make them in the 8080 days though.
Why no DEC Rainbow in your collection? Dopey question I guess, but you mean to tell me you never came across one?
Did you ever see a Northstar Dimension? I haven't even been able to find a picture of one.
I want the Leading Edge Model M, which is the same as a unit offered by Sperry. Had one in my clutches but muffed it :(

The the AmproLB+ I have is a 1984 4mhz Z80 powered model.

I have limited interest in 8088 or 80186/188 machines. I do have a handful of 8088 and 80186/188 including some cmos parts. I also have some V20s loose too. The only machine I have of that class I've kept is a Tandy 1000. I have a working one with manuals and software with a V20 in it.

I used to have leading edge model D for and gave it away. Never considered it collectable. Just an old PC XT compatable that worked well.
It was my first nonPDP-11 or VAXen for connecting to the 'net.

DEC Rainbow, every time I get one I give it away. No real intrest in PCs or their sort other than as current or near current application engines. To me other than a few clones and odditites they are not my cup of tea. Even for that I only have a few old Modular System 486/50s, dell 486/66 pizza boxes, color 486 laptops plus a few slow Pentiums.

I do have several DEC Robins as they are CP/M machines. My focus is PDP-8s (small 8s), Qbus PDP-11s, desktop VAXen and of course anything that runs CP/M-80.

I've seen a NS* Dimension but don't have one, nor have actively sought one. I stopped with the Horizon and Advantage.

If I were to seek something to add my list would seem rather odd. Examples would be Bob's (spare time gizmos) SBC6120 or a 1802 based system that has disk and OS, and just about any small single board 8085/z80 system. Other candidates would be a S100 LSI-11 board or
TI9900 for S100.

Allison

Terry Yager

April 11th, 2005, 01:28 PM

Allison

April 11th, 2005, 02:38 PM

Chris,

That's funny, caz when I read "Ampro LB", the one that pops into my mind is the Z80 (Z180?, eZ280?...) model. That's the one on my wish-list. BTW, one of the V-40 models showed up on eBay a couple days ago. It's a complete system, built into a dual 5.25 drive enclosure. (I dunno if it has a handle tho).
--T

Ampro had been around for a while and the and LB+ was z80 but at that same time you could buy a similar board with 80188 on it. Before they disappeard off my radar they'd made it to the V40 class SBC for embedded use.

Another name that should be significant is Micromint with the BCC and SB
series single board microcontrollers. I have a BCC180 (z180) and a SB180 (64180) with the big difference in them being the BCC is for embedded use(Eprom based) and the SB180 series ran CP/M by virtue of a floppy controller being present.

Allison

Terry Yager

April 11th, 2005, 05:54 PM

Allison

April 11th, 2005, 06:51 PM

Before they disappeard off my radar they'd made it to the V40 class SBC for embedded use.

AFAIK, they're still around, selling Pentium-based SBCs in PC/104 (and others) form-factors. Wonder if www.ampro.com will be thier webpage? <disappearing into the datastream> Yup! Still there all right.

--T

The clue should have been dissapeard off "MY" radar. IE: I stopped tracking what they did because they went full PC. PCs and PC/104 to me are YASPC (Yet Another Stupid PC). If I needed YASPC I'd look around at names I know.

Allison

Chris2005

April 12th, 2005, 04:50 PM

Terry Yager

April 12th, 2005, 05:16 PM

Allison

April 12th, 2005, 06:54 PM

Regard that as "stupid pc's" if you will, but they were the first machines where you could accomplish something useful. Was my and alot of other's intro to computers. Sorry, hex pads and toggle switches just don't seem to have the selling point that bit mapped graphics and addressable memory capacities greater than 32k have. A computer is meant to amplify a person's abilities, not simply act as a paper weight.

Well lets disect that.

First I was running CP/M-86 in early 1982 on an 8mhz 8086 multibus system with 1mb floppy drives and a reall megabyte of ram. Compared to the PC-xt it's was light years ahead. The S100 crate with 64k of ram and a hard disk of 5mb I owned at that same time was running database software, multiplan and my favorite tool word processing with spell checking. The PC didn't invent crap, at best the name IBM helped get small computing firmly off the ground. However, it was going that way already.

Actually I abandoned "hex pads and toggle switches" by early 1979 with a NS* Horizon and a H19 terminal. With a disk system, three drives and a full complment of software that included CP/M, NS*dos and UCSD Pascal P-system it was a working tool.

New a friend of mine went a different path with an H11[DEC lsi-11] and H27 disk running a varient of RT-11 and UCSD Pascal. By 1979 standards it was already 16bit and a it would take a Turbo PC AT to match it for basic performance. By time the PC hit a full 640k the 11/23 was commonly running 4mb.

Sorry, the PC as a serious tool only hit stride with the midlife 286s and was wanting for the later 386s and a marginaly useful OS. By then we
were in the latter half of the 80s.

My running joke for years about PCs is the PDP-8 used to do what PC users wished for. The only difference is size. That and for security I'll run VAX/VMS against a PC anytime.

Actually the breaking point for me to switch from CP/M as a working environment full time was 1988, The replacement was an 11/23 with
RD52 (30mb) disk and RSTS, RT11 and RSX-11. That was followed by a MicroVAX-II and by 1991 I was running DECwindows on an eithernet connected Xterminal. They live on the network in my room that's now reaching 15 years old. It was nearly 1996 before I started seriously using PCs and even now they are still also ran for some things.

I use PCs when being first to the scene of a crash is not to inconvenient or when others who were of the true believers wanted theirs to work and paid me to make that happen. Remember even the newest PC has two buttons in easy reach power and reset.

Look back hard, the PC was not the be all. It was only the sloppy continuation of what started with 8080/z80. I believe the PC XT is historical in that it represented the evolotionary wrong turn that needed
10 years of legacy pain into the Pentium era to winnow out.

Allison

patscc

April 13th, 2005, 04:33 PM

[quote="Chris2005"]Regard that as "stupid pc's" if you will, but they were the first machines where you could accomplish something useful. Was my and alot of other's intro to computers. Sorry, hex pads and toggle switches just don't seem to have the selling point that bit mapped graphics and addressable memory capacities greater than 32k have. A computer is meant to amplify a person's abilities, not simply act as a paper weight.[/quote

Well, I don't know about bit mapped graphics.

Stock PC's came with MDA, which was a text-only card, or you could buy a CGA card, with crappy resolution.
It wasn't until Hercules brought out it's card that bit-mapped graphics were conveniently available for the PC.
Sure, you could run a PC with an MDA, and a CGA, but that was expensive, and required you use "mode" to switch between monitors. A few apps like 123 supported dual-monitor use.
The "stock" OS for the PC. DOS, certainly didn't give a hoot about bit-mapped graphics.
And for for many,many years, IBM graphics lagged far behind almost everything else out there.
The real graphics machines were Amigas or Macs, if you were into bit-mapped graphics.

About doing useful things.

The Apple II had been out for several years, and there were tons of educational and scientific apps out there for it. And games, you could get some pretty good graphics on Apples.

Ditto for the TRS-80 line.

WordPerfect 1.0 was released, I think for DataGeneral systems.
VisiCalc, a spreadsheet, was releaesd for the Apple II.

Morrow shipped 26 MByte hard drive systems for their CP/M 80 systems.

HP's model 85 luggable became quite popular for controlling scientific equipment, it was programmable, and very accurate due to it's BCD math.

And all this before the first IBM PC ever shipped, and surely something in the above list can be considered useful.

So, IBM decides to jump on the bandwagon:

The PC used a 8088, which was mainly since they had already licensed it to manufatcture. Legend has it IBM originally wanted to use it in a "smart" typewriter.

The 8088 has got a half-assed method of addressing memory, which for the next 10 years had every programmer who had to deal with their segmented memory model cursing IBM and Intel.

For those of you never having to deal with segmented memory, it basically limited you to 64k chunks of memory for a lot of what you do. You have to "shift" your addressing around to get at the next 64k chunk. (I know, this is really simplified, so please don't flame me)

This is similar to bank-switching, which is how a lot of other computers at the time were able to access larger amounts of memory than their address bus would normally permit. All Intel did was to jam the bank-switching into the chip, so they could be somewhat competitive with it.

So, yes, you could deal with more than 32k. A lot of other machines could as well.This is where in IBM land near, far, and huge pointers started happening.
To further confuse the issue, bank-switching on the PC platform was made official with the expanded memory boards, non of which were compatabile, but finally brought about the LIM-EMS 4.0 standard.

The reason for all this memory need ? DOS has no provision for virtual memory, and the killer app of DOS, Lotus 1-2-3, required that the entire app space be contained in memory. Hence, more memory was required.
This trend of throwing more memory at the problem( for the PC platform) continues to this day, look at Windows with it's Unicode and MCBS character sets. Sure, it's nice for internationalization, but that was managed with ANSI long ago, and you only needed 1 byte per character.

The MC68000, the other 16-bit processor of choice at the time, certainly didn't muck around with segments. (There were other 16-bit chips and sets as well, 8086 and 68000 are the only ones I have direct 'coding' experience with)
MC68000 machines had linear memory. Very simple, very elegent.
Here's a good link that goes into this:

http://linux.cis.monroeccc.edu/~paulrsm/doc/dpbm68k1.htm

Here IBM had a chance to give us all a workstation-class machine offering flat addressing, and instead we got a glorified "smart" typewriter.(Anything more powerfull might have cut into their mainframe and mini market)
And they didn't even use the 8086, they used the 8-bit data bus version so they could use more off-the-shelf parts.

Any time you make it harder for a programmer to do something, or more complicated than it has to be, application and platform stability suffer, and I'm sure all of us that used PC's remember how often they crashed, and how buggy a lot of the apps were(are).

To me, the ability to do something useful requires your tool be reliable.
You'll notice that IBM and Microsft even gave us the "three-finger-salute" for free to help us deal with reliability issues.
Speed is important as well, and the CISC, multiple cycle instruction set of the 8086 wasn't really the quickest chip out there at the time.

I guess it depends on how you define useful.

They missed the wagon, as far as I'm concerned. Even when laptops started happening, here's what IBM did:

In '83, the Kyocera-driven laptops where coming onto the market( Think TRS-80 Model 100)
They ran off batteries, for up to twenty hours. They had a real keyboard. Journalist loved them. Biologists loved them for data entry in remote (no power) locations.

Epson sold it's HX-20, same size, nicad's, printer, microcassete drive.

IBM (in 86) comes out with the PC convertible, an oddball of machine where you stacked "packs" onto the back to expand the unit. This increased the length of the unit. I have one, with the RS232 module, CGA module, and Ribbon Printer Module, and it's close to 2 feet long.

To me, IBM at the time clearly weren't the most cutting edge guys.

About the only thing IBM got right with it was that ISA bus didn't require any licensing, so virtually everyone could make cards for it. (And later they rethought even that, with the PS/2's; the MCA architecture required licensing from IBM)That, and the manuals. The old technical manuals were so complete( including BIOS listing), the PC practically had "clone me" written all over it.

I think what sold a lot PC's was Lotus-123, and the fact that it came from a "reputable" company.
The big purchasers of PC's, businesses, probably were already running an IBM mainframe, and IBM had an eviable reputation for quality due to their mainframes, and their typewriters.

Hex pads and toggle switches:

Well, I'd have actually prefered hex pads and toggle switches to the nightmare of jumper settings you had to fiddle with everytime you wanted to add an expansion card to your PC. Cards were very jealous about the address space the wanted, the interrupts they required, the list goes on and on.
I mean, it was a great way to make a living, because most people didn't want to fuss with jumpers.

The only reason I got one in the first place is because clones were so cheap, and I couldn't afford anything better at the time.

patscc

Allison

April 13th, 2005, 05:12 PM

>>Well, I don't know about bit mapped graphics.

For S100 in 1976 there was the Comemco Dazler, color graphics for S100
and on a par with Herc in the PC 5 years later. You got that right. the PCs were a throw back and took years to get on track. I'd say the XT
and 286ATs really didn't leave the scene till the early 90s.

>>The Apple II had been out for several years, and there were tons of educational and scientific apps out there for it. And games, you could get some pretty good graphics on Apples.

Considering in 1980 two others plus myself did a cross platform forcasting program using UCSD Pascal. the platforms, S100 NS*, TRS80m1, Apple][.
The programming was the same for all three at the source level.

>Morrow shipped 26 MByte hard drive systems for their CP/M 80 systems.

When the IBM PC hit the street I bought my ST506 (5mb) and a Teltek HDC for S100. The IBM equivilent was a nearly a year later. It was also slower.

>>HP's model 85 luggable became quite popular for controlling scientific equipment, it was programmable, and very accurate due to it's BCD math.

It had GPIB and it set a standard for labratory systems.

>>The 8088 has got a half-assed method of addressing memory, which for the next 10 years had every programmer who had to deal with their segmented memory model cursing IBM and Intel.<<

That was the evolutionary wrong turn I refered to. I've worked with the 8088/6 and it was horrible to use past 64k.

>For those of you never having to deal with segmented memory, it basically limited you to 64k chunks of memory for a lot of what you do. You have to "shift" your addressing around to get at the next 64k chunk. (I know, this is really simplified, so please don't flame me)<<

Even the Z280 was better in that it was not only 16bit address with MMU
taking it to 24 it had the seperate I&D space thing that DEC did on the PDP-11. Still the Z280 was an 8bitter but a lot better than 8088. It's failing was it was way late. The only one that had a real linear address
was the 68000.

>>They missed the wagon, as far as I'm concerned. Even when laptops started happening, here's what IBM did:

In '83, the Kyocera-driven laptops where coming onto the market( Think TRS-80 Model 100) They ran off batteries, for up to twenty hours. They had a real keyboard. Journalist loved them. Biologists loved them for data entry in remote (no power) locations. <<

And you could buy new batteries anyther in the world!
My Epson PX-8 is still better than on batteries compared to any PC laptop for data logging and other tasks. the PX8 even had graphics.

>>About the only thing IBM got right with it was that ISA bus didn't require any licensing, so virtually everyone could make cards for it. (And later they rethought even that, with the PS/2's; the MCA architecture required licensing from IBM)That, and the manuals. The old technical manuals were so complete( including BIOS listing), the PC practically had "clone me" written all over it.<<<

They got ISA wrong too. ISA-8 is basically the same signal set as the Intel multibus with one BIG error. IBM inverted the interrupts and that made "wired ORing" of interrupts very difficult. Big oops in my book.

>>Hex pads and toggle switches:

Fabulous debug tool they were. However by 1980 it was more optional that standard fare on S100 stuff.

I'd tack on one thing. I took a 8085 and added a simple MMU and tossed that on a PC board to plug in where a 8088 went in a XT clone. Fiairly straight forward actually as theinterface for 8088 in minimum mode and 8085 are nearly the same save for 4 address bits. Pulled the rom and created a boot monitor and it became a really nice expanded memory, two floppy CP/M system that OBTW was faster than the 8088. I'll admit the 8085 was running at 6.144mhz (8085-6 hmosII part). Looking back I'll have to find another XT and dig out that kluge as it was fun. For word
processing it was faster than the PC running the 8088 at 4.77mhz!

Allison

Terry Yager

April 13th, 2005, 06:06 PM

I'd tack on one thing. I took a 8085 and added a simple MMU and tossed that on a PC board to plug in where a 8088 went in a XT clone. Fiairly straight forward actually as theinterface for 8088 in minimum mode and 8085 are nearly the same save for 4 address bits. Pulled the rom and created a boot monitor and it became a really nice expanded memory, two floppy CP/M system that OBTW was faster than the 8088. I'll admit the 8085 was running at 6.144mhz (8085-6 hmosII part). Looking back I'll have to find another XT and dig out that kluge as it was fun. For word
processing it was faster than the PC running the 8088 at 4.77mhz!

Sounds great. Are you going to make the plans & source available...or even better yet, kits? Put me down as your first customer, I've always thought there must be some way to make all those old XTs useful.

--T

Allison

April 14th, 2005, 05:40 AM

Sounds great. Are you going to make the plans & source available...or even better yet, kits? Put me down as your first customer, I've always thought there must be some way to make all those old XTs useful.

--T

I did this 12 years ago. No scanner and also the sources for the boot monitor were lost to a crash. If I did it again I'd revisit the whole show.

If I get a suitable candidate XT I may do it again. I haven't decided if the Tandy 1000 I have is a good candidate. Though that has a V20 in it already.

Allison

Terry Yager

April 14th, 2005, 07:14 AM

Damn! 12 years too late...story of my life...

--T

Allison

April 14th, 2005, 08:03 AM

Damn! 12 years too late...story of my life...

--T

Well by now it should be obvious that I really like to experiment. Some I keep, some are just because.

Actually I've considered doing it again. I do have the parts and all. The
tandy 1000 is a good one as it works and would easily look cool running CP/M. Then I start thinking.. I also have a handful of T-11 chips, those are base PDP-11 and their external interface was somewhat similar to the
8085. A Tandy 1000 running RT-11 would be cool and unusual!

I have to find my notebook and a pen.

Allison

Terry Yager

April 14th, 2005, 08:43 AM

No...! Don't butcher the T1K, it already knows how to run CP/M natively on it's V-20. I have a couple of different versions of CP/M for the V-20. I can send 'em to if ya want.

--T

Allison

April 14th, 2005, 09:41 AM

No...! Don't butcher the T1K, it already knows how to run CP/M natively on it's V-20. I have a couple of different versions of CP/M for the V-20. I can send 'em to if ya want.

--T

I know that. about the V20. Though CP/M for it is a desireable item. I'd want to run CP/M-80 and I guess I should check to see if a Tandy 1000 version is avalible on line.

Allison

Terry Yager

April 14th, 2005, 01:51 PM

Allison

April 14th, 2005, 03:55 PM

I dunno about CP/M-80 on a T1K, but I do recall running CP/M-86 v.1.1 for the IBM on mine back in the day. Worked pretty smooth, but I could never get it to see any more than 128K of memory, even though the 1000 had the full 640K. IIRC, the first 128K was mounted on the mainboard, and the rest on an expansion card, but I don't know if that has to do with it or not. Mebbe you could add that to your project list: figger out how to see all the memory under CP/M-86.

--T

Well getting cpm-86 to see all of available memeory is easy. You have to basically relocate it. CP/M-86 is a fixed system in that it doesn't autosize.
The fact that the additional 128k is on a physically different board has no meaning.

I know CP/M-86 is available for it but no interest there. I'd want to run
CP/M-80 as that were all the programs are. However, it would ahve to have a hybrid bios as the sytem would ahve to boot as 8088 a load the
bios and then switch to 8080 most likely.

Myself I'd rather run completely 8080/8085 or Z80.

Allison

Terry Yager

April 14th, 2005, 06:01 PM

Allison

April 14th, 2005, 06:43 PM

What's really weird is that the same disk in a Tandy 1000EX would see 256K on that hardware. That's why I thought there might be a connection, something weird about the way the T1K uses it's memory, beause the EX had 256K on the mainboard. Just coincidence then, eh?

--T

May well be. I have T1000 and 1000HX techmanual sets. Maybe I'll look
if I get a chance. I dont think I have the EX though.

Is the EX the one with the 80188?

Allison

patscc

April 14th, 2005, 07:23 PM

I think the EX had a 8088. I honestly can't remember if any of them used a 8088, there were so many different members of the T1K family. I know the TL's used 80286, but where really more XT clones than AT clones.

Terry, you're right about the T1K (pre-HX) and memory. Since the original was more or less a PCjr clone, never mind that the PCjr stopped selling right around the time the T1K came out, Tandy decided to do the same the PCjr did, and that was to share main and video memory. They also dropped the DMA and used the video controller to generate RAM refresh. Of course, being Tandy, they threw in the extra twist that in low-mem systems, the BIOS stole back some of the memory from the video controller, and to make things worse, I think access to this "stolen" memory actually had to occur through the video controller.
There used to be a utulity that let you change the memory split on the fly, or at boot, or something like that, but damned if I can remember this.

I'm guessing this "split" memory is why CP/M doesn't see all memory. I've never actually tried CP/M on a T1K, but of course, will now have to dig around for a version and play with it.

The original Tandy memory expansion boards did contain a DMA controller, and eventually they split memory subsystems, which made everyone's live easier.
patscc

Terry Yager

April 14th, 2005, 07:25 PM

Terry Yager

April 14th, 2005, 07:39 PM

pat,

I'm pretty sure the EX is a 8088, I know it's not a V-20. Yes, the TX had a '286, but I don't recall what was in the TL. The TL was a pretty late model, so it may have been a '286. Even the TX's '286 was "crippled" by Tandy, when they cut costs by tying it to an 8-bit bus. They got around the video memory "theft" problem by providing sockets on-board for extra memory, up to 1Mb, but the extra RAM could only be used as video RAM, which freed up the main memory again.

--T

patscc

April 14th, 2005, 07:50 PM

Okay, so at least 2 out 3 think the EX ran on a 8088, so I guess we can call that official.
What was odd was that you could actually run a V20 in some of them, but of course not all of them. I think the general idea was try it, if it worked, you lucked out, if it didn't, oh well.
Yeah, see, I remembered there was something funky with the vram, but I'd forgotten about the limitation about the extra memory being only for video.
I totally forgot about the T2K series, I don't think I ever owned one.
patscc

Allison

April 15th, 2005, 07:48 AM

No, the EX is an all-in-one console, just like your HX, except with a 5.25" drive mounted in the side, instead of the front-mounted 3.5" that the HX has. I have an extra one on-hand, if you'd like one to practice on, I could send it. The TRS-80 Model 2000 (T2K) is the one with the '186 CPU (@12MHz, IIRC).

--T

Took a quick peek at my docs. I have the 25-1504 (t1k) and 25-1513(t1kHX) docs. I have a console, an HX. I rememberd when looking at the console why I kept it. It was small and used a Video
(rs170) monitor or the usual PC tube. My only comment beside being
fully working and cute is it's mostly useless.

As to practicing on one, nope, if I do like the last one hack it was pull the 8088 (v20) out and plug in the kluge board. Change the BIOS Eprom and work out the firmware and timing bugs. I don't know if I will anytime soon
as I have my head inside some S100 hardware doing things.

Other models were the 1200, 1400 and 2000 but I have no info on them
besides knowing one used the 80186 and was very fast but not very compatable.

Allison

Terry Yager

April 15th, 2005, 09:14 AM

Allison

April 15th, 2005, 03:30 PM

Took a quick look at www.radioshack.com. DaShack is kind enough to provide on-line documentation for almost everything they ever sold. The TL & TL/2 were definitely '286-powered, and the EX was an 8088.
The 1200s were straght-up IBM clones, unlike the rest of the 1000 family, which, as pointed out already, were not-quite-compatible. The 1400 series (LT, FD, HD) were V-20 powered laptops, with 768K of RAM, but unlike the 1000TX, the extra RAM could only be used as a RAMdisk, not VRAM. I don't know if it had the same memory limitation as the T1K or not. I do know that CP/M-80 runs great on my HD.

--T

I know they had a lot of "flavors". I only have that lonely T1000HX. I upgraded it to a V20 as I had a few around. Its a nice enough box but,
near impossible to expand as the connectors are not ISA despite the signals being identical on first pass.

Like I said if it weren't for the Composite video connector I'd have traded or sold it. It's oddly inconsistant with most of my other items. The only other stranger is the Commie 128 but that runs CP/M so it has an excuse.

Allison

Terry Yager

April 17th, 2005, 03:01 PM

There certainly were a lot of them sold. At one point, the T1K held the title as best-selling computer in America. As for expansion options, the "Plus Cards" were few and far between (I think there was a memory board, a serial port, a (300-baud?) modem, and not much else). However, Tandy did market a Plus Card adaptor that would accomodate a short ISA card, so you still had a little flexibility. Third-parties also made other options, like external hard drives, but they all cost waaayyy too much.

--T

Terry Yager

April 17th, 2005, 03:13 PM

Allison

April 17th, 2005, 03:47 PM

Couldn't resist the price, I hadda buy the LittleBoard on eBay.

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5184961374&rd=1&sspagename=STRK%3AMEWN%3AIT&rd=1

I think I'll add a nice handle to it and bundle it with an Informer portable terminal that I have, then it'll fit in with the other portables in my collection.

--T

I never check Ebay items. I presume it's a z80 version. I have the LB+ which has SCSI via a 5380. It bit of time spent back when a 45mb 3.5"
SCSI was still desireable stuff netted me a nice system. FYI: I use 3.5" floppies as media is more accessable and they give about 780kb of storage. the whole thing is packages in a aluminum case 4x12x13 inches
though it requires an external terminal anything will serve such as a PC
running Procom. The usual terminal is a VT320, it sits nicely on top of that
case. It's my most used CP/M system second to a modded kaypro4/84
and my 10mhz z80 s100 crate.

Allison

Terry Yager

April 17th, 2005, 04:10 PM

No, the Z80 LB is still on my wish list. This one is the V-40 (yaspc) version, built into an external floppy drive enclosure. I got it for the minimum bid of $5.

--T

Allison

April 17th, 2005, 07:14 PM

No, the Z80 LB is still on my wish list. This one is the V-40 (yaspc) version, built into an external floppy drive enclosure. I got it for the minimum bid of $5.

--T

A yaspc is a PC or PC clone that runs winders or Dos out of the box.
I'd bet a V40 doesn't fully qualify. Besides for a V20 with DMA and
interrupt logic and can still run 8080 code, it's gotta be cool.

And for that price way coooler.

Allison

Terry Yager

April 17th, 2005, 07:43 PM

Gotta try running CP/M on it, no bought adoubt it.

--T

patscc

April 17th, 2005, 08:31 PM

patscc

April 17th, 2005, 08:35 PM

Terry Yager

April 17th, 2005, 08:53 PM

LeatherFace, my K-10 runs both CP/M & MSDOS (well, sorta). Yaspc is Allison's word, from earlier in this thread (fire up your WayBack Machine & set it for a page or two backwards). The rest of us just adopted it.

--T

patscc

April 17th, 2005, 09:44 PM

Whew, I thought I missed something. I got yaspc, I didn't realize it was introduced here.

Question: if a system can run DR-DOS, can it be called yaspc ?

patscc

Allison

April 18th, 2005, 06:15 AM

V40 more than qualifies.
What you want is a Olivetti Prodest PC1.
Ran both MS-DOS, and CP/M

Say thanks to my ex for me not having one anymore.
I suspect it would also have run Windows 1.x (not that it took much), but didn't have a chance to try this out.

yaspc is a new term on me, though.
For some reason, yaspc sounds like something out of a Slackware distribution.

Should anyone have one that needs a home....
patscc

The idea of YA**** is old unixism. IE: YALOE is Yet Another Line Oriented Editor (part of pascal P system!), YASBEC was a Z180 SBC and it stoof for Yet Another Single Board Computer.

You can run Win3.0 on 8088. However you really don't as it will be slooowwwwww.

Allison

Allison

April 18th, 2005, 06:25 AM

Sort of an addendum, since we're on the topic of NEC's V-series.
Why V20 ? Was there a V10 ? What ended up happening to the V-series, anyway ? I seem to remember a V70, altough I'm not at all sure about this. I'm pretty sure there was at least a V50, though.

patscc

No V10 I know of, I was with NEC when the V20 was introduced and did see the internal product planing for the series through V70. As far as a my books go I have data for V60. I'll bet a penny they did do V70.

Allison

Allison

April 18th, 2005, 07:15 AM

updated list of stuff:

Collection of operational hardware:

PDP-8f
Intersil sampler (6100 chipset)
6120 based board, homebrew
2 Decmate-IIIs OS/278

1 LSI-11/03 rx02
3 PDP11/23 various hardware configs
2 PDP11/23B
1 PDP11/73 with DEC sound card (gigilo)
BA11 with 11/23 +tu58
homebrew design using T-11 (the 40pin PDP11)
OSs Rt-11, XXDP-11 and unix V6

Microvax-II (ba23 based)
Microvax-II/GPX (Ba123 based, SCSI disks, TK50)
3 Microvax2000 all with RD53 or 54 drives, one with ultrix
2 Microvax3100/m76/gpx
TLZ04 tape
3 Microvax3100/server (not M10e)
+ BA42 SCSI disk expansion cases.
OSs VMSv5.4-4, V7.2 netbsd, Ultrix 4.2

Altair 8800(pre-A) Built jan 1975 SN200!
Altair 8800B-t Complete with Altair disk system and drives ca1978
2 Northstar horizon, CP/M, NS*dos (one I built in '77)
CCS-2200 CP/M2.2 complete system
Compupro full boat with 8085/8088 card and MPX-1 (CCPM)
1 Vt180 complete (inside a VT100)
2 Vt180 CP/M board built up as standalone one modded for 6mhz
1 Vt185 Thats a Vt125 with CP/M board.
AmproLB+ Cmos modded and running with 45mb 3.5" SCSI
3 Visual technolgies 1050, CPM-3 two with outboard 10mb SCSI disk
SB180 with SCSI adaptor, adptec scsi bridge and 20mb CPM2.2
Kaypro 4/84 w/handyman and Advent turborom+personality card
Kaypro-II vanilla complete with Vinyl cover case.
Osborne 1
Epson PX-8 with 120k ram wedge and 300bd modem wedge
NS* Advantage
Hurikon Z80 Multibus system CP/M2.2
Netronics 8085 (with expansions A thru E)

Motorola 6800D1 SBC with TBX
National SC/MP board
National Nibble basic [sc/mpII] SBC
NEC TK80A 8080a SBC with protocard
8 DEC VT100 printerbuffer cards modded as SBCs (8085 based).
BCC180
Cosmac ELF orgional (built back in '78)
Basys-1 1802 with memory and rom on STD sized board
IMSAI IMP48 (8035 based SBC)
Homebrew 8039 based SBC (switches and leds pannel!)
Prompt-48 804x development tool/system
NEC EVAkit-48 8048/9 development system/board
KIM-1
Trackstar 128 (dual 6502 for ISA PC improvement)
Technico superstarter system with assembler roms
2 TI99A (with disks and SW)
DEC ADVICE uVAX-ii chipset on an SBC for in circuit emulation.

Commodore 128, I keep forgetting that one..
COCO-III no case, runs.

Terminals:
H19 terminal
Vt100/125 terminal
Vt1200 Xterminal
3 Vt320 terminal White, amber and green!
Vt340 Color Terminal

PCs of interest, generally I dont bother but these are interesting.
Tandy 1000hx (V20)
4 Modular systems 486/50 with NIC,and 16meg ram. (runs W95!)
5hx3wx12d form factor, 12V power.
2 DELL 466/np 486/66, 32mb, NIC, pizza box one runs NT4, other W95b

MISC boards:
Intel Inboard386
AST 6pack pro
DC Hayes S100 1200 baud modem
Compupro 68000 CPU S100
SWTP Ct1024 built in 1975 (exactly 30 years ago) by me.
8 Vt180 base CPU boards (goes in VT100) Z80,64k, FDC, 4 serial
Imagewise video reciever standalone version
Extensive selection of S100 boards and Qbus Boards.

MISC CPU and procesor chips:
8008, also have my 8008 manual from 1972.
8080s various old ones.
8085s many vendors all old
Z80s
8088s and 8086s
V20s
80C188s and 80C186
6502s
6800s
6809s
Z8002, Z8000DMA, Z8000mmu
Z8001
2901s (Q 20) 4bit slice
29116 16bit slice
upd7801 ucom78
upd78PG11 ucom78 piggyback Eprom
1802s
TI9900s
T-11 (DEC 40 pin LSI-11/pdp11)
upd372 (early FDC chip that predats 1771)
upd371 Tape controller chip

I have a small mountain of Ram Eprom and TTL to
maintian with.

Allison

Terry Yager

April 18th, 2005, 12:11 PM

Allison

April 18th, 2005, 02:36 PM

I went a-googling last night, but couldn't find anything authoritative re: NEC V-series. Even NEC seems to disavow any knowledge of such, although they do have a current product called V850, but no info whether or not it owes it's ancestry to the original V-xx series or not. Actually, I had never heard of the V-40 until last week, then the ref came up in a couple different places at once, which got me intrigued at the prospect of running CP/M on one. (Now I can hardly wait to try it).

--T

Try NEC Electronics USA. Tey are the markeing and sales arm for NEC microprocessor and memory products.

My 1991 data boot list up to the V55.

They tended to target embedded and telco.

Among the odd Vseris I do have I have the V20H, thats a fully statinc CMOS version of the V20 (very low power). The V40 is a MS-dos compatable version With periperal IO, The V40H is the static CMOS version. V50 is the 8086 flavor. The numbering is Vx=odd it's x86
flavor bus and Vx=even it's 8088 flavor bus.

Allison

Chris2005

April 19th, 2005, 12:29 PM

Somebody could stand to spend some time dissecting a dictionary lol.

I really shouldn't bother with any of this, seeing that it may just drag on for 5 years or more, and at least one person most definately has to "wind" to sustain it. But moving right along anyway...

>First I was running CP/M-86 in early 1982 on an 8mhz 8086 multibus >system with 1mb floppy drives and a reall megabyte of ram. Compared >to the PC-xt it's was light years ahead.

I'm not familiar with the multibus thing, but just met someone who used to own one. If you read down, you seem to consider the 8086 a kludge, so how is this even relevant? The Victor 9000 packed 1.2 megs onto double density disks, and even the Tandy 2000 used quad density drives. I'm interested to know what you mean by "a reall meg of ram". Light years ahead? I think that's an overstatement.

>The S100 crate...

You said it. Many built inside of milk crates lol

>The PC didn't invent crap...

No one said it did, but just did everything the others did faster generally and w/less constraints.

>...at best the name IBM helped get small computing firmly off the >ground. However, it was going that way already.

Right, but as with lots of things, unless one of the big guns get into the game, standardization usually takes lots longer. It may not have been perfect, but you don't have to hate it just because it was popular.

>Actually I abandoned "hex pads and toggle switches" by early 1979...

Lol I was actually kidding about that stuff. I never suspected peeps were actually using them as late as 1979!

>My running joke for years about PCs is the PDP-8 used to do what PC >users wished for. The only difference is size. That and for security I'll >run VAX/VMS against a PC anytime.

What's the point of comparing a micro w/a mini? Remember, a pc will always fit in yer trunk, a mini never will. You have to drag it behind. VMS only runs on big iron.
And I hope you're not talking about the PDP-8 I saw at the MARCH exhibit at the Trenton Computer Fest this past weekend. Because that would indeed be a very big joke. A fascinating early beast, which I'll bet was unparalleled, and smaller than the alternatives (ahem like the PC). Particularly interesting to people, like me, who have a smattering of assembler under their belt. I especially got a kick out of watching the operator load the boot sequence with the toggle switches, whereby the OS? would subsequently be loaded from the paper or mag tape. But this thing definately was your grandaddy's PDP-8 lol. Weighed in at about 600 or so pounds, I help load it back in the truck, I would know. And I know for a fact a Mindset - Y-A-S-P-C - would run rings around it. Prolly even a Peanut lol. Sorry Mike.

>Actually the breaking point for me to switch from CP/M as a working >environment full time was 1988, The replacement was an 11/23 with
>RD52 (30mb) disk and RSTS, RT11 and RSX-11....

Yada yada yada. Blah blah blah. Whatever lol.

> Remember even the newest PC has two buttons in easy reach power >and reset.

Ok...fine. But in reality that's a good thing. I'd hate to have to wind up that crank up everytime to fire up my PDP-?, like an old Model T.

Next...

>Stock PC's came with MDA, which was a text-only card, or you could buy a CGA card, with crappy >resolution. It wasn't until Hercules brought out it's card that bit-mapped graphics were conveniently available for the PC. Sure, you could run a PC with an MDA, and a CGA, but that was expensive, and required you use >"mode" to switch between monitors. A few apps like 123 supported dual-monitor use. The "stock" OS for the PC. DOS, certainly didn't give a hoot about bit-mapped graphics. And for for many,many years, IBM graphics lagged far behind almost everything else out there. The real graphics machines were Amigas or Macs, if you were into bit-mapped graphics.

There were loads of aftermarket graphics cards for the pc. Are you forgetting IBM's PGA? Sure it was expensive but years before VGA came out, you could have nearly 500 lines of resolution and analog color. Remember, the original pc's were designated for business use (iBm). And the routines for producing bit maps were built into the bios, so I don't know where you're coming from there. The Mac came out years later, was slow, expensive, miniscule, and monochrome. For $2500 I don't think you even got a dedicated CRT controller, the 68000 absorbed all the overhead. The Amiga was certainly capable for certain types of work, but also required expensive add-ons, and high resolutions only came with a trade off - interlacing.
And I ain't running down Macs or Amigas here. Let's just be fair with our comparisons.

>The Apple II had been out for several years, and there were tons of educational and scientific apps out there for it. And games, you could get some pretty good graphics on Apples.

I wasn't suggesting you couldn't get anything done with the alternate platforms. There day was coming to end though. The Apple's graphics were attrocious compared to the pc's. Perhaps a few more colors, and may have had some redemption with hardware sprites like the C64, I'm not sure about that though. I looked at all of these before I bought me first puter. A PC was the only logical choice.

>Ditto for the TRS-80 line.

TRS-80 means alot of things to alot of people. To some it just meant trash lol. My Tandy 2000 was a TRS-80 - 640 x 480 x 8, with a pallet of 16 colors. The earlier TRS-80's and Coco's had miserable graphics. I happen to have a fondness for both, in particular the Model 3/4, but there was no comparison with any of them.

>The 8088 has got a half-assed method of addressing memory, which for the next 10 years had every programmer who had to deal with their segmented memory model cursing IBM and Intel.

The PDP-8 also used segmented addressing I'm told. The 8088 wasn't the first chip to do that. And remember - no one foresaw software requiring a full meg of memory. The original PC mobo had 64k soldered on, and don't quote me but I think you needed a card to advance it any further. As a side point, the original mobo also had a provision for cassette storage. Those were early days. It's always so easy w/20-20 hindsight to point out what could have been done differently.

>The MC68000, the other 16-bit processor of choice at the time, certainly >didn't muck around with segments...MC68000 machines had linear >memory. Very simple, very elegent.

I guess you weren't aware that IBM came out with a 68000 based micro. You had the oppurtunity to buy one, but bungled it. Yer problem lol.

>Here IBM had a chance to give us all a workstation-class machine >offering flat addressing, and instead we got a glorified "smart" >typewriter.(Anything more powerfull might have cut into their mainframe >and mini market). And they didn't even use the 8086, they used the 8->bit data bus version so they could use more off-the-shelf parts.

That's business. And if they had come out with the super duper whatever, it probably would have cost 3x more. The PC was something of a bridge between the 8 and 16 bit worlds. There were loads of stuff for cpm and whatnot still floating around. One or two of their attempts to introduce a pc flopped, so they wisely decided on allowing for the use of earlier os's and apps. That trend still continues, because as you know, compatibility is everything.

>Any time you make it harder for a programmer to do something, or >more complicated than it has to be, application and platform stability >suffer, and I'm sure all of us that used PC's remember how often they >crashed, and how buggy a lot of the apps were(are).

There's a steeper learning curve with the Intel uP's then some other I'll grant. But once you start compiling code, instead of assembling, the chips architecture is totally transparent. Apps are buggy for lots of reasons. Software for any platform is going to have problems early on. The burden has to be placed on the programmers, then and today, not the computer.

>About the only thing IBM got right with it was that ISA bus didn't require >any licensing, so virtually everyone could make cards for it. (And later >they rethought even that, with the PS/2's; the MCA architecture >required licensing from IBM)That, and the manuals. The old technical >manuals were so complete( including BIOS listing), the PC practically >had "clone me" written all over it.

Actually, the original tech manuals were loaded with errors lol. Yeah, they opted for an open architecture, and thank goodness. Few other companies did. Apple tried years later, but it was too little too late.

>Well, I'd have actually prefered hex pads and toggle switches to the >nightmare of jumper settings you had to fiddle with everytime you >wanted to add an expansion card to your PC. Cards were very jealous >about the address space the wanted, the interrupts they required, the >list goes on and on.

We're talking about 2 entirely different things here. You'd rather boot your computer - manually - then pop in a disk? And plug and play wasn't around in 1981, so you're complaining?

>The only reason I got one in the first place is because clones were so >cheap, and I couldn't afford anything better at the time.

That goes for alot of us. If it wasn't economical enough, we wouldn't have one at home, on our desk, etc. All in all it worked out pretty well wouldn't you say?

>the PCs were a throw back and took years to get on track. I'd say the >XT and 286ATs really didn't leave the scene till the early 90s.

I don't know, does anyone else see the suggestion of longevity in that?

>>Hex pads and toggle switches:

>Fabulous debug tool they were. However by 1980 it was more optional >that standard fare on S100 stuff.

All the development systems, even Intels, came with hexadecimal keypads, even into the mid 80's or beyond I guess. But if you talk to engineers, they'll tell you they rarely designed boards that way. I personally didn't learn assembler that way, though lots of anal college courses insisted on doing it that way. I consider it a waste of time.

And with all this talk about how the Intel micros are so terrible, tell me why there's probably still untold thousands of '186's in use out there on sbc's? I can imagine someone will say that it's easy to develop for it, being that you can write code on a stupid pc, but give some credit where it's due at least.

Allison

April 19th, 2005, 01:11 PM

Somebody could stand to spend some time dissecting a dictionary lol.

There infer that typing on a PCkyboard with it's weird keyscan doesn't help. Note many spelling flubs are right chars wrong order.

I'm not familiar with the multibus thing, but just met someone who used to own one. If you read down, you seem to consider the 8086 a kludge, so how is this even relevant? The Victor 9000 packed 1.2 megs onto double density disks, and even the Tandy 2000 used quad density drives. I'm interested to know what you mean by "a reall meg of ram". Light years ahead? I think that's an overstatement.

No surprize to me. Multibus was first used by intel for the MDS them marketed. There were many companies that sold baord and integrated systems using the bus. The 8088/8086 were put on that bus by 1979
and ran at typically 8mhz though faster were available when the chips showed up.

>The S100 crate...

You said it. Many built inside of milk crates lol.

S100 was as rotton a bus as existed but it was popular and offered lots of CPU possibilities.

>The PC didn't invent crap...

No one said it did, but just did everything the others did faster generally and w/less constraints.

Early on that really was slower and with more reboots.

>Actually I abandoned "hex pads and toggle switches" by early 1979...

Lol I was actually kidding about that stuff. I never suspected peeps were actually using them as late as 1979!

Actually it was a diagnostic and development resource that died hard.

>My running joke for years about PCs is the PDP-8 used to do what PC >users wished for. The only difference is size. That and for security I'll >run VAX/VMS against a PC anytime.

What's the point of comparing a micro w/a mini? Remember, a pc will always fit in yer trunk, a mini never will. You have to drag it behind. VMS only runs on big iron.

That was a serious error and vastly incorrect. By 1981 the PDP-8 was
a desktop system in the same realm of the PC. By 1989 the VAX in the form of the MicroVAX was smaller than a PC and loads faster with networking.

And I hope you're not talking about the PDP-8 I saw at the MARCH exhibit at the Trenton Computer Fest this past weekend. Because that would indeed be a very big joke. A fascinating early beast, which I'll bet was unparalleled, and smaller than the alternatives (ahem like the PC). Particularly interesting to people, like me, who have a smattering of assembler under their belt. I especially got a kick out of watching the operator load the boot sequence with the toggle switches, whereby the OS? would subsequently be loaded from the paper or mag tape. But this thing definately was your grandaddy's PDP-8 lol. Weighed in at about 600 or so pounds, I help load it back in the truck, I would know. And I know for a fact a Mindset - Y-A-S-P-C - would run rings around it. Prolly even a Peanut lol. Sorry Mike.

Older nearly pre-70s machine. Slow but then the PC was yet to be by 12 years.

>Actually the breaking point for me to switch from CP/M as a working >environment full time was 1988, The replacement was an 11/23 with
>RD52 (30mb) disk and RSTS, RT11 and RSX-11....

Yada yada yada. Blah blah blah. Whatever lol.

What you dont understand you discard.

> Remember even the newest PC has two buttons in easy reach power >and reset.

Ok...fine. But in reality that's a good thing. I'd hate to have to wind up that crank up everytime to fire up my PDP-?, like an old Model T.

Never had to, Core never forgot on power fail. So craking up often was turning on the power and hitting run. Often if the system was shut down
in a clean state a simple power up was to set the restart address in the switch register, hit load address and then run. That put you right where you stopped last session.

I wasn't suggesting you couldn't get anything done with the alternate platforms. There day was coming to end though. The Apple's graphics were attrocious compared to the pc's. Perhaps a few more colors, and may have had some redemption with hardware sprites like the C64, I'm not sure about that though. I looked at all of these before I bought me first puter. A PC was the only logical choice.

Yes everyone forgets the Commie thing that could make a PC look seriously poor especially for graphics. Had a funny name like Amiga.

TRS-80 means alot of things to alot of people. To some it just meant trash lol. My Tandy 2000 was a TRS-80 - 640 x 480 x 8, with a pallet of 16 colors. The earlier TRS-80's and Coco's had miserable graphics. I happen to have a fondness for both, in particular the Model 3/4, but there was no comparison with any of them.

Yep and some of the most incompatable versions too. Though some were far more creative than big blue.

The COCO was an interesting anomoly as the 6809 was really a bridge cpu between the 6800 8bitter and the high potential 68000.

>The 8088 has got a half-assed method of addressing memory, which for the next 10 years had every programmer who had to deal with their segmented memory model cursing IBM and Intel.

The PDP-8 also used segmented addressing I'm told. The 8088 wasn't the first chip to do that. And remember - no one foresaw software requiring a full meg of memory. The original PC mobo had 64k soldered on, and don't quote me but I think you needed a card to advance it any further. As a side point, the original mobo also had a provision for cassette storage. Those were early days. It's always so easy w/20-20 hindsight to point out what could have been done differently.

Right you are the PDP-8 was nearly a hack for addressing more memory
though it did also do I&D addressing. However for the 8088 to do that many years later show a lack of creativity or plain understanding history.

That's business. And if they had come out with the super duper whatever, it probably would have cost 3x more. The PC was something of a bridge between the 8 and 16 bit worlds. There were loads of stuff for cpm and whatnot still floating around. One or two of their attempts to introduce a pc flopped, so they wisely decided on allowing for the use of earlier os's and apps. That trend still continues, because as you know, compatibility is everything.

We're talking about 2 entirely different things here. You'd rather boot your computer - manually - then pop in a disk? And plug and play wasn't around in 1981, so you're complaining?

And they didn't get it working till when? And even then I find systems
trying to put the soundcard and the NIC at the same int and adddress.

>The only reason I got one in the first place is because clones were so >cheap, and I couldn't afford anything better at the time.

That goes for alot of us. If it wasn't economical enough, we wouldn't have one at home, on our desk, etc. All in all it worked out pretty well wouldn't you say?

Yep, it was cheap. Took a lot of years to get where I was. Good thing I had all that old clanky stuff running to get useful work done.

>the PCs were a throw back and took years to get on track. I'd say the >XT and 286ATs really didn't leave the scene till the early 90s.

I don't know, does anyone else see the suggestion of longevity in that?

If it were then I'd say it was eclipsed by the 30+ years of PDP-8, the better than 30 years of PDP-11 (they are still made!) and the lowly 1802
chips ca 1976 did get the first rover around Mars. Longevity?

All the development systems, even Intels, came with hexadecimal keypads, even into the mid 80's or beyond I guess. But if you talk to engineers, they'll tell you they rarely designed boards that way. I personally didn't learn assembler that way, though lots of anal college courses insisted on doing it that way. I consider it a waste of time.

Err no. Your thinking of SDKs (single board development kits) those crude demonstrators. The development systems were all disk based and had ICE.

And with all this talk about how the Intel micros are so terrible, tell me why there's probably still untold thousands of '186's in use out there on sbc's? I can imagine someone will say that it's easy to develop for it, being that you can write code on a stupid pc, but give some credit where it's due at least.

If they were so great why are they outnumberd by Z80s. Actually of the intel cpus the 8048/9, 8051 class were far more commonly used. Figure there was an 804x and later the 805x in most every car with electronic controls and those that weren't were other vendors cpus. I'm talking millions.

Allison

Allison

patscc

April 19th, 2005, 03:45 PM

Sorry this isn't quoted, doesn't seem to work properly on my browser.

Okay, what BIOS calls are you talking about ? Everything video funneled through call 0x010 on a standard IBM BIOS.

If you're talking about function 0x00C and 0x00D, well, I don't really call the ability to read and set an individual pixel "bit-mapped" graphics. To me, bit-mapped graphics entail some kind of graphics primitives suppoerted by the machine. If to you the ability to read/set a pixel is all you need advertise "bit-mapped" graphics, then perhaps you should have worked for IBM way back when. Perhaps you would have been able to help them turn the series 9000 workstation into a success. Go ahead, google it. See what a piece of crap it was. See how well it sold. It also wasn't compatible with the PC. There's more about it further down.

But, maybe there's there's "secret" BitBlt built into the BIOS that I've been missing all these years. If so, please share it with us. At least tell us what routines for for "producing bitmaps' where actually built into an IBM bios.
And, of course, call 0x010 functions 0x00C and 0x00D weren't available if you had an MDA, which is what a lot of the stock "business" PC's at the time shipped with, since CGA was just plain awful for business use.

Bit planes weren't even available until EGA, (or PGA, possible, I'll grant you that, never had to program for a PGA) and that meant any attempt at drawing graphics usually involved writing in system memory, and then copying to graphics memory with plain 'ol MOV type code.
This is incidently the cause of the infamous CGA 'snow' problem.

If you call this bit-mapped graphics, then perhaps you should investigate how some of the other systems of the time worked.

The MDA and CGA are built around a 6845, really just a CRT controller. It is not a graphics controller, there are no onboard system callable routines that off-load graphics creation from the central CPU. In other words, the CRT controller and IBM BIOS combination of early IBM cards doesn't provide any graphics primitives, whereas other systems of the time did, and they actually went further, they had a GUI available that made use of these.

I don't remember stock VGA as having any graphics primitives, either( I could be wrong ), although at least it offered bit planes, and it's organization made them a lot easier to write to the EGA's bit planes.

Yes, there were lots of aftermarket graphics cards for the PC. There were lots of aftermarket graphics controllers for just about anything out there.

PGA came out 3 years after the PC did, sometime in '84, which means it was probably mid '85 before it was on shelves, and took up two or three slots, I can't remember, which was a lot of space back then.

It was a interesting card, considereing how much real estate took up, and how much it cost. I've never seen on used in a business setting, but then I certainly haven't seen everything. If we're comparing cards at the PGA level, which, if you think about it, was probably IBM's attempt to give the PC an air of "workstation", go look at some of the other workstation-class cards available out there.

I still stand by what I say, that for many years IBM graphics lagged far behind almost everything else out there. Of course, after a while, there wasn't that much else left.

Macs, Amigas, Ataris (the 16-bitters) had tons of graphics primitives built into ROM, which all their OS's used.

Okay, so the early Mac's were a bit slow on the video side.
Of course, the Mac ran a full GUI. Try running a GUI on the PC,XT, or AT, and let's see how fast it is.

Remember, though, once they actually implemented a seperate video card (I think Mac II, but I'm not a Mac expert) this changed.

When the original PC shipped there was only MDA and CGA. Sure, aftermarket cards came out, but look at the cards that were coming out at the same time for Macs and Amigas.

If we start comparing aftermarket cards, especially ultra-high end ones that cost more than the base computer, we'll be here for a while. Perhaps we should go back to the S100- MicroAngelo card as a baseline.
And yes, the stock Apple graphics weren't great. Neither were the PC's. That's why you went out and bought the aftermarket cards you keep talking about.
For instance, one for the Apple II that was available was called the Graphics Tool Kit made by Damco Electronics.
It did 640x768, and to work with the stock monitor, let you have a 'virtual desktop' of 384 in the vertical.
Not bad considering how much longer than the PC the Apple II had been around.

Actually, I'm probably the wrong person to talk to about early personal and workstations. Perhaps some of the users of these early systems could sched some light on this, you know, the people that did desktop publishing, animation, medical imaging, video editing. Let's see how manyof them used PC's for their work.

By the way, if you're talking about the 68000 based beast IBM put out for the scientific market( this was their series 9000 stuff), and ran BASIC, and I think FORTAN, and used a custom mask for the 68000 to mimic the System/370 instuction set, look how many of them were sold, period.
As far as I'm concerned, that was a workstation.
I don't think I bungled it since at the time I had no need for a brick that ran a custom OS and mimicked a s/370.
I don't recall that particular model being much of a success for IBM, since HP and others at the time were also putting out 68000-based workstations, and anyway,
I thougt we were talking about PCs, not workstations.

They also brought out the RT series, but again, I thought we're talking about PC's here.

Oh, and I don't think CP/M 80 runs on a 8088. There was CP/M 86 available, but IBM didn't want to pay for it. Go look it up, it's history, as they say. Apps written for CP/M don't run on DOS, so I'm not real sure what you're talking about when you say '...wisely decided on allowing for the use of earlier os's and apps'
Perhaps a few examples, and then I might be able to remember.

One more thing, what languages did you use ? I'm primarily a C/C++ programmer, with some assembler thrown in( though my 80x86 assembler is rusty these days)
In C, on 80x86 architecture, until 80x86 offered linea-mode addressing, you had to worry about near, far and huge pointers. I don't call this this 'transperant'
One of the burdens of segmented architecture isn't just on the programmer, it's on the actual performance of the machine in question, since shfting your base segment address around all the time takes up instructions, which take up CPU time. Why do you think Intel finally introduced flat access to the 80x86 family.
When PC's finally had more than a meg available, because of the way DOS is stuctured, you couldn't even access it until you either linked in a DPMI memory extender lib or loaded an external one.
I've never met a programmer, at assembler level, or C-level, that actually prefers segemted architecture. But then, I'm sure there's one out there somewhere.

There's also thousands of other microprocessors and microcontrollers out there on SBC's, at least I hope so, it'd be a shame to think I'd been wasting my time reading 'Ciarcia's Circuit Cellar' all those years ago.
Pull out some SBC's and do a head count, and let's see what you find.

patscc

Allison

April 20th, 2005, 04:24 AM

Sorry this isn't quoted, doesn't seem to work properly on my browser.

Is ok, no problem. A couple of points.

If we start comparing aftermarket cards, especially ultra-high end ones that cost more than the base computer, we'll be here for a while. Perhaps we should go back to the S100- MicroAngelo card as a baseline.

Actually for S100 that was one of the late arrivals. Early arrivals were comemco dazzler and that was 1977. Better Color graphics that the PC had till CGA.

Actually, I'm probably the wrong person to talk to about early personal and workstations. Perhaps some of the users of these early systems could sched some light on this, you know, the people that did desktop publishing, animation, medical imaging, video editing. Let's see how manyof them used PC's for their work.

Actually the PC was a mid entry into the real "workstation wars". Then
the metric for a workstation was 3M, Megapixals, Megabytes and MIPS.
For those that dont remember it was about high resolution screens(1024x1024 or better), Memory in the several megabyte range
and processors that were in the multi-MIPS range. Names like Sun, Apollo, DEC, Stratus and Even IBM were in there but the PC would have to wait till the 386+FPU got fast enough.

By the way, if you're talking about the 68000 based beast IBM put out for the scientific market( this was their series 9000 stuff), and ran BASIC, and I think FORTAN, and used a custom mask for the 68000 to mimic the System/370 instuction set, look how many of them were sold, period.
As far as I'm concerned, that was a workstation. I don't think I bungled it since at the time I had no need for a brick that ran a custom OS and mimicked a s/370. I don't recall that particular model being much of a success for IBM, since HP and others at the time were also putting out 68000-based workstations, and anyway, I thougt we were talking about PCs, not workstations.

Spectacular failure is the best word for the RT, By time Motorola got the mask right everyone lost interest.

Oh, and I don't think CP/M 80 runs on a 8088. There was CP/M 86 available, but IBM didn't want to pay for it. Go look it up, it's history, as they say. Apps written for CP/M don't run on DOS, so I'm not real sure what you're talking about when you say '...wisely decided on allowing for the use of earlier os's and apps' Perhaps a few examples, and then I might be able to remember.

Lessee I ran Dbase-II for CP/M-86 (aka concurrent) we used to run it under DOS and later under win98 off a small drive{less than 500mb}.
The small disk thing was because under FAT-16 you still could use FCBs
and FCBs were needed for file handling under CP/M. FCBs disappeared under FAT32. And FAT32 is something like 10years old?

One more thing, what languages did you use ? I'm primarily a C/C++ programmer, with some assembler thrown in( though my 80x86 assembler is rusty these days)In C, on 80x86 architecture, until 80x86 offered linea-mode addressing, you had to worry about near, far and huge pointers. I don't call this this 'transperant'. One of the burdens of segmented architecture isn't just on the programmer, it's on the actual performance of the machine in question, since shfting your base segment address around all the time takes up instructions, which take up CPU time. Why do you think Intel finally introduced flat access to the 80x86 family.
When PC's finally had more than a meg available, because of the way DOS is stuctured, you couldn't even access it until you either linked in a DPMI memory extender lib or loaded an external one.
I've never met a programmer, at assembler level, or C-level, that actually prefers segemted architecture. But then, I'm sure there's one out there somewhere.

Got that right. segments were a programming nightmare in everything but BASIC. There it was not a problems as programs that size didn't happen.

There's also thousands of other microprocessors and microcontrollers out there on SBC's, at least I hope so, it'd be a shame to think I'd been wasting my time reading 'Ciarcia's Circuit Cellar' all those years ago.
Pull out some SBC's and do a head count, and let's see what you find.

The whole micromputer show started when Intel tried [successfully] to
generalize the calculator chip to something programmable. That was the
start with the 4004, then 8008 then 8080 in the space of three years. But,
they were not alone and by 1978 there were more cpus from more vendors than I can completely name. It has grown from there.

Now you can by a programmable single cpu thats fast enough to do slow signal processing, with as much or as little IO as needed including USB and
cheap enough to put in toys. For every X86 sold there must be dozens of
PIC and similar micros on a chip stuffed in talking dolls, microwave ovens, coffee pots, smart chargers for emergency equipment, Tube and LCD monitors, TV sets, car radios, and the list goes on.

I'm sorry for our PC centric friend as PCs weren't the center of the world, they were a result of the pressures that had already seen in other parts of computing. The PC itself was the commercial fall out from that.

Allison

Chris2005

April 20th, 2005, 02:14 PM

>There infer that typing on a PCkyboard with it's weird keyscan doesn't >help. Note many spelling flubs are right chars wrong order.

I guess I'll have to discard that, because it don't make a lick of sense. I'm typing on a PCkyboard too you know. But that's hitting below the belt I'm sure. There's really no point expecting an IT pro to know how to type lol.

>No surprize to me. Multibus was first used by intel for the MDS them >marketed. >There were many companies that sold baord and integrated >systems using the >bus. The 8088/8086 were put on that bus by 1979
>and ran at typically 8mhz though faster were available when the chips >showed up.

I would like to find one though. In case you haven't noticed, I'm into that sort of thing! :)

>S100 was as rotton a bus as existed but it was popular and offered lots of CPU >possibilities.

No doubt. As stupid as they were though, they were probably quite useful. ;)

>
>>The PC didn't invent crap...

>No one said it did, but just did everything the others did faster generally and w/less >constraints.
>

>Early on that really was slower and with more reboots.

Yeah but they were cheap and even punks like me could buy one if they gave up a whopper now and again. Actually I had to wait until the T2K went on clearance. That stupid p.o.s. was pretty pricey early on.

>
>>Actually I abandoned "hex pads and toggle switches" by early 1979...

>Lol I was actually kidding about that stuff. I never suspected peeps were >actually using them as late as 1979!
>

>Actually it was a diagnostic and development resource that died hard.

Yeah, took way way way too long huh?

>That was a serious error and vastly incorrect. By 1981 the PDP-8 was
>a desktop system in the same realm of the PC. By 1989 the VAX in the ?>form of the MicroVAX was smaller than a PC and loads faster with >networking.

Ok, I was a bit off, I for one have no problem admitting that. But I'd like to know what the desktop PDP-8 went for. And DEC didn't waste any time coming out with a stupid pc. Problem is you're always giving 'em away. Hold on to one...who knows, maybe you'll fall in love.

>
>And I hope you're not talking about the PDP-8 I saw at the MARCH >exhibit
>Older nearly pre-70s machine. Slow but then the PC was yet to be by 12 >years.

Not a moment too soon though.

>What you dont understand you discard.

If it were possible, I could ramble on about things you never heard of either. From the sound of it though, you've done it all Allison. I'm in deep water here. Help!

>Never had to, Core never forgot on power fail. So craking up often was >turning on the power and hitting run. Often if the system was shut down
>in a clean state a simple power up was to set the restart address in the >switch register, hit load address and then run. That put you right where >you stopped last session.

Whatever lol.

>Yes everyone forgets the Commie thing that could make a PC look >seriously poor especially for graphics. Had a funny name like Amiga.

Somewhere in my response I touched on the Amiga. Came out years after the pc. The 1000 kind of flopped, imagine that. The Amiga probably found it's ancestry in the Mindset, which flopped too, it wasn't it's time. And personally I have an aversions to computers, even in my collecting, that have alot of custom ic's. Too hard to get replacements, and too expensive initially. The whole mobo would have to be replaced. But my Tandy *RIP* I could get back up and running after a memory parity error, or blown capacitor in the power supply - for between 20 cents and 2 dollars. And all of it's ic's were more or less commonly available.

>[quote=Chris2005]
>TRS-80 means alot of things to alot of people. To some it just meant trash lol. My >Tandy 2000 was a TRS-80 - 640 x 480 x 8, with a pallet of 16 colors. The earlier >TRS-80's and Coco's had miserable graphics. I happen to have a fondness for both, >in particular the Model 3/4, but there was no comparison with any of them.
>

>Yep and some of the most incompatable versions too. Though some were far more >creative than big blue.

Yep, it's easy to improve on an original. But compatibility is everything. Funny all those aftermarket graphic cards didn't last long. Must have been expensive to support them all. Come to think of it, economics is everything.

>The COCO was an interesting anomoly as the 6809 was really a bridge cpu between >the 6800 8bitter and the high potential 68000.

Coco's are fun. And cheap if you want to play with the innards. But the original point was that the graphics were pretty lousy, no contest between it and even the cga.

>Right you are the PDP-8 was nearly a hack for addressing more memory
>though it did also do I&D addressing. However for the 8088 to do that many years >later show a lack of creativity or plain understanding history.

Well the 8086/88 came out in 1979. And you said yourself that the only chip that was truly linear back in the day was the 68000.

>Yep, it was cheap. Took a lot of years to get where I was. Good thing I had >all that old clanky stuff running to get useful work done.

Even my highly incompatible 2000 had all the basics, and then some. And 896k of ram to run it in.

>
>the PCs were a throw back and took years to get on track. I'd say the >XT and >286ATs really didn't leave the scene till the early 90s.

>I don't know, does anyone else see the suggestion of longevity in that?
>

>If it were then I'd say it was eclipsed by the 30+ years of PDP-8, the better than 30 >years of PDP-11 (they are still made!) and the lowly 1802
>chips ca 1976 did get the first rover around Mars. Longevity?

Ok, and I saw a Pro 350 (350 Pro?), which had the PDP-11, at the fest last weekend. The guy wanted 5 bucks for it. I should have grabbed it right then, but figured who else was going to buy that p.o.s. lol, and someone else did. Provided it worked, where would I get any software for it? Or parts? Yet if I had found a pc, and there wasn't a one strangely, I could pull down apps off the net all day long. Longevity means more than just being able to turn something on. I saw one guy scarffing up early copies of Lotus 1-2-3, and would run them on a Pentium, because he felt he could get the job done easier that way.

>
>All the development systems, even Intels, came with hexadecimal keypads, even >into the mid 80's or beyond I guess. But if you talk to engineers, they'll tell you they >rarely designed boards that way. I personally didn't learn assembler that way, >though lots of anal college courses insisted on doing it that way. I consider it a >waste of time.
>

>Err no. Your thinking of SDKs (single board development kits) those crude >demonstrators. The development systems were all disk based and had ICE.

"SDK" usually mean software development kit I thought. But there's no point in arguing that. The point was that keypads are a tad prehistoric - very collectible though - and a tad anal lol.

>If they were so great why are they outnumberd by Z80s. Actually of the intel cpus >the 8048/9, 8051 class were far more commonly used. Figure there was an 804x >and later the 805x in most every car with electronic controls and those that weren't >were other vendors cpus. I'm talking millions.

Maybe I'm a bit over my head here, but this I do know, the 8048 series was used in most pc keyboards. And was an Intel. If more of them were used then '186/'188's, it must have been because it would have been overkill. The '48 had very very little on board ram and rom. I had thought that auto's used their own chips, but I'd be happy to find out that it did use stock ic's so if one blew, I could replace it. The 186 was used ALOT, that's all I know. Strangely in precious few desktop puters, the T2K being one of them, although the 186 was supposed to drastically reduce the chip count, and yet there's like double the amount as in the original PC.

>Okay, what BIOS calls are you talking about ? Everything video funneled through >call 0x010 on a standard IBM BIOS.

>If you're talking about function 0x00C and 0x00D, well, I don't really call the ability to >read and set an individual pixel "bit-mapped" graphics. To me, bit-mapped graphics >entail some kind of graphics primitives suppoerted by the machine. If to you the >ability to read/set a pixel is all you need advertise "bit-mapped" graphics, then >perhaps you should have worked for IBM way back when. Perhaps you would have >been able to help them turn the series 9000 workstation into a success. Go ahead, >google it. See what a piece of crap it was. See how well it sold. It also wasn't >compatible with the PC. There's more about it further down.

I thought bit mapped graphics referred to the ability to turn each itty bitty pixel on and off, and change it's color if present. Goes hand in hand with the term all points addressable. It's just that. As far as I know, the "primitives" you're referring to weren't built into the machines, but came as kits. Set me strait if you know better. I have a package for the NEC APC III called GSX or something, and it's disked based. Was there that much demand, just to make a few pie charts? Your original words indicated that bit mapped graphics weren't possible, unless presumably you programmed the hardware directly. This simply isn't true. Yeah int 10h or whatever, it's been a while. And let's not talk about the Amiga again (...and again). It came out years later.
You were compaining that they didn't build THE PC w/a 68000. IBM had a couple of earliy flops, and they did make a 68000 based unit. That's all I was saying. I know it was a flop, but that's just the point. Putting one in a computer didn't make a success. And now pc compatibility is a good thing? I think we're going around in circles.

>But, maybe there's there's "secret" BitBlt built into the BIOS that I've been missing >ll these years. If so, please share it with us. At least tell us what routines for for >"producing bitmaps' where actually built into an IBM bios.
>And, of course, call 0x010 functions 0x00C and 0x00D weren't available if you had >an MDA, which is what a lot of the stock "business" PC's at the time shipped with, >since CGA was just plain awful for business use.

Right, whatever. NO you absolutely can't produce graphics on with a card that doesn't support graphics. Irrelevent point. There were options. Many. The original pc was like $3-5000, and many of the advanced options were pricey too. The people/businesses that had the need and the money bought them. And how much did you want to be built into 8k of rom???

>Bit planes weren't even available until EGA, (or PGA, possible, I'll grant you that, >never had to program for a PGA) and that meant any attempt at drawing graphics >usually involved writing in system memory, and then copying to graphics memory >with plain 'ol MOV type code.
>This is incidently the cause of the infamous CGA 'snow' problem.

In all of this, all I keep hearing is IBM, IBM, o God, IBM. There were many stupid pc's remember. And the truth is, I've been told, alot of people didn't understand how to use the bios subroutines, yet this amazes me, because for simple tasks they were simple to use. And the reality is most commercial software didn't use them, because there were slow (The same deal with the C64 and other computers too I'm sure). The main reason why all the "incompatibles" were just that. If software was "well behaved", most everything would run on these machines. It was an absolute no no to do anything of the sort on a Macintosh. Everthing had to use the API's that Apple provided. One of the reasons Macs were slow.
The problem of "snow" was circumventable. You know that.

>It was a interesting card, considereing how much real estate took up, and how much >it cost. I've never seen on used in a business setting, but then I certainly haven't >seen everything. If we're comparing cards at the PGA level, which, if you think about >it, was probably IBM's attempt to give the PC an air of "workstation", go look at >some of the other workstation-class cards available out there.

The EGA also came out in '84. The PGA was loads more powerful, yes took up 2 slots, was in effect a seperate onboard computer, with it's own 8088. Vermont Microsystems had a similar board (x2), with an 80186. But the point is if you needed the power, you had to pay for it. All the high end cad/cam stuff was pricey, regardless of who made it. Tens of thousands of dollars. You needed drivers and support for that stuff too. It's awful stupid to spend money on systems that would be obsolete in a year or two.

>I still stand by what I say, that for many years IBM graphics lagged far behind >almost everything else out there. Of course, after a while, there wasn't that much >else left.

>Macs, Amigas, Ataris (the 16-bitters) had tons of graphics primitives built into ROM, >which all their OS's used.

They all came out years later too, and each had their issues. Each found ia small niche. For some reason I liked the Atari's especially, but it's graphics capability, in terms of resolution, was considerably less than a pc with something like a Sigma 400 (the T2K's were better also, considerably). Many of the clones, lets not forget about them, early on had high resolutions. You didn't have to buy an IBM, but that's all I'm hearing. Many of the semi's were superior. Way better graphics, speed, everything. But people weren't inclined to buy them, alot of times not even full compatibles that didn't have the IBM moniker.

>Okay, so the early Mac's were a bit slow on the video side.
>Of course, the Mac ran a full GUI. Try running a GUI on the PC,XT, or AT, and let's >see how fast it is.

But the mac didn't even have a dedicated CRT controller. And towards the end of 1985 it was $2500. One of those things you'd love to have around, stylish as all hell, but had very limited internal expandability, came with 128k initially, and had to be upgraded to the, look out, Fat Mac just to get to 512k. A very interesting yuppy toy early on, I bought one in the early 90's for 50 bucks, had a ball with it.

>Remember, though, once they actually implemented a seperate video card (I think >Mac II, but I'm not a Mac expert) this changed.

1987, maybe '88. The SE's could accept some sort of card too, not sure about color, but then you'd get a little black and white mini-mi sitting next to a giant color monitor forever.

>When the original PC shipped there was only MDA and CGA. Sure, aftermarket >cards came out, but look at the cards that were coming out at the same time for >Macs and Amigas.

No. Years later. You could buy quite respectable card for pc's by probably 82-83.

The other things excelled in the graphics arena. But it depends on what you were doing. Again, the Amiga used interlaced video in the higher modes. The color macs were expensive, and the stock Apple color card was no better than vga.

>For instance, one for the Apple II that was available was called the Graphics Tool Kit >made by Damco Electronics.
>It did 640x768, and to work with the stock monitor, let you have a 'virtual desktop' of >384 in the vertical.
>Not bad considering how much longer than the PC the Apple II had been around.

I'm not familiar with that card, but if the thing wasn't operating at the monitor's native resolution, it must have had to interleave the video. Not too kind on the eyes. And how many colors could it produce?

>Actually, I'm probably the wrong person to talk to about early personal and >workstations. Perhaps some of the users of these early systems could sched some >light on this, you know, the people that did desktop publishing, animation, medical >imaging, video editing. Let's see how manyof them used PC's for their work.

There were specialized workstations at the time. No one said the pc's were an answer for everything. Tektronix made some. Dazzling. Expensive too. If you needed it, you had to pay.

>By the way, if you're talking about the 68000 based beast IBM put out for the >scientific market( this was their series 9000 stuff), and ran BASIC, and I think >FORTAN, and used a custom mask for the 68000 to mimic the System/370 >instuction set, look how many of them were sold, period.
>As far as I'm concerned, that was a workstation.
>I don't think I bungled it since at the time I had no need for a brick that ran a custom >OS and mimicked a s/370.
>I don't recall that particular model being much of a success for IBM, since HP and >others at the time were also putting out 68000-based workstations, and anyway,
>I thougt we were talking about PCs, not workstations.

You went on about how the pc's didn't use the 68000. Well, they did it. And you didn't even buy one, so where's the argument? You bought a pc you said.

>They also brought out the RT series, but again, I thought we're talking about PC's >here.

I don't know anything anymore. My head hurts lol. The point is if the pc was such a piece of crap, what are you comparing it to? Most of us bought one because it was the most logical, economical choice.

>Oh, and I don't think CP/M 80 runs on a 8088. There was CP/M 86 available, but >IBM didn't want to pay for it. Go look it up, it's history, as they say. Apps written for >CP/M don't run on DOS, so I'm not real sure what you're talking about when you say >...wisely decided on allowing for the use of earlier os's and apps'
>erhaps a few examples, and then I might be able to remember.

There were other OS's availabe for the pc's. Including the USCD p-system previously mentioned. Dos was the most popular. There was some degree of compatibility with earlier stuff, this I remember reading.

>One more thing, what languages did you use ? I'm primarily a C/C++ programmer, >ith some assembler thrown in( though my 80x86 assembler is rusty these days)
>n C, on 80x86 architecture, until 80x86 offered linea-mode addressing, you had to >orry about near, far and huge pointers. I don't call this this 'transperant'
>ne of the burdens of segmented architecture isn't just on the programmer, it's on the >ctual performance of the machine in question, since shfting your base segment >ddress around all the time takes up instructions, which take up CPU time. Why do >ou think Intel finally introduced flat access to the 80x86 family.
>hen PC's finally had more than a meg available, because of the way DOS is >tuctured, you couldn't even access it until you either linked in a DPMI memory >xtender lib or loaded an external one.
>'ve never met a programmer, at assembler level, or C-level, that actually prefers >segemted architecture. But then, I'm sure there's one out there somewhere.

I'm not even a programmer. I do know a little C, and do know that it's a far cry from programming in assembler. You spend alot of time talking about how difficult to program an Intel in assembler, as if you and everyone else were doing it everyday. No, no one is going to prefer a segmented architecture. Such is life. One platform dominated and you had to take the good with the bad.

>There's also thousands of other microprocessors and microcontrollers out there on >SBC's, at least I hope so, it'd be a shame to think I'd been wasting my time reading >'Ciarcia's Circuit Cellar' all those years ago.
>Pull out some SBC's and do a head count, and let's see what you find.

S. Ciarcia published an article for a sbc based on the 8088 in late '82 lol. Of course there were others. But loads of 80186/88 sbc's are still out there. I can't count that high.

>Now you can by a programmable single cpu thats fast enough to do slow >signal processing, with as much or as little IO as needed including USB >and cheap enough to put in toys. For every X86 sold there must be dozens >of PIC and similar micros on a chip stuffed in talking dolls, microwave >ovens, coffee pots, smart chargers for emergency equipment, Tube and >LCD monitors, TV sets, car radios, and the list goes on.

Pics came out years later. Intels don't find themselves in as many sb apps as in the past, but we were talking about the 80's (and 90's) remember. It evidently became very cheap and easy to build a small, single purpose chip *blob* to fill a calculator or a hand held chess game.

>I'm sorry for our PC centric friend as PCs weren't the center of the world, >they were a result of the pressures that had already seen in other parts of >computing. The PC itself was the commercial fall out from that.

I don't know, I just bought a old Sparc at the show this past weekend to play with. I still have my C64, and an sx-64, and plan on firing them up one of these days. I have pieces and parts of macs lying about. If you're talking "pc centric" with regard to everyday usage, who isn't? Remember, this is a collectors forum. I don't give 2 shits what everyone else finds interesting - with regard to my preferences - I'll collect what's interesting to me. Yet I'll show there stuff the respect it's due. You two can go on and on, and I'm sure you will, about how they were these awful miserable pieces of crap. Had their problems, no arguments. But me and other people owned one as our first puter, and for that reason we find it - collectilble. And the next time I run into a pdp-whatever, I'm going to pay the dude, then stomp on it! lmao

Allison

April 20th, 2005, 03:44 PM

>

I guess I'll have to discard that, because it don't make a lick of sense. I'm typing on a PCkyboard too you know. But that's hitting below the belt I'm sure. There's really no point expecting an IT pro to know how to type lol.

;) Yep, used to be able to type then using crappy PC keyboards killed
the wrists.

I would like to find one though. In case you haven't noticed, I'm into that sort of thing! :)

Seriously? worked with quite a few. I have a bunch in the box that I'm not using.

[quote="Chris2005"]
Yeah but they were cheap and even punks like me could buy one if they gave up a whopper now and again. Actually I had to wait until the T2K went on clearance. That stupid p.o.s. was pretty pricey early on.

Later on that was true, in 82/83 that was not true. PC did create the idea of toilet tissue computers, use and toss.

Ok, I was a bit off, I for one have no problem admitting that. But I'd like to know what the desktop PDP-8 went for. And DEC didn't waste any time coming out with a stupid pc. Problem is you're always giving 'em away. Hold on to one...who knows, maybe you'll fall in love.

Actually their first error was the DECmate, then Robin(Vt180) and then Rainbow. of all the rainbow was by far the most interesting. Never thought badly of it. I don't collect them mostly due to space as they
otherwise do interest me. I fix them and pass them to good homes.

If it were possible, I could ramble on about things you never heard of either. From the sound of it though, you've done it all Allison. I'm in deep water here. Help!

Quite possible on both counts.

Even my highly incompatible 2000 had all the basics, and then some. And 896k of ram to run it in.

Tandy and DEC the few that decided than 640k was not enough. Where be they now?

Ok, and I saw a Pro 350 (350 Pro?), which had the PDP-11, at the fest last weekend. The guy wanted 5 bucks for it. I should have grabbed it right then, but figured who else was going to buy that p.o.s. lol, and someone else did. Provided it worked, where would I get any software for it? Or parts? Yet if I had found a pc, and there wasn't a one strangely, I could pull down apps off the net all day long. Longevity means more than just being able to turn something on. I saw one guy scarffing up early copies of Lotus 1-2-3, and would run them on a Pentium, because he felt he could get the job done easier that way.

Apples and pigs. However the pro was just another marketing disaster
though the Pro380 had the J-11 cpu and with a couple of megs of ram and a NIC it's was a decent DEC biased workstation.

"SDK" usually mean software development kit I thought. But there's no point in arguing that. The point was that keypads are a tad prehistoric - very collectible though - and a tad anal lol.

Collectable yes. I have a collection of SBCs and SDKs. Software development hardly. Most of them were so poor for ram as to be
useless.

Maybe I'm a bit over my head here, but this I do know, the 8048 series was used in most pc keyboards. And was an Intel. If more of them were used then '186/'188's, it must have been because it would have been overkill. The '48 had very very little on board ram and rom. I had thought that auto's used their own chips, but I'd be happy to find out that it did use stock ic's so if one blew, I could replace it. The 186 was used ALOT, that's all I know. Strangely in precious few desktop puters, the T2K being one of them, although the 186 was supposed to drastically reduce the chip count, and yet there's like double the amount as in the original PC.

Yep, 8048 was in PC keyboards, 8042 is the other side one the motherboard. Most of the early car cpus were 8048/9 but extended temp range as they had to work at -40 to +80. Then it was the 80860 and friends.

No. Years later. You could buy quite respectable card for pc's by probably 82-83.

For S100 maybe. Based on the Byte and other mags I still have for the era your about 1.5 and some cases 2 years earlier than intro or availability.

>For instance, one for the Apple II that was available was called the Graphics Tool Kit >made by Damco Electronics.
>It did 640x768, and to work with the stock monitor, let you have a 'virtual desktop' of >384 in the vertical.
>Not bad considering how much longer than the PC the Apple II had been around.

That capability was featured in September 1982 Byte. One problem then was finding a monitor that could handle 640x480 or better!

In the middle of the article there's an IBM ad with Charlie C kicking his cane. Danged if a 360k floppy wasn't an option by then. In '82 I was running losely coupled Z80s and 5mb hard disks.

I'm not even a programmer. I do know a little C, and do know that it's a far cry from programming in assembler. You spend alot of time talking about how difficult to program an Intel in assembler, as if you and everyone else were doing it everyday. No, no one is going to prefer a segmented architecture. Such is life. One platform dominated and you had to take the good with the bad.

C is nice. It's close to the metal but not too close. That is until you run
into an application that starts pressing addrssing large arrays. Then you pray to the compiler dogs to hopefully manage it for you. If not you get
real close to the metal again.

Pics came out years later. Intels don't find themselves in as many sb apps as in the past, but we were talking about the 80's (and 90's) remember. It evidently became very cheap and easy to build a small, single purpose chip *blob* to fill a calculator or a hand held chess game.

PICs are a revisit of the GI CP1600 devices. they are old.

As to calcs, yes they were easy and cheap. However there was a growing market for the "programable calcs" and there were few chips for that.

I don't know, I just bought a old Sparc at the show this past weekend to play with. I still have my C64, and an sx-64, and plan on firing them up one of these days. I have pieces and parts of macs lying about. If you're talking "pc centric" with regard to everyday usage, who isn't? Remember, this is a collectors forum. I don't give 2 shits what everyone else finds interesting - with regard to my preferences - I'll collect what's interesting to me. Yet I'll show there stuff the respect it's due. You two can go on and on, and I'm sure you will, about how they were these awful miserable pieces of crap. Had their problems, no arguments. But me and other people owned one as our first puter, and for that reason we find it - collectilble. And the next time I run into a pdp-whatever, I'm going to pay the dude, then stomp on it! lmao

Me it wasn't my first, not even close by 10 years. The result is I had a base to compare it to and was disappointed. I expected better knowing the technology. Are the pre 386s collectable, your bet. Some were pretty interesting and unique.

It's your money, waste it how you may. But while your doing that I'll be scoping something worth putting effort into to understand.

Allison

Chris2005

April 20th, 2005, 04:30 PM

>Seriously? worked with quite a few. I have a bunch in the box that I'm >not using.

I'll have to do some research. I don't suppose you have pics? I know you just want to find out where I live anyway ;). I'll ask my friend Sam, he's giving me some Epson thingees wif both '88's and z80's or 8080's...not sure.

>Actually their first error was the DECmate, then Robin(Vt180) and then >Rainbow. of all the rainbow was by far the most interesting. Never >thought badly of it. I don't collect them mostly due to space as they
>otherwise do interest me. I fix them and pass them to good homes.

Computers like the Rainbow created alot of solidarity though back in the day (i.e. user groups). And others like it...TI Pro, Victor 9000, and the T2K even had a newsletter called the Whimper. The dude that owned the PDP-8/e (oops) ACTUALLY OWNS THIS RARE RARE COMPATIBILITY CARD FOR THE TI PC THAT I SWORE MUST HAVE BEEN VAPORWARE! BUT...NOT FOR SALE!!!

>Tandy and DEC the few that decided than 640k was not enough. Where >be they now?

Muh Tandy is in muh garage :).

>Apples and pigs. However the pro was just another marketing disaster
>though the Pro380 had the J-11 cpu and with a couple of megs of ram >and a NIC it's was a decent DEC biased workstation.

That thing was remarkably similar in appearance to the Rainbow, as is the Decmate II (a feller wants to send me one). It would have been worth buying it for the drives if nothing else.

>Yep, 8048 was in PC keyboards, 8042 is the other side one the >motherboard. Most of the early car cpus were 8048/9 but extended >temp range as they had to work at -40 to +80. Then it was the 80860 >and friends.

I think only AT's had the '42 on the motherboard. Something else handled it on the PC/XT's...8255 or something.

>For S100 maybe. Based on the Byte and other mags I still have for the >era your about 1.5 and some cases 2 years earlier than intro or >availability.

Omg omg omg. I have Bytes from '83. I guess I'd have to go through them again. But then again maybe not...

>PICs are a revisit of the GI CP1600 devices. they are old.

I don't know what those are. But PIC apps were popular starting maybe 10 years ago.

>As to calcs, yes they were easy and cheap. However there was a >growing market for the "programable calcs" and there were few chips >for that.

I want a HP49g+, but the 49g was prettier. Stupid calc teacher made me buy a stupid TI86 for one freaking problem on the final exam! I could have done it in my head!

>Me it wasn't my first, not even close by 10 years. The result is I had a >base to compare it to and was disappointed. I expected better knowing >the technology. Are the pre 386s collectable, your bet. Some were >pretty interesting and unique.

Disappointing maybe. Economical, definately. The reality was that the Amiga's, Atari's, etc. disappointed me. I looked at them all. My first puter was actually the T1000, but it seemed buggy to me, and I returned it Their monitors sucked too. I liked the graphics on the 2000, though most people would argue it was one of the stupidest