Re: [9fans] /sys/include/ip.h 5c(1)

Re: [9fans] /sys/include/ip.h 5c(1)

[erik quanstrom]

> lucio@proxima.alt.za wrote:
> >>> but by 1990 with microchannel &c. things were much more closed off.
> >> i thought only one company ever really made microchannel,
> >> and even they weren't terribly in earnest in the end,
> >> except on non-PC things like RS6000.
> >
> > IBM tried to recover control over the PC market by introducing MCA,
> > bargaining that public sentiment would swing in their favour.
>
> They might have had that in mind as a secondary reason - but I doubt even that.
>

wikipedia agrees with lucio on this point
http://en.wikipedia.org/wiki/Micro_Channel_architecture#Marketshare_issues

> The majority within IBM never wanted into that part of the market in the first
> place, as it was seen as cannibalizing not only 3XXX terminal sales, but the
> entire, highly profitable, big-iron+interface+network+services infrastructure
> behind said terminals.

do you have a reference for this?  i worked at a company around 1990
that was heavily into ibm mainframes.  (so much so, that they
sold PROFS to ibm.)  we all had 3270 terminals, and if you were lucky,
you had a pc.  email, calandaring, all that great stuff was done centrally
1500 miles away on ibm mainframes.  the pc could do none of the
criticial functions that the mainframe system could perform.  we didn't
have networking for the pc.  heck, there was only one machine fat enough
to run windows 3.1, which didn't even do networking.

so even 3 years after the release of microchannel, we would never
have considered pcs as 3270 replacements.  i don't remember any
machines that could have even run 3270 emulators, if they existed.
perhaps we were the wiredest ibm site ever, but i think not.  and
judging from what i saw, the mca guys would have wasted time
thinking about 3270 emulators.

ah, the summer of broken arrows.  good times.

- erik

Re: [9fans] /sys/include/ip.h 5c(1) LONG POST

[W B Hacker]

erik quanstrom wrote:
>> lucio@proxima.alt.za wrote:
>>>>> but by 1990 with microchannel &c. things were much more closed off.
>>>> i thought only one company ever really made microchannel,
>>>> and even they weren't terribly in earnest in the end,
>>>> except on non-PC things like RS6000.
>>> IBM tried to recover control over the PC market by introducing MCA,
>>> bargaining that public sentiment would swing in their favour.
>> They might have had that in mind as a secondary reason - but I doubt even that.
>>
>
> wikipedia agrees with lucio on this point
> http://en.wikipedia.org/wiki/Micro_Channel_architecture#Marketshare_issues

'Wikipedia' does not always compute.

That article ass u me s that the MCA was needed ONLY for PC's.

Wrong answer as far as IBM goes.

>
>> The majority within IBM never wanted into that part of the market in the first
>> place, as it was seen as cannibalizing not only 3XXX terminal sales, but the
>> entire, highly profitable, big-iron+interface+network+services infrastructure
>> behind said terminals.
>
> do you have a reference for this?

Probably an hpfs-386 disk or three full - but half a century on [1] from my
first IBM 701 run, and with Irish Alzheimer's and senile dementia don't expect
me to *find* them.

> i worked at a company around 1990
> that was heavily into ibm mainframes.  (so much so, that they
> sold PROFS to ibm.)  we all had 3270 terminals, and if you were lucky,
> you had a pc.  email, calandaring, all that great stuff was done centrally
> 1500 miles away on ibm mainframes.  the pc could do none of the
> criticial functions that the mainframe system could perform.  we didn't
> have networking for the pc.  heck, there was only one machine fat enough
> to run windows 3.1, which didn't even do networking.

You waz bein' robbed.

The secret to high 'PC" peformance as at 1990-94?

Fast private network for the WAN. No 'Weendows'

100 MBps TCNS for the LAN. No 'Weendows'

Hercules monochrome graphics and DRDOS, else ATI SVGA and OS/2 2.11. No 'Weendows'

Did I forget anything? Oh ..

** NO effing 'Weendows' ** No way. No how. No where.

>
> so even 3 years after the release of microchannel, we would never
> have considered pcs as 3270 replacements.  i don't remember any
> machines that could have even run 3270 emulators, if they existed.

A year-one IBM 'PC-1' 8-bit ISA could run 3270 emu just fine - part of what it
needed was in the BIOS, and the onboard 64K RAM was enough for block-mode
buffers. Just needed a NIC appropriate to the local concentrator, optionally a
keyboard swap. Ditto OS/2.

3270 emulation only got 'difficult' if you wanted to run it on *Windows*.

> perhaps we were the wiredest ibm site ever, but i think not.  and
> judging from what i saw, the mca guys would have wasted time
> thinking about 3270 emulators.
>

They didn't have to.

MCA-bus machines could emulate the central 370 itself - and anything earlier -
that those terminals once connected to. 308X and newer mainframes were another
matter.

> ah, the summer of broken arrows.  good times.
>
> - erik
>

Yah - well... I can't edit plaintext files remotely any faster today (Joe, Pico,
Nano, Mined) over cable modem ssh internet than I could (BRIEF) over dedicated
56 Kbps fifteen years ago, so 'progress' has been eaten by TCP/IP overhead,
Ethernet overhead (world's second worst protocol), congestion, throttling,
packet-loss ..and .... GUI's.

Died in the wool Plan 9 guys are no doubt ROFL by now at that last part...

;-)

Bill

[1] In order of first use, IBM 701, WECO M33, Burroughs AN/GSA-51, IBM/MIT
Whirlwind II, (AKA MITRE AN/FSQ-7) ....

Re: [9fans] /sys/include/ip.h 5c(1)

[lucio]

> wikipedia agrees with lucio on this point
> http://en.wikipedia.org/wiki/Micro_Channel_architecture#Marketshare_issues
>
>> The majority within IBM never wanted into that part of the market in the first
>> place, as it was seen as cannibalizing not only 3XXX terminal sales, but the
>> entire, highly profitable, big-iron+interface+network+services infrastructure
>> behind said terminals.
>
> do you have a reference for this?

There's truth on both sides, IBM had committed to the PC because Apple
was stealing the show, but within IBM there were definitely movements
keeping the PC at bay.  My understanding was that the 8250, crappy
UART that it was, was used specifically because SDLC required
synchronous RS-232 and the 8250 didn't have it.  Note that the 8251
was compatible with the 8088 (both Intel designs, if I'm not
mistaken), where the 8250 came from National Semiconductors and
required additional glue logic (and had write-only registers and no
RESET, shudder!).

Fact is, IBM had distinct, sometimes contrasting marketing objectives.
I suspect that fighting the Taiwanese menace was as high on the agenda
as anything could possibly get.  In "Big Blues" (I think that is the
book title) it is suggested that IBM did not have a proper focus and
the PC was a knee-jerk reaction that should have been planned
considerably better.

++L

Re: [9fans] /sys/include/ip.h 5c(1) - LONG POST

[W B Hacker]

lucio@proxima.alt.za wrote:
>> wikipedia agrees with lucio on this point
>> http://en.wikipedia.org/wiki/Micro_Channel_architecture#Marketshare_issues
>>
>>> The majority within IBM never wanted into that part of the market in the first
>>> place, as it was seen as cannibalizing not only 3XXX terminal sales, but the
>>> entire, highly profitable, big-iron+interface+network+services infrastructure
>>> behind said terminals.
>> do you have a reference for this?
>
> There's truth on both sides, IBM had committed to the PC because Apple
> was stealing the show,

No - they neither knew nor cared about that.

Thinking here is being boxed by a 'personal computer' worldview.

Big bucks NOW - small change THEN. Vanishing margins now.

Let's talk about IBM's middle initial - where the money was.

'Business'

They 'committed' to the PC because it was 3270-capable, yet was far more
flexible than a 3270 AND could go head-to-head with everything from an ADM-3A,
SOROQ IQ-120, TVI-9xx, Wyse 1XX thru 3XX (color) DEC VT-52 up (including color)
  HP terminals, Sun workstation (low-end), Xerox, Wang, Data General, [GE,
Honeywell, Bull] (eventually merged), CDC, ICL/Fujitsu/Siemens, Nixdorf,
Burroughs, Univac --> Unisys, etc.. any or all of whom were lining up to eat at
least a slice of what IBM saw as 'their' lunch, low, middle, or high end.

DEC alone DID eat a very large chunk of it at their peak, and HP MPE-3000 still
won't die.

Having a configurable box that could be a dumb OR smart terminal OR go
head-to-head with a Xerox, Wang, or baby DEC - moreover a box with an IBM logo -
made selling the mainframe and midframe and support much easier, simplified
inventory, and started to erode the low-end of all the OTHER competitors.

Poisoning THE OTHER GUY's well more than IBM's well, if you will.

Apple, CP/M, AT&T Bellmac and the entire Unix movement - were not even a blip on
the radar *commercially*.

The 'enemy' was DEC .. and to a lesser extent, HP, SUN, Unisys, and even Wang's
quite competent office networks.

 >  but within IBM there were definitely movements
> keeping the PC at bay.  My understanding was that the 8250, crappy
> UART that it was, was used specifically because SDLC required
> synchronous RS-232 and the 8250 didn't have it.

The INS8250 most certainly DID have it. All you had to do was connect the lines.

I've run them on the George Morrow Designs 'Empire' S-100 board at 112KBps
synchronous and 56 KBps async for hours on end.

The only 'glue' needed was level-shifters - discrete transistors on my OSI
Challenger II, Motorola 1488 & 1489 diode-coupled-logic on everything up until
the 16XXX derivative of the 8250 was sucked into a 'bridge' chipset.

The 'can't do more than 9.6' issue was purely an OS problem on the PC.

The same 240-odd byte Forth routine that ran the chipsets on the Morrow S-100
boar ported with no more than a base-address change to the IBM PC. Under LMI
Forth, the PCDOS and its brain-dead interrupt handling used to boot into LMI
Forth were pushed aside, and the 'can't do' IBM PC MB also ran well at baud
rates not to be seen again until OS/2 on 386 with 16550.

>  Note that the 8251
> was compatible with the 8088 (both Intel designs, if I'm not
> mistaken), where the 8250 came from National Semiconductors and
> required additional glue logic (and had write-only registers and no
> RESET, shudder!).

Not accurate. The problem is that there were TWO chips with the '8250' name, and
they were not alike.

The INS8250 adopted by IBM had only ONE byte of buffering - but it was enough if
you were a machine-code, ASM, or Forth programmer.

PCDOS was lousy at I/O, and Windows no better.

In telco logging gear, OS/2, by comparison, could drive dozens of the same
chipset serial ports at the same time - every one of them at speeds Windows
wouldn't be able to match on ONE port for a decade or more.

BTW - The 'Empire' also used a pair of the Intel 8259 PIC, but correctly
cascaded: 8 X 8 = 64 less the cascade line = 63 hardware interrupt lines.

IBM Boca not only FUBARRED by tying the NMI line to a (generally non-existent)
memory parity checker, insuring there was NFW to analyze or recover from such an
error, they ran the 8259 PIC on the AT in add-on instead of cascade:

8 + 8 = 16, less the cascade line (2/9) = 15 usable interrupts.

Dumb.
>
> Fact is, IBM had distinct, sometimes contrasting marketing objectives.
> I suspect that fighting the Taiwanese menace was as high on the agenda
> as anything could possibly get.

IBM's turnover exceeds that of many entire nations. MS edges them as a 'software
company' - but software is nearly 100% of MS biz, and only 20% of IBM's.

IBM's global headcount exceeds that of many US cities, and even a few entire
state populations.

That is going to insure a good deal of 'sometimes contrasting' contradiction.
It wasn't Taiwan back then, but the *Japanese* who were seen as a threat to IBM
revenue - especially after ex-IBM'er Gene Amdahl got Hitachi to make
plug-compatible mainframes.

NEC and Hitachi/Fujitsu (same parent Keidanren) are about the only competition
IBM have now for volume production 'Big iron'.

>  In "Big Blues" (I think that is the
> book title) it is suggested that IBM did not have a proper focus and
> the PC was a knee-jerk reaction that should have been planned
> considerably better.
>

'Pee Cee' centric thinking. IBM was 'dumb like a Fox' ...again.

The PC did what it was strategically intended to do - left IBM as last-man
standing in the US and European mainframe and midframe business, and able to buy
PC and server parts cheaper than make them.

IBM will not go there again in the same way.

This go, they are using Linux to suck Sun (accomplished) and HP (in work) into a
black hole and cause Redmond to spend a ton of cash. All the while, service &
support is ever more important.  IBM has more of that than all others combined.

If/as/when it makes business sense to do so, IBM will exit the computer biz
entirely in favor of food production, electric cars, space travel - Hell *time*
travel - or whatever else they see a solid one-hundred-year profit run in.

You don't have to love 'em. Few do.

Just never ignore the resources they command.

Bill

Re: [9fans] /sys/include/ip.h 5c(1) - LONG POST

[Ethan Grammatikidis]

On Sat, 10 Oct 2009 04:15:59 +0800
W B Hacker <wbh@conducive.org> wrote:

>
> The only 'glue' needed was level-shifters - discrete transistors on my OSI
> Challenger II, Motorola 1488 & 1489 diode-coupled-logic on everything up until
> the 16XXX derivative of the 8250 was sucked into a 'bridge' chipset.
>

I remember being quite surprised by the first UARTs which had level
shifters on the chip, and they came out around 1993 or so didn't they?
As far as I know it was hardly possible to handle RS-232's 25V signals
on the same die as logic functions back in the 80s.


> PCDOS was lousy at I/O, and Windows no better.

Linux either. Makes me sad. ;-J


--
Ethan Grammatikidis

Those who are slower at parsing information must
necessarily be faster at problem-solving.

Re: [9fans] /sys/include/ip.h 5c(1) - LONG POST

[W B Hacker]

Ethan Grammatikidis wrote:
> On Sat, 10 Oct 2009 04:15:59 +0800
> W B Hacker <wbh@conducive.org> wrote:
>
>> The only 'glue' needed was level-shifters - discrete transistors on my OSI
>> Challenger II, Motorola 1488 & 1489 diode-coupled-logic on everything up until
>> the 16XXX derivative of the 8250 was sucked into a 'bridge' chipset.
>>
>
> I remember being quite surprised by the first UARTs which had level
> shifters on the chip, and they came out around 1993 or so didn't they?
> As far as I know it was hardly possible to handle RS-232's 25V signals
> on the same die as logic functions back in the 80s.

It wasn't 'hard' - especially since the 25V, though more common then than now -
was more likely to be between 7 to 15 volt, and 5V would usually get the job done...

But it just wasn't smart.

The separate circuitry - or even its PC board traces - all too often had to act
as a fuse - protecting the more expensive part of the kit with a cheap and easy
to replace module.

>
>
>> PCDOS was lousy at I/O, and Windows no better.
>
> Linux either. Makes me sad. ;-J
>
>

Don't be too harsh. All Linux lacks is a decent kernel.

;-)

Bill