[comp.sys.next] How did they make the printer so cheap?

jshelton@deimos.ads.com (John L. Shelton) (10/18/88)

How can NeXT possibly make a postscript printer to sell for $2000?
I'm afraid to find out, but I suspect that the printer has no CPU and
memory.  More likely, the printer uses the NeXT CPU and memory.  I got
suspicious when I heard that there was a specific PRINTER port for the
machine.  Anyone got the real scoop?

=John=

ilan343@violet.berkeley.edu (10/18/88)

In article <5807@zodiac.UUCP> jshelton@ads.com (John L. Shelton) writes:
>
>How can NeXT possibly make a postscript printer to sell for $2000?
>I'm afraid to find out, but I suspect that the printer has no CPU and
>memory.  More likely, the printer uses the NeXT CPU and memory.  I got
>suspicious when I heard that there was a specific PRINTER port for the
>machine.  Anyone got the real scoop?
>
>=John=
It is not a postscript printer. Is this a change in Adobe's licensing
policy? I thought you had to have postscrip in the printer.

wb1j+@andrew.cmu.edu (William M. Bumgarner) (10/18/88)

The printer is _NOT_ a postscipt printer. It is a modified Cannon print
engine (accpets LaserWriter II toner cartridges) that prints pure bit images.

Since the computer is running Display Postscript, it builds the image and
then bit-blasts it out the printer port to the printer.  In either 400DPI
or 300 DPI (draft-- HAH!) mode.

That is why the printer is so cheap-- no networking, no CPU, no internal
memory (small buffer, anyways), nothing other than the propieatary interface
and the engine...

b.bum
wb1j+@andrew.cmu.edu

bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) (10/18/88)

In article <5807@zodiac.UUCP> jshelton@ads.com (John L. Shelton) writes:
>How can NeXT possibly make a postscript printer to sell for $2000?
>I'm afraid to find out, but I suspect that the printer has no CPU and
>memory.  More likely, the printer uses the NeXT CPU and memory.

Correct.  The PostScript imaging happens in the cube, then the bits
get blasted, real fast, across the cable to the Canon engine in the
printer.  I'm curious about, in a Bezerkeley remote line printer
daemon environment (one cube with a printer per cluster), whether the
imaging happens in the print server cube or in the cube that
originated the print job, and what are the tradeoffs in each one's CPU
and network bandwidth, etc.
-=-
Zippy sez,								--Bob
Th' MIND is the Pizza Palace of th' SOUL

jensen@gt-eedsp.UUCP (P. Allen Jensen) (10/18/88)

In article <5807@zodiac.UUCP>, jshelton@deimos.ads.com (John L. Shelton) writes:
> How can NeXT possibly make a postscript printer to sell for $2000?
> I'm afraid to find out, but I suspect that the printer has no CPU and
> memory.  More likely, the printer uses the NeXT CPU and memory.....

Correct, John !   The $2000 printer has no CPU or memory.  There is
a special high-speed laserprinter port on the NeXT box that connects
to the laserprinter.  The laserprinter is basically just the cannon
print engine with no control (ie cpu and memory) electronics.

Since NeXT uses Display Postscript, a (hopefully) good postscript
interpreter is already there waiting to be used.  As long as one
remembers that this is supposed to be a one-person-one-machine
workstation, that should not be a big problem unless you print lots
of large documents.

P. Allen Jensen
-- 
P. Allen Jensen
Georgia Tech, School of Electrical Engineering, Atlanta, GA  30332-0250
USENET: ...!{allegra,hplabs,ulysses}!gatech!gt-eedsp!jensen
INTERNET: jensen@gteedsp.gatech.edu

cramer@optilink.UUCP (Clayton Cramer) (10/19/88)

In article <15626@agate.BERKELEY.EDU>, ilan343@violet.berkeley.edu writes:
> In article <5807@zodiac.UUCP> jshelton@ads.com (John L. Shelton) writes:
> >
> >How can NeXT possibly make a postscript printer to sell for $2000?
> >I'm afraid to find out, but I suspect that the printer has no CPU and
> >memory.  More likely, the printer uses the NeXT CPU and memory.  I got
> >suspicious when I heard that there was a specific PRINTER port for the
> >machine.  Anyone got the real scoop?
> >
> >=John=
> It is not a postscript printer. Is this a change in Adobe's licensing
> policy? I thought you had to have postscrip in the printer.

Not a requirement -- the IBM PostScript printer has the PostScript in
the controller card in the computer.

Does anyone really have any specs about the NeXT PostScript printer?
What sort of interfaces does it have?  (When will I be able to buy
one for my PC?)




-- 
Clayton E. Cramer
..!ames!pyramid!kontron!optilin!cramer

barry@confusion.ads.com (Barry Lustig) (10/19/88)

In article <5807@zodiac.UUCP> jshelton@ads.com (John L. Shelton) writes:
	
	How can NeXT possibly make a postscript printer to sell for $2000?
	I'm afraid to find out, but I suspect that the printer has no CPU and
	memory.  More likely, the printer uses the NeXT CPU and memory.  I got
	suspicious when I heard that there was a specific PRINTER port for the
	machine.  Anyone got the real scoop?
	
	=John=


All of the postscript is in the CPU box.  A raster image is blasted
down to the printer for printing.

barry

swilson%thetone@Sun.COM (Scott Wilson) (10/19/88)

>Since the computer is running Display Postscript, it builds the image and
>then bit-blasts it out the printer port to the printer.  In either 400DPI
>or 300 DPI (draft-- HAH!) mode.

I don't know squat about printers, so I'm confused about something.  If
the computer is going to build the image then for an 8 1/2 x 11" sheet
at 400 dpi wouldn't you need 8.5x400x11x400 bits of memory in which
the image is built?  This works out to be roughly 1.8 meg of memory
(assuming I didn't screw up somewhere).  Does this mean that to
compose a page for the printer the NeXT machine has to briefly use
this much of its memory?  What kind of transfer rates are expected
out the printer port?


--
Scott Wilson		arpa: swilson@sun.com
Sun Microsystems	uucp: ...!sun!swilson
Mt. View, CA

nghiem@ut-emx.UUCP (nghiem) (10/19/88)

In article <5807@zodiac.UUCP>, jshelton@deimos.ads.com (John L. Shelton) writes:
> 
> How can NeXT possibly make a postscript printer to sell for $2000?
> I'm afraid to find out, but I suspect that the printer has no CPU and
> memory.  More likely, the printer uses the NeXT CPU and memory.  I got


Your absolutely correct. The NeXT CPU controls the printer. A Business
Week article (this weeks) says that it was meant to keep the price of
the printer down. But, it also means that only the NeXT printer can be
used with the machine at the moment. Also, having the Printer CPU on
board could have something to do with the Display Postscript video.

nghiem@emx.utexas.edu

Why?...Because we LIKE you.

.

ronc@fai.UUCP (Ronald O. Christian) (10/20/88)

In article <73489@sun.uucp> swilson@sun.UUCP (Scott Wilson) writes:
>[...]  This works out to be roughly 1.8 meg of memory
>(assuming I didn't screw up somewhere).  Does this mean that to
>compose a page for the printer the NeXT machine has to briefly use
>this much of its memory?  What kind of transfer rates are expected
>out the printer port?

If I were Jobs, I would set things up so I could do a DMA of the video
memory and blast it directly out to the printer.  No fuss, no muss.


				Ron
-- 

      Ronald O. Christian (Fujitsu America Inc., San Jose, Calif.)
      {amdahl, pyramid, sun, unisoft, uunet}!fai!ronc -or- ronc@fai.com

      Calling all Fujitsu Usenet sites!  Contact fai!ronc or
      ronc@fai.com to establish uucp connection.

muir@postgres.uucp (David Muir Sharnoff) (10/20/88)

In article <73489@sun.uucp> swilson@sun.UUCP (Scott Wilson) writes:
>this much of its memory?  What kind of transfer rates are expected
>out the printer port?

1Mbit/s in 300dpi mode and 3Mbit/s in 400dpi mode.   The printer gets
one of the 12 "channels" on the DMA chip.  

-Dave

tswift@well.UUCP (Theodore John Swift) (10/20/88)

>>How can NeXT possibly make a postscript (PS) printer to sell for $2000?
>>I'm afraid to find out, but I suspect that the printer has no CPU and
>>memory.

>It is not a postscript printer. Is this a change in Adobe's licensing
>policy? I thought you had to have postscript in the printer.

I dunno about whether the printer is PS or not (why have DPS and *not* have
a PS printer? -other than parts, assembly, and licensing costs?).

However, on p.8 of the 18 Oct issue of MacWeek, they mention "A 400 dot per
inch laser printer, controlled by the 68030 processor inside the NeXT box,
will sell for $2000."
  This probably means that all your fonts and other printing resources are in
the CPU box, also.  Sigh.  I was hoping that Apple would come out with a new
LaserWriter at half price :-).
-- 
Ted Swift                        "Why is there a watermelon there?"
{hplabs,lll-crg/lcc, pacbell}               "I'll explain later"
               !well!tswift             - from "Buckaroo Banzaii"

bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) (10/20/88)

In article <24895@tut.cis.ohio-state.edu> bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) writes:
>In article <5807@zodiac.UUCP> jshelton@ads.com (John L. Shelton) writes:
>>How can NeXT possibly make a postscript printer to sell for $2000?
>>I'm afraid to find out, but I suspect that the printer has no CPU
>>and memory.  More likely, the printer uses the NeXT CPU and memory.
>
>Correct.  The PostScript imaging happens in the cube, then the bits
>get blasted, real fast, across the cable to the Canon engine in the
>printer.  I'm curious about, in a Bezerkeley remote line printer
>daemon environment (one cube with a printer per cluster), whether the
>imaging happens in the print server cube or in the cube that
>originated the print job, and what are the tradeoffs in each one's
>CPU and network bandwidth, etc.

It turns out that the remote lpr daemons ship PostScript across the
network (using the standard TranScript stuff), which is clearly a more
efficient way of describing a picture that was already in PostScript
than shipping a bitmap.  The imaging is done on the cube that hosts
the printer.  The CPU running the imaging software will hardly sweat,
even with everything else going on, so the user won't complain much
about that.  The imager can image much faster than this printer can
print (8ppm), so the bottleneck is in the printer.

At 400dpi resolution, a full page bitmap is almost 2Mb.  If a cube is
a printer host for several heavy printer users, it will be
advantageous to add extra memory (in 4Mb SIMM increments).  Otherwise,
you might start (horrors! :-) exercising the VM system and swapping.
If the bitmap only uses 2Mb of that 4Mb addition, then the user of the
printer server gets 2Mb for free, and will hardly complain.
-=-
Zippy sez,								--Bob
Didn't I buy a 1951 Packard from you last March in Cairo?

sean@ms.uky.edu (Sean Casey) (10/20/88)

In article <73489@sun.uucp> swilson@sun.UUCP (Scott Wilson) writes:
>I'm confused about something.  If
>the computer is going to build the image then for an 8 1/2 x 11" sheet
>at 400 dpi wouldn't you need 8.5x400x11x400 bits of memory in which
>the image is built?  This works out to be roughly 1.8 meg of memory
>(assuming I didn't screw up somewhere).  Does this mean that to
>compose a page for the printer the NeXT machine has to briefly use
>this much of its memory?

Not exactly. Imagen printer resident software, for example, allocates
memory in small blocks. I believe someone in the company told me 32
byte blocks. A block isn't allocated unless there is some black in it.
This is how a 512K 8/300 can reasonably print text and graphics.  I
suspect that text operations (i.e. fonts) are rasterized on the fly,
and don't require preallocations. This is only a guess.

This kind of technique would lend itself well to NeXT, and I suspect
that it's being done this way already.

My concern is how fast the Postscript processing is done. I understand
that Postscript is horribly cpu intensive. I'd hate to think of what
happens if your rasterization software goes virtual.

Sean
-- 
***  Sean Casey                        sean@ms.uky.edu,  sean@ukma.bitnet
***  The Hacker from Hell.             {backbone|rutgers|uunet}!ukma!sean
***  U of K, Lexington Kentucky, USA  ..where christian movies are censored.
***  ``The World... she's a flat! She's a round! Flat! Round! Flat! Round!''

sho@pur-phy (Sho Kuwamoto) (10/20/88)

In article <574@optilink.UUCP> cramer@optilink.UUCP (Clayton Cramer) writes:
<In article <15626@agate.BERKELEY.EDU<, ilan343@violet.berkeley.edu writes:
<< In article <5807@zodiac.UUCP< jshelton@ads.com (John L. Shelton) writes:
<< <
<< <How can NeXT possibly make a postscript printer to sell for $2000?
<< <I'm afraid to find out, but I suspect that the printer has no CPU and
<< <memory.  More likely, the printer uses the NeXT CPU and memory.  I got
<< <suspicious when I heard that there was a specific PRINTER port for the
<< <machine.  Anyone got the real scoop?
<< <
<< <=John=
<< It is not a postscript printer. Is this a change in Adobe's licensing
<< policy? I thought you had to have postscrip in the printer.
<
<Not a requirement -- the IBM PostScript printer has the PostScript in
<the controller card in the computer.
<
<Does anyone really have any specs about the NeXT PostScript printer?
<What sort of interfaces does it have?  (When will I be able to buy
<one for my PC?)

I think the printer gets its own DMA channel, so good luck getting one
for your PC. 

-Sho

dill3360@neptune.uucp (Tom Dillon) (10/20/88)

In article <7049@ut-emx.UUCP> nghiem@ut-emx.UUCP (nghiem) writes:
>Your absolutely correct. The NeXT CPU controls the printer. A Business
>Week article (this weeks) says that it was meant to keep the price of
>the printer down. But, it also means that only the NeXT printer can be
>used with the machine at the moment.
>nghiem@emx.utexas.edu

Right. You have two serial ports on the cube; the cube runs mach; mach
supports I/O redirection.  Therefore, if you have a postscript file and
a non-NeXT serial postscript printer, you can print postscript.  Maybe 
not very fast, but you can do it.  And you'll have to show me a postscript
speaking printer for <$2000 to get me to do it too(unless you have one
laying around 8})

kean
-------------------------------------------------------------------------------
Oregon State University                           Kean Stump
College of Oceanography "Cruise Support"          kean@cs.orst.edu 
"Where's the floppy?"                            {tektronix,hp-pcd}!orstcs!kean
-------------------------------------------------------------------------------

wb1j+@andrew.cmu.edu (William M. Bumgarner) (10/21/88)

>> If I were Jobs, I would set things up so I could do a DMA of the video
memory and blast it directly out to the printer.  No fuss, no muss.

no resolution.

rwhite@nusdhub.UUCP (Robert C. White Jr.) (10/21/88)

Guess what this line does  (smack gurgle blurb!)  ;-)

in article <15626@agate.BERKELEY.EDU>, ilan343@violet.berkeley.edu says:
> It is not a postscript printer. Is this a change in Adobe's licensing
> policy? I thought you had to have postscrip in the printer.

According to what I read, the NEXT's printer driver uses display
postscript in a virtual buffer in memory to construct the immage
(using the CPU et al) in a virtual screen memory buffer, and then
transcribes the bit-image into a VERY simplified dump protocol for the
printer.  The brains of the printer live on a single VLSI (custom)
chip which controls an otherwise standard engine (don't know the
manufacturer).

No muss, no fuss, no bother.

(this from info-world)

Rob.

kent@lloyd.camex.uucp (Kent Borg) (10/21/88)

In article <25141@tut.cis.ohio-state.edu> bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) writes:
...
>             The imager can image much faster than this printer can
>print (8ppm), so the bottleneck is in the printer.
...
>-=-
>Zippy sez,								--Bob
>Didn't I buy a 1951 Packard from you last March in Cairo?

Not so fast!  

I have seen unimpressive pages on which a LaserWriter cranks well over
an hour to output.  For the NeXT machine to output a 1
LaserWriter-hour page at 8 pages a minute at the same 300 dpi
resolution (never mind 400 dpi) it would have to interpret PostScript
more than 28000 times as fast does a LaserWriter.

Certainly for simple pages the NeXT laser printer might be the
bottleneck, but let's not pretend there is no limit to Jobs' magical
powers.  Someone can always throw a more complicated PostScript
program at his NeXT cube.  There is no way he can guarantee how fast
it will be run.

Kent Borg
kent@lloyd.uucp
or
hscfvax!lloyd!kent

bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) (10/22/88)

In article <231@lloyd.camex.uucp> kent@lloyd.UUCP (Kent Borg) writes:
>In article <25141@tut.cis.ohio-state.edu> bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) writes:
>>The imager can image much faster than this printer can print (8ppm),
>>so the bottleneck is in the printer.
>
>I have seen unimpressive pages on which a LaserWriter cranks well
>over an hour to output... Someone can always throw a more complicated
>PostScript program at his NeXT cube.

Quite so.  That 8ppm, as with any printer that prints images described
by an interpreted language, will of course be achieved when imaging
simpler pages.  For more complex pages, the bottleneck will move from
the print engine into the imager.

Still, given the speed improvement between a 68000 driving a
LaserWriter imager and a 68030 driving a NeXT cube imager, I'd expect
that as we move to more complex pages, the bottleneck stays in the
print engine longer - that is, a NeXT cube/printer can probably
maintain its full rated speed on more complex pages than those on
which a LaserWriter can maintain its.

Now, if you're also doing SQL retrievals of your Sybase multi-composer
style guide so your Common Lisp engine will compose a Sousa-like march
in real time to send to your DSP to listen to while you wait for your
document, you might find your printer slowing down just a bit :-)
-=-
Zippy sez,								--Bob
.  I don't know why but, suddenly, I want to discuss declining I.Q.
 LEVELS with a blue ribbon SENATE SUB-COMMITTEE!

casseres@Apple.COM (David Casseres) (10/26/88)

In article <73489@sun.uucp> swilson@sun.UUCP (Scott Wilson) writes:

>I don't know squat about printers, so I'm confused about something.  If
>the computer is going to build the image then for an 8 1/2 x 11" sheet
>at 400 dpi wouldn't you need 8.5x400x11x400 bits of memory in which
>the image is built?  This works out to be roughly 1.8 meg of memory
>(assuming I didn't screw up somewhere).  Does this mean that to
>compose a page for the printer the NeXT machine has to briefly use
>this much of its memory?  What kind of transfer rates are expected
>out the printer port?

You didn't screw up, it is roughly 1.8 meg.  I presume that they do indeed
pig up this much RAM in order to print.  Remember they have a virtual
memory system as well as lots of RAM, so they can do this; at worst, it
might mean that applications would have to do more swapping while printing
is in progress.  At best, the 1.8 meg would be "free" in the 8 meg total
system, and since the I/O to the printer from RAM is DMA, the applications
would be totally unaffected.  Nice.  I don't recall the transfer rate (it's
in the Byte article) but it's about what you'd need to hose that 1.8 meg
out to the printer within the 6 seconds or so that the printer allows, once
the paper has begun to roll.

David Casseres

gnu@hoptoad.uucp (John Gilmore) (10/29/88)

casseres@Apple.COM (David Casseres) wrote:
>                  At best, the 1.8 meg would be "free" in the 8 meg total
> system, and since the I/O to the printer from RAM is DMA, the applications
> would be totally unaffected.  Nice.  I don't recall the transfer rate (it's
> in the Byte article) but it's about what you'd need to hose that 1.8 meg
> out to the printer within the 6 seconds or so that the printer allows...

I'm not sure what he means by "free" but the RAM is certainly there to
be "used"; that's what RAM is for.  However, just having DMA between
memory and the printer does not guarantee that "applications would be
totally unaffected".  The interface to a print engine looks a lot like
video going to a monitor; you send each scan line as bits on a serial
line.  The Byte article quotes 3.2M bits/sec, or about 300 ns per bit.
If they can fetch 128 bits (16 bytes) from main memory in 9 cycles,
that's 128 bits/360ns, (at 25MHz, a cycle is 40ns), so the printer
interface, when running, burns about 1/100th of the memory bandwidth,
which is not bad; applications would not notice.

But don't forget the CPU time burned by the PostScript interpreter.
We sell one, we know; high quality imaging of 1.8MB of bits in an
interpreted language definitely runs the CPU at peak for a while.

The point of the NeXT design is that the CPU subsystem now has enough
resources that when printing, it can spare a significant chunk of them
to drive the printer (and with dense gate arrays, the interface to the
printer can come standard because it's close to free).  So all you
have to pay for is the print engine and software.  This was not an
option on a 512K Mac with a slowly clocked 68000 controlled by PALs, so
they had to put the 12.5MHz 68000 and 1.5MB inside the printer.  Of
course, if the original Mac had had a bus, they could've put the extra
1.5MB and the fast 68K on the bus, so when the printer was idle you
could use it for your applications!  But that's design by hindsight.
-- 
John Gilmore    {sun,pacbell,uunet,pyramid,amdahl}!hoptoad!gnu    gnu@toad.com
		Noriega-Bush in '88 -- a *crack* team.  
Let's put the white powder (CIA = Cocaine Import Agency) in the white house!

casseres@Apple.COM (David Casseres) (11/01/88)

In article <5784@hoptoad.uucp> gnu@hoptoad.uucp (John Gilmore) writes:
>casseres@Apple.COM (David Casseres) wrote:
>>                  At best, the 1.8 meg would be "free" in the 8 meg total
>> system, and since the I/O to the printer from RAM is DMA, the applications
>> would be totally unaffected.  Nice.  I don't recall the transfer rate (it's
>> in the Byte article) but it's about what you'd need to hose that 1.8 meg
>> out to the printer within the 6 seconds or so that the printer allows...
>
>I'm not sure what he means by "free" but the RAM is certainly there to
>be "used"; that's what RAM is for.

I just meant that in the best case, the space could be allocated without
having to swap anything else out.

>...  However, just having DMA between
>memory and the printer does not guarantee that "applications would be
>totally unaffected".

Correct; I should have said virtually unaffected, and I should have added
"by the I/O requirements."

>But don't forget the CPU time burned by the PostScript interpreter.
>We sell one, we know; high quality imaging of 1.8MB of bits in an
>interpreted language definitely runs the CPU at peak for a while.

I don't think this will have a very large impact on applications running
on the same machine.  We know their implementation of Display PostScript
can paint the NeXT screen fast enough for a whizzy interface, and that
screen is a million pixels X 2 bits/pixel = 2 megabits or about 0.25 MB.
To draw a page of about 2 MB would take about 8 screen times, which just
doesn't sound to me like a big hit.

David Casseres