bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) (10/18/88)
In article <10639@reed.UUCP> mdr@reed.UUCP (Mike Rutenberg) writes: >But where is the RJ-11? That's the little plug that is on the back >of your modem that connects it to the phone system. > >The DSP will make a wonderful 9600 bps modem, but only if it can >connect to the phone line to do modem like things. How is this >intended to be done? Did NeXT just run out of space on the system >board? The 9600bps modem was built into the early revs of the board, but was later moved outside the case in a little box that just hangs there like an appendage. That was a concession to internationalization because not all countries' modems and telco interfaces are the same, so it seemed better to keep the portable-hardware signal processing inside the case and the variant stuff on the outside. This also leaves more flexibility for use as a FAX modem, etc. -=- Zippy sez, --Bob I'm pretending that we're all watching PHIL SILVERS instead of RICARDO MONTALBAN!
david@ms.uky.edu (David Herron -- One of the vertebrae) (10/18/88)
I had a bit of a disappointment last night ...
I opened up my trailblazer to see which DSP chip it had. Telebit
had one marked "DSP34010", is that the TI chip? Anyway, it's certainly
not the same one as in the NeXT machine.
It woulda been kinda neat to have a software trailblazer :-)
--
<-- David Herron; an MMDF guy <david@ms.uky.edu>
<-- ska: David le casse\*' {rutgers,uunet}!ukma!david, david@UKMA.BITNET
<--
<-- "Smarter than the average pagan god ... "ken@gatech.edu (Ken Seefried iii) (10/18/88)
In article <10383@s.ms.uky.edu> david@ms.uky.edu (David Herron -- One of the vertebrae) writes: >I opened up my trailblazer to see which DSP chip it had. Telebit >had one marked "DSP34010", is that the TI chip? Anyway, it's certainly >not the same one as in the NeXT machine. The 34010 is actually the TI graphics chip. Interestingly, it is also a pretty sharp 5MIPS or so general purpose processor. They have a CCITT Group 3 & 4 (aka fax) image compression package that runs as fast as the dedicated AMD fax compression chips. I never thought of it as a modem chip, but it makes a kind of sense... > >It woulda been kinda neat to have a software trailblazer :-) I don't see why its not possible. My understanding (sure to be flawed) is that the PEP protocol used by the Trailblaser is a software only protocol. It should not be difficult to code a 56001 version of PEP, if this is true. >-- ><-- David Herron; an MMDF guy <david@ms.uky.edu> ken seefried iii ...!{akgua, allegra, amd, harpo, hplabs, ken@gatech.edu masscomp, rlgvax, sb1, uf-cgrl, unmvax, ccastks@gitvm1.bitnet ut-ngp, ut-sally}!gatech!ken
jbs@eddie.MIT.EDU (Jeff Siegal) (10/18/88)
In article <24824@tut.cis.ohio-state.edu> bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) writes: >In article <10639@reed.UUCP> mdr@reed.UUCP (Mike Rutenberg) writes: >>But where is the RJ-11? [...] >[... it] was later moved outside the case in a little box that just hangs >there like an appendage. [...] This also leaves more flexibility for use as >a FAX modem, etc. I don't see why this would be true. Every FAX I've ever seen has an RJ-11, just like every modem I've ever seen. I can see how external phone interface circuitry could simplify internationalization, but not how it would help the NeXT talk FAX. Jeff Siegal
dtynan@sultra.UUCP (Der Tynan) (10/18/88)
In article <24824@tut.cis.ohio-state.edu>, bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) writes: > > The 9600bps modem was built into the early revs of the board, but was > later moved outside the case in a little box that just hangs there > like an appendage. [...] > > This also leaves more flexibility for use as a FAX modem, etc. > -=- > Zippy sez, --Bob The thing I have against all this "9600 baud modem emulator" stuff, is if you try to do any "real" work with the system, besides just a terminal emulator, you'll probably get bogged down real quick. I mean, can you imagine using such an arrangement as a newsfeed? So, here we have uucico running, which forks uuxqt, which forks sh, which forks rnews, which forks compress and another rnews, not to mention any residual network daemons. Into this environment, you're going to throw in a full Telebits PEP emulation??? I don't think so. Gone are the days of FSK modems. The amount of work a Trailblazer does must be gastronomical, what with up to 511 channels, each modulated at 7.58 baud? I think personally, I'd just fork over the extra spondoolix, and get a 'real' modem. - Der -- Reply: dtynan@sultra.UUCP (Der Tynan @ Tynan Computers) {mips,pyramid}!sultra!dtynan Cast a cold eye on life, on death. Horseman, pass by... [WBY]
merrill@bucasb.bu.edu (John &) (10/18/88)
In article <2583@sultra.UUCP>, dtynan@sultra (Der Tynan) writes: >In article <24824@tut.cis.ohio-state.edu>, bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) writes: >> >> The 9600bps modem was built into the early revs of the board, but was >> later moved outside the case in a little box that just hangs there >> like an appendage. [...] >> > >The thing I have against all this "9600 baud modem emulator" stuff, is if you >try to do any "real" work with the system, besides just a terminal emulator, >you'll probably get bogged down real quick. I mean, can you imagine using >such an arrangement as a newsfeed? Sure. Remember that the proposed modem emulator runs *on the DSP chip*---not on the main CPU. (I agree that you couldn't run such a system on the main CPU and handle a full news-feed simultaneously.) --
bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) (10/18/88)
In article <2583@sultra.UUCP> dtynan@sultra.UUCP (Der Tynan) writes: >The thing I have against all this "9600 baud modem emulator" stuff, >[... my system's already busy with all the real work ...] >Into this environment, you're going to throw in a full Telebits PEP >emulation??? I don't think so. I suspect that the 68030/68882 has enough juice to handle a PEP thread without sweating too hard, if you really want it to. Someone may be clever enough to put PEP on the DSP someday, but 9600 V.32 is already better than anything I've got. And it all runs on the chip, not bothering my uucico/uuxqt/rnews/compress/rnews/whatever (which will run more efficiently in Mach's IPC environment to begin with). Again, in a University environment, many of these cubes will be NNTP clients and won't do the drudgery of unbatching the news. Give that onerous task to the grant 3B2/400s that ATT was throwing away a couple of years ago (like osu-cis) or the spare VAX that everyone has lying about. For that matter, these cubes probably won't have many modem connections, except maybe to run SLIP from my study at home to my office on campus :-) If the DSP is busy being a modem, you may not get the full 44.1Mhz (?) sampling on your real-time harpsichord synthesis. That's probably the biggest reason I'd go buy a Trailblazer :-) -=- Zippy sez, --Bob It's OKAY --- I'm an INTELLECTUAL, too.
dorn@fabscal.UUCP (Alan Dorn Hetzel) (10/20/88)
In article <2583@sultra.UUCP> dtynan@sultra.UUCP (Der Tynan) writes: >In article <24824@tut.cis.ohio-state.edu>, bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) writes: >The thing I have against all this "9600 baud modem emulator" stuff, is if you >try to do any "real" work with the system, besides just a terminal emulator, >you'll probably get bogged down real quick. I mean, can you imagine using ... Not necessarily, the modem emulation runs mostly in the DSP56001 chip, not in the main CPU (at least if they did things right) Dorn
henry@utzoo.uucp (Henry Spencer) (10/26/88)
In article <17500@gatech.edu> ken@gatech.UUCP (Ken Seefried iii) writes: >>I opened up my trailblazer to see which DSP chip it had. Telebit >>had one marked "DSP34010", is that the TI chip? ... > >The 34010 is actually the TI graphics chip... I think the original poster made a typo; the processor inside the Telebits is the 32010, not the 34010. At least, it was when I opened mine up, unless I've got bit rot in my memory. -- The dream *IS* alive... | Henry Spencer at U of Toronto Zoology but not at NASA. |uunet!attcan!utzoo!henry henry@zoo.toronto.edu
wtm@neoucom.UUCP (Bill Mayhew) (10/27/88)
Yes, the chip used in the Trailblazer is a standard part made by Texas Instrments. Most of the Trailblazers I've seen use a General Instruments second-souce part that is the functional equivalent of TI's. I can't find the datasheet, so I'll let somebody else qoute the performance data. The Motorola part in the Next machine is a higher performance DSP, I believe it is safe to say. It would be neat if somebody would put together a software based Trailblazer for the NeXT machine. I know NeXT is working on some sort of 9600 bps modem, but I haven't heard any details on wheter or not it will conform to one of the "v." [proposed] standards. --Bill