gillies@p.cs.uiuc.edu (05/24/89)
1.5 years ago cheapo external 2400-baud modems were selling for
$180-$220. Now the cheapest 2400-baud modems are selling for about
$130.
My question is, when will faster modems replace the 2400-baud
standard? It's been a couple of YEARS and the progress does not seem
great. I expected that by now, 4800 baud modems would be standard.
Where are these modems? How much do they cost? Why don't they cost
~$250 by now?
Don Gillies, Dept. of Computer Science, University of Illinois
1304 W. Springfield, Urbana, Ill 61801
ARPA: gillies@cs.uiuc.edu UUCP: {uunet,harvard}!uiucdcs!gilliesgrr@cbmvax.UUCP (George Robbins) (05/25/89)
In article <78700004@p.cs.uiuc.edu> gillies@p.cs.uiuc.edu writes: > > 1.5 years ago cheapo external 2400-baud modems were selling for > $180-$220. Now the cheapest 2400-baud modems are selling for about > $130. > > My question is, when will faster modems replace the 2400-baud > standard? It's been a couple of YEARS and the progress does not seem > great. I expected that by now, 4800 baud modems would be standard. > Where are these modems? How much do they cost? Why don't they cost > ~$250 by now? Huh? Which planet are you on? There's lots of activity in the 9600+ range with volume prices down in the 500-800 dollar price range. It takes at least a couple of years for "leading edge" technology to become made-in-taiwan, almost free technology. You may have not "noticed" how long "expensive" 2400 baud modems were avilable before the first "consumer" (Hayes/USR, etc) modems hit the market and started the first price decrease to the ~500 dollar range. Implementations are also not particularly cheap at this point. A lot of these modems are using DSP chips of one sort or another - all of the V.32 modems are currently using the same Rockwell DSP chipset, which being single sourced, isn't immediately subject to price competition. Also included are random amounts of memory, unneeded in the economy 1200/2400 buad designs. Hang on for another year or two and maybe you'll see $89.95 9600 baud modems, but I wouldn't hold your breath... -- George Robbins - now working for, uucp: {uunet|pyramid|rutgers}!cbmvax!grr but no way officially representing arpa: cbmvax!grr@uunet.uu.net Commodore, Engineering Department fone: 215-431-9255 (only by moonlite)
john@zygot.UUCP (John Higdon) (05/25/89)
In article <78700004@p.cs.uiuc.edu>, gillies@p.cs.uiuc.edu writes: > My question is, when will faster modems replace the 2400-baud > standard? It's been a couple of YEARS and the progress does not seem > great. I expected that by now, 4800 baud modems would be standard. With 9600+ bps technology now readily available, it would seem unlikely that anyone would be putting much effort into developing 4800 bps standards. The gap between 2400 and 9600+ bps is very wide and the differences in technology is very great, but I would expect to see the price of 9600+ units come down rather than "cheap" modems get faster. -- John Higdon | P. O. Box 7648 | +1 408 723 1395 john@zygot.uucp | San Jose, CA 95150 | M o o !
dave@westmark.UUCP (Dave Levenson) (05/25/89)
In article <78700004@p.cs.uiuc.edu>, gillies@p.cs.uiuc.edu writes: > 1.5 years ago cheapo external 2400-baud modems were selling for > $180-$220. Now the cheapest 2400-baud modems are selling for about > $130. > My question is, when will faster modems replace the 2400-baud > standard? It's been a couple of YEARS and the progress does not seem > great. I expected that by now, 4800 baud modems would be standard. > Where are these modems? How much do they cost? Why don't they cost > ~$250 by now? I would suggest that you read this newsgroup! Look at the discussion surrounding the several 9600 and 19,200 and faster modems currently on the market. Why anyone would bother with 4800 at this point is beyond me. Predicting the future is always risky, but at the risk of having to eat a few words in a year or two, I will predict that two current needs will be met in the near future: A standard for high-speed modems will be adopted, and they'll get less expensive. Today, a 9600 or faster modem costs well over $1,000 (yes, I know they are sometimes discounted to around $600 - $800). The bigger problem with buying one is that there are several incompatible families of them. I would prefer to sit on the sidelines and talk 2400 bps until there is a high-speed standard as universal as the current crop of 1200 and 2400 speed units. I would also expect the prices to come down as the market expands. When that happens, your 2400 modem will probably be sold used at computer flea markets for $20 or so -- as 300-speed units are today. Progress? It's usually not visible until it's already happened. -- Dave Levenson {uunet | rutgers | att}!westmark!dave ...the man in the mooney
elg@killer.DALLAS.TX.US (Eric Green) (05/26/89)
in article <78700004@p.cs.uiuc.edu>, gillies@p.cs.uiuc.edu says: > 1.5 years ago cheapo external 2400-baud modems were selling for > $180-$220. Now the cheapest 2400-baud modems are selling for about > $130. > > My question is, when will faster modems replace the 2400-baud > standard? It's been a couple of YEARS and the progress does not seem In 1983, when I first got into telecommunications, a Hayes 1200 Smartmodem costed over $500, and the best deal around for a 1200 baud modem was around $350. The standard, of course, was still 300 baud. I bought my first 300 baud modem for $120. I have a boxfull of'em in the closet... I occasionally pull one out to use as a junction box when I want to add a new section to my phone cordage. In 1985, 1200 baud modems dipped below $200 for the first time, and 2400 baud modems got down to around $400-$500. People still had a hard time believing that it was possible to go 2400 baud over the phone lines, and were often saying, "Hmph, it already goes fast enough, why do I want to upgrade to a more expensive modem?" 2400 baud modems got cheap real quick. The Taiwan clone manufacturers got ahold of them, apparently, and did their usual. But 9600 isn't so easy. The problem is the #$%#$% analog-based phone system we have here in the U.S., which has extremely limited bandwidth. As a result, your beautiful digital data is analogized, spit onto the phone line where it is then re-digitized and spit over fiber optics at 64kbaud or so, then re-analogized at the other end into a nice slow analog signal. I believe the signal bandwidth is 7khz. A telephone expert could be more exact. In any event, the whole process is a hack and a kludge, and as you press the limits, it naturally becomes more and more difficult to make progress. It's much like the top end microprocessors... even now, <n> years later, you can't get 80286 processors for the $5 apiece that you can get 8086 processors. -- Eric Lee Green P.O. Box 92191, Lafayette, LA 70509 ..!{ames,decwrl,mit-eddie,osu-cis}!killer!elg (318)989-9849 bcase: "I have seen or heard "designer of the 68000" attached to so many names that I can only guess that the 68000 was produced by Cecil B. DeMile."
mhyman@hsfmsh.UUCP (Marco S. Hyman) (05/26/89)
In article <6977@cbmvax.UUCP> grr@cbmvax.UUCP (George Robbins) writes: > Implementations are also not particularly cheap at this point. A lot > of these modems are using DSP chips of one sort or another - all of the > V.32 modems are currently using the same Rockwell DSP chipset, which > being single sourced, isn't immediately subject to price competition. Not true. Several of the V.32 modems out there pre-date the Rockwell V.32 chipset. Check out the Codec, NEC, UDS, and Hayes V.32 entries for example. --marc -- //Marco S. Hyman //UUCP: ...!sun!sfsun!hsfmsh!mhyman //Domain: sfsun!hsfmsh!mhyman@sun.com
ron@ron.rutgers.edu (Ron Natalie) (05/27/89)
Actually, Anderson Jacobson made a full duplex 4800 baud modem over three years ago. However, technology jumps exponentially frequently. We didn't stop at 600 on the way to 1200 baud modems either. -Ron
snoopy@sopwith.UUCP (Snoopy) (05/27/89)
In article <8196@killer.DALLAS.TX.US> elg@killer.DALLAS.TX.US (Eric Green) writes: | In 1983, when I first got into telecommunications, a Hayes 1200 | Smartmodem costed over $500, and the best deal around for a 1200 baud | modem was around $350. The standard, of course, was still 300 baud. In '83 stupid 1200 baud modems were common inside Bell Labs. (pick up data phone, dial computer, push button on modem/cord, hang up phone) I have no idea what they paid for them. I found it a bit annoying to be limited to 1200 baud talking to a machine in the same building. (In their defense, it was a convienant way to provide switching.) | As a result, your | beautiful digital data is analogized, spit onto the phone line where | it is then re-digitized and spit over fiber optics at 64kbaud or so, | then re-analogized at the other end into a nice slow analog signal. I | believe the signal bandwidth is 7khz. A telephone expert could be more | exact. In the Bell/AT&T worlds, phone lines are sampled at 8 kHz. Samples are 8 bits. The bandwidth is thus guaranteed to be less than 4kHz. I recall numbers of something like 3400, I suspect this varies a lot with the age of equipment and so on. Don't expect to see very high order filters to squeak out every last Hz like you do on CD players, lots of phone equipment is old. Like it or not, the phone system was not designed for high fidelity stereo or computer data transmission. It was designed for voice communication, period. Unfortunately it doesn't always accomplish even that. :-( Sort of amusing remembering my EE communications prof stating that 1200 baud was the absolute maximum one could get over a phone line. *grin* _____ /_____\ Snoopy "My dot-matrix does Postscript." /_______\ |___| qiclab!sopwith!snoopy |___| sun!nosun!illian!sopwith!snoopy parsely!sopwith!snoopy
grr@cbmvax.UUCP (George Robbins) (05/27/89)
In article <753@hsfmsh.UUCP> sfsun!hsfmsh!mhyman@sun.com (Marco S. Hyman) writes: > In article <6977@cbmvax.UUCP> grr@cbmvax.UUCP (George Robbins) writes: > > Implementations are also not particularly cheap at this point. A lot > > of these modems are using DSP chips of one sort or another - all of the > > V.32 modems are currently using the same Rockwell DSP chipset, which > > being single sourced, isn't immediately subject to price competition. > > Not true. Several of the V.32 modems out there pre-date the Rockwell V.32 > chipset. Check out the Codec, NEC, UDS, and Hayes V.32 entries for example. All right! I confess to overgeneralization and ignorance of tru-facts. I'll modify my statment to say the the current "cook-book" solutions to building V.32 modems are limited to (a) using the Rockwell V.32 chipset or (b) using a "module" from UDS. The real price cuts don't start until somebody can avoid paying the piper on proprietary technology, and is willing to sell products without feeling a compulsion to jack up the margin to cover R&D or other business issues. -- George Robbins - now working for, uucp: {uunet|pyramid|rutgers}!cbmvax!grr but no way officially representing arpa: cbmvax!grr@uunet.uu.net Commodore, Engineering Department fone: 215-431-9255 (only by moonlite)
henry@utzoo.uucp (Henry Spencer) (05/28/89)
In article <182@sopwith.UUCP> snoopy@sopwith.UUCP (Snoopy) writes: >Sort of amusing remembering my EE communications prof stating that 1200 baud >was the absolute maximum one could get over a phone line. *grin* All the more so when you consider that Telebit is rumored to be able to do 28 kbps with a souped-up internal version of PEP. Not bad for lines that are sampled at 8 kHz at (as far as the modem is concerned) random times! -- Van Allen, adj: pertaining to | Henry Spencer at U of Toronto Zoology deadly hazards to spaceflight. | uunet!attcan!utzoo!henry henry@zoo.toronto.edu
pmb@swituc.UUCP (Pat Berry) (05/28/89)
In article <7002@cbmvax.UUCP>, grr@cbmvax.UUCP (George Robbins) writes: > The real price cuts don't start until > somebody can avoid paying the piper on proprietary technology, and is > willing to sell products without feeling a compulsion to jack up the > margin to cover R&D or other business issues. I'm afraid I must take exception to this statement. Companies do not produce products for their health. All companies have the right to expect the recovery of costs (including 'R&D or other business issues') plus a decent return on their investment (aka profit). In fact, feduciary laws require the CEO to do his best to make the company profitable!
grr@cbmvax.UUCP (George Robbins) (05/29/89)
In article <132@swituc.UUCP> pmb@swituc.UUCP (Pat Berry) writes: >In article <7002@cbmvax.UUCP>, grr@cbmvax.UUCP (George Robbins) writes: >> The real price cuts don't start until >> somebody can avoid paying the piper on proprietary technology, and is >> willing to sell products without feeling a compulsion to jack up the >> margin to cover R&D or other business issues. >I'm afraid I must take exception to this statement. Companies do not produce >products for their health. All companies have the right to expect the recovery >of costs (including 'R&D or other business issues') plus a decent return on >their investment (aka profit). In fact, feduciary laws require the CEO to do >his best to make the company profitable! My main point was how things become cheap, not why they are often initially expensive. Companies may do as they wish as long as they have some corner on the market. The prices may then decrease a few increments as the result of peer competition, but the reduction to bargain-basement prices doesn't really begin until other competitors come along - either with different technology or operating on a different basis (zero R&D, off-shore manufacture, low overhead sales channels, etc.) -- George Robbins - now working for, uucp: {uunet|pyramid|rutgers}!cbmvax!grr but no way officially representing arpa: cbmvax!grr@uunet.uu.net Commodore, Engineering Department fone: 215-431-9255 (only by moonlite)
jeff@b11.ingr.com (Jeff Kilpatrick) (06/02/89)
In article <1400@westmark.UUCP>, dave@westmark.UUCP (Dave Levenson) writes: > > needs will be met in the near future: A standard for high-speed > modems will be adopted, and they'll get less expensive. > > problem with buying one is that there are several incompatible > families of them. I would prefer to sit on the sidelines and talk > 2400 bps until there is a high-speed standard as universal as the ^^^^^^^^^^^^^^^^^^^ V.32 is the standard for high speed dial up. Prices have dropped quite a bit in the last couple years. Most V.32 modems are compatible (I think Hayes does a half duplex V.32) We will possibly be seeing higher data rates with V.42 (standard which includes data compression). ----------------------------------------------------|------------------------- Jeff Kilpatrick | American by birth... Communications Hardware Test and Evaluation | Southern by the | Grace of God | | Opinions don't necessary | reflect my own... | Much less my companies' ----------------------------------------------------|-------------------------
mhyman@hsfmsh.UUCP (Marco S. Hyman) (06/02/89)
In article <5173@b11.ingr.com> jeff@b11.ingr.com (Jeff Kilpatrick) writes: > Most V.32 modems are compatible > (I think Hayes does a half duplex V.32) Hayes has two 9600 bps modems. The V-Series Smartmodem 9600 is a proprietary half-duplex modem with quick turnaround, compression, etc. The Smartmodem 9600 (not V-Series) is a Full-duplex full compliance V.32 modem. --marc (I work for Hayes but can not speak for them) -- //Marco S. Hyman //UUCP: ...!sun!sfsun!hsfmsh!mhyman //Domain: sfsun!hsfmsh!mhyman@sun.com
gpwrdcs@gp.govt.nz (Don Stokes, Govt Print, Wellington) (06/05/89)
In article <182@sopwith.UUCP>, snoopy@sopwith.UUCP (Snoopy) writes: > Sort of amusing remembering my EE communications prof stating that 1200 baud > was the absolute maximum one could get over a phone line. *grin* Just being picky, but my V22bis 2400 *bps* modem runs at 600 *baud*, ie four bits per baud. Your EE communications prof wasn't that far wrong. I'm not sure what the modulation is on faster modems, but I suspect that the baudrate is <= 1200. Anyone know what the modulations are on V32/HST/TB+PEP? Don Stokes, Systems Programmer / / Domain: don@gp.govt.nz Government Printing Office, /GP/ PSImail: PSI%0530147000028::DON Wellington, New Zealand / / UUCP: ...!munnari!vuwcomp!windy!gpwd!don -------------------------------------------------------------------------------- A shortcut is the longest distance between two points.
paul@csnz.co.nz (Paul Gillingwater) (06/06/89)
In article <581@gp.govt.nz> gpwrdcs@gp.govt.nz (Don Stokes, Govt Print, Wellington) writes: +In article <182@sopwith.UUCP>, snoopy@sopwith.UUCP (Snoopy) writes: +> Sort of amusing remembering my EE communications prof stating that 1200 baud +> was the absolute maximum one could get over a phone line. *grin* + +Just being picky, but my V22bis 2400 *bps* modem runs at 600 *baud*, ie +four bits per baud. Your EE communications prof wasn't that far wrong. +I'm not sure what the modulation is on faster modems, but I suspect that +the baudrate is <= 1200. + +Anyone know what the modulations are on V32/HST/TB+PEP? + +Don Stokes, Systems Programmer / / Domain: don@gp.govt.nz My PEP uses up to 512 carriers equally spaced over the 4 kHz assumed bandwidth. Each individual carrier might use 2-bit QPSK, 4-bit QAM or 6-bit QAM depending upon line quality. -- Paul Gillingwater, Computer Sciences of New Zealand Limited Bang: ..!uunet!dsiramd!csnz!paul Domain: paul@csnz.co.nz Call Magic Tower BBS V21/23/22/22bis 24 hrs +0064 4 767 326
csg@pyramid.pyramid.com (Carl S. Gutekunst) (06/12/89)
In article <581@gp.govt.nz> gpwrdcs@gp.govt.nz (Don Stokes, Govt Print, Wellington) writes: >I'm not sure what the modulation is on faster modems, but I suspect that the >baudrate is <= 1200. Anyone know what the modulations are on V32/HST/TB+PEP? V.32 is 2400 baud, 4 bits per baud. (Actually, 5 bits per baud, including the trellis coding bit. But 4 *user* bits per baud.) <csg>
mcp@ziebmef.uucp (Marc Plumb) (06/13/89)
In article <581@gp.govt.nz> gpwrdcs@gp.govt.nz (Don Stokes, Govt Print, Wellington) writes: >Anyone know what the modulations are on V32/HST/TB+PEP? PEP works at something like 7 baud; it also has an interactive mode that gets lower throughput but lower latency at about 80 baud. 7 baud, but up to 6 bits/baud on 511 channels! HST is the same as V.32; going to half-duplex lets you avoid all the nasty echo-cancellation stuff. -- -Colin Plumb