"Strings" <strings@example.net> wrote in
news:462e45d8$0$97244$892e7fe2@authen.yellow.readf reenews.net:

IMHO he only deserves a $25 hooker.


--
Disagreements and the usual insults expected and welcomed.

thanatoid <waiting@the.exit.invalid> wrote:
There are some rather dramatic differences between 14.4/28.8
(v.32 or v.34) modems and a v.90 or v.92 modem.

The actual bit rate over the modem link does in fact approach
53Kbps (though it is an extremely rare circumstance that it
actually gets that high, and more typical links would run at 42
to 49 Kbps).

The actual throughput depends on how much the data can be
compressed, and might actually peak at 4 times the link rate,
though that again is very rare. More common rates would be just
less than 2x compression on regular text files, and no
compression at all on many binary files.

--
Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@apaflo.com

thanatoid wrote:
Not so. The bit rate can seriously exceed 56k. You will also find
that such things as .zip files do not achieve this.

--
<http://www.cs.auckland.ac.nz/~pgut001/pubs/vista_cost.txt>
<http://www.securityfocus.com/columnists/423>
<http://www.aaxnet.com/editor/edit043.html>
cbfalconer at maineline.net



--
Posted via a free Usenet account from http://www.teranews.com

"thanatoid" <waiting@the.exit.invalid> wrote in message
news:Xns991CA2F8F87Athanexit@66.250.146.158...
Not true...

Take a 1 megabyte file full of zeros and transfer it over a 56k modem...
You'll see VERY high throughput. Compression CAN take the data stream faster
than the signalling rate.

Much of this is moot as dialup data is going the way of the dinosaur. I find
it funny that nobody has come up with a CHEAP and SIMPLE modem that does
strictly Fax and advanced voice.

I agree. What we need is another world war. I can't decide if I'd want to be
a winner or loser in that fight though.

On Apr 23, 8:17 pm, ibupro...@painkiller.example.tld (Moe Trin) wrote:

I was asking whether it is theoretically possible to get such a high-
speed by using those multiple dial-up connections.

I use Windows 98 SE

Actually, I was thinking of multiple ISPs. A different ISP for each
line.

On Apr 24, 3:24 am, "Noozer" <dont.s...@me.here> wrote:

http://img56.imageshack.us/img56/242...example8is.gif

If each clock signal is 1 GHz, and you have a 4 of them, staggered
such that every 1ns part of the CPU can start, and finish, an
instruction - making the effective 'clock rate' 4 GHz.

I wondering if something analogous could be done with dial-up modems.
IOW, using more phone lines/modems/phone #'s to get a faster speed.

Radium wrote:
It has.

--
<http://www.cs.auckland.ac.nz/~pgut001/pubs/vista_cost.txt>
<http://www.securityfocus.com/columnists/423>
<http://www.aaxnet.com/editor/edit043.html>
cbfalconer at maineline.net



--
Posted via a free Usenet account from http://www.teranews.com

"Noozer" <dont.spam@me.here> wrote in
news:JAwXh.123949$6m4.88170@pd7urf1no:

Another very good question. I was originally going to say that
dying would be a plus since one would be done with this stinking
world, but then I thought, MAYBE with 2 billion people life
would be slightly more bearable. Then I remembered the problem
is in MY head so it doesn't matter whether 6 billion people
annoy me or only 2 billion.

I recently saw a quote from a European philosopher which
basically said, although in very different words, "once you're
born, you're screwed".


--
Disagreements and the usual insults expected and welcomed.

CBFalconer <cbfalconer@yahoo.com> wrote in
news:462E8321.182A78CC@yahoo.com:

I realize whether the data transmitted is raw, zipped, rar'd,
yEnc'd or mimed doesn't matter, but I do not believe bitrate can
exceed 56KB (if that) - and most of the others in the thread
seem to feel the same way.

I HAVE noticed something interesting with MY 33.6 modem (laugh -
it works great, and I'm in no hurry).

When dl'g *headers* from groups XNews *sometimes* - not always -
shows bitrates approaching 25000 - fast as heck (well, for a
dialup guy). Actual posts usually dl at about 3500. But most of
the time the headers dl at about 4500-5000 - still faster than
posts, but nothing like 25000.

I think this may have something to do with MRU - several headers
probably fitting in one packet. (Am I using the right terms? The
thing that used to be about half of 1500 and you had to modify
settings for better performance, but Microsoft insisted on 1500
and now the MRU my ISP uses is 1524 ...)

But if it IS the packets, why is the very rapid header download
speed not standard but quite intermittent?


--
Disagreements and the usual insults expected and welcomed.

thanatoid wrote:
But it does matter, highly. Your modem may not do compression, or
may be configured to avoid it. Zipped files are already highly
compressed, and cannot be much further processed.

--
<http://www.cs.auckland.ac.nz/~pgut001/pubs/vista_cost.txt>
<http://www.securityfocus.com/columnists/423>
<http://www.aaxnet.com/editor/edit043.html>
cbfalconer at maineline.net


--
Posted via a free Usenet account from http://www.teranews.com

this is clearly wrong. there are multiple bits per baud in faster modems.
for instance the 9600 baud used for PCs used quadrature modulation.

it's pretty clear, that on a clean analog line you can get at least 30 kbps.
this is due to the bandwidth and SNR, and the Shannon theorem sets an upper
bound on line speed.

56k modems go well past 56 kbps on material which can be compressed heavily.

this is definitely wrong since you can download zipped files (bz2, 7z) at 50
kbps on a clean line. those types of files cannot be compressed much by
v.44.

there is no *general* compression algo that can magically compress any file.
text, HTML, xls, can be compressed because they are highly redundant i.e.
low entropy.

"CBFalconer" <cbfalconer@yahoo.com> wrote in message
news:462ED5C1.EEB107A@yahoo.com...
some stuff that is misunderstood by computer newbies:

1. "bps = baud." this was only true in the early modems.

2. Bps vs bps. Bps usually refers to bytes per second.

3. the line speed is limited by Shannon's theorem.

http://en.wikipedia.org/wiki/Shannon's_theorem

4. line speed vs throughput.

http://en.wikipedia.org/wiki/V.44#Er...ta_compression

5. UART caps..

http://en.wikipedia.org/wiki/UART

CBFalconer <cbfalconer@yahoo.com> wrote in
news:462ED5C1.EEB107A@yahoo.com:

No it does not. If a modem can only receive 25kb per second due
to whatever reasons, it does NOT matter whether that 25kb is raw
text or a zipped file. It will do whatever compression it can,
and give you whatever amount of data it can handle.

Even my nearly 10 year old modem does compression. As I said,
all modem protocols except the very earliest ones - before they
started using the term bps because at the time baud was more
accurate and realistic - use compression. Mine only gives me the
option to turn off IPX header compression. I don't even know
what that means, it's NOT turned off, and everything works fine.

Huh?

I an aware of the fact you can drink water but not sand, thank
you.


--
Disagreements and the usual insults expected and welcomed.

"Strings" <strings@example.net> wrote in
news:462f31bd$0$97228$892e7fe2@authen.yellow.readf reenews.net:

You know more than I but you are not being logical. A modem can
NOT handle more than, say - in my case - 3500bps of data. Of
course, if that data was compressed to begin with or if it is
sent using yEnc, I will get it more of it in less time. And
certain text files can be compressed by 85% or more as anybody
who has ever zipped one knows. A bmp file will take forever
while a visually equivalent jpg sent using yEnc will take 1/15
of the time.

All that, however, has NOTHING to do with the subject. A modem
can only do SO MUCH with ANY data it receives.


--
Disagreements and the usual insults expected and welcomed.

thanatoid <waiting@the.exit.invalid> wrote:
CBFalconer is correct, it matters.

Not necessarily.

Ten years ago is when modems ceased to be a significant factor
in computer networking for consumers. That is the mere end of
their history.

Keep in mind that CBFalconer was a grizzled old fart even then,
never mind now! ;-)

But, seriously, he is correct about modems that optionally do
not compress.

The term bps was in use, even then. M-ary encoding schemes have
existed since at least the 1930's. Keep in mind that the PCM
encoding used in digital telephony was fully specified by the
mid-1940's.

The term "baud" was used with FSK modems because it is correct.
But the first modems introduced as consumer products where FSK
modems, and when consumer products moved to PSK using multiple
bits per symbol the consumer advertizing continued to use the
term "baud" because consumers were familiar with it.

It was not a technically correct usage, and it had not been used
in describing the many earlier professional modems that used the
same techniques.

I suspect that your modem allows non-compressed data transfers
too.

Disabling compression is not commonly used, but is typically
available as a configuration option.

Are you aware that enabling compression increases latency? If
your application is sensitive to latency and the data is not
compressable, the most reasonable modem configuation is to
disable compression.

--
Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@apaflo.com

I think folks are arguing two different things here...

Physically, a 56k modem can download at max, with a perfect line at 53kbit
per second. (53k here in Canada and US. Not sure if other areas limit power
the same as here). That is the raw data between modems.

Data can be transmitted THROUGH the modem much faster if the data is easily
compressed WITHIN the modem (yes, dialup modems do compression on the fly).
For example, send a large text file full of space characters only and the
modem will squash that data BEFORE putting it on the line to the other end.
While the data stream never goes above the 53kbps limit between the modems,
the actual data throughput is much higher.

If you want to test efficiency, create a text document, about one megabyte
in size, that contains nothing but space characters. This file contains
8388608 bits of data. After adding a start bit, parity bit and stop bit you
end up with 11534336 bits of data. At 100% efficiency (which will never
happen), this file could be dowloaded in 217.6 seconds.

Zip this file using your compression ratio of choice. This file should be a
LOT smaller than the original.

At this point you should have "bigfile.txt" and "bigfile.zip"

Put both of these files onto webspace on the internet somplace and try to
download them with your 56k modem. What kind of download speeds and download
times do you actually see?

.... and there is one other item to take into consideration here. A 56kbit
modem will ONLY run faster than 33.6kbit if the answering end of the system
is digital. It can't be an analog phone line, etc. This means you'll never
get 56kbit dialing between PC modems.

"Noozer" <dont.spam@me.here> wrote:
What is required is that there be only one codec in the
connection. It makes no difference which side answers and which
side originates.

However, your conclusion that 56Kbps cannot happen "dialing
between PC modems" is correct, because there is necessarily a
requirement for two codecs if each "PC modem" is a modem
connected to a POTS telephone line. That is a function of the
POTS lines, not of a "PC modem", unless of course you define it
to be a modem that only uses a POTS line, as opposed, for
example, to being an ISDN modem). That's obviously a lot of
hair splitting... :-)

The "analog phone line" is not actually used as an analog phone
line when there is only one codec. Instead the codec becomes a
digitial level shifting mechanism (which converts digital PCM to
digital PAM, and there is nothing analog about the data! If it
is, however, decoded by a 2nd codec... it becomes a
"quasi-analog" signal, and can no longer be used in any way that
will produce a bit rate approaching 56Kbps.

--
Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@apaflo.com

floyd@apaflo.com (Floyd L. Davidson) wrote in
news:87zm4w8djo.fld@barrow.com:

Thanks for a comprehensive and polite reply...

<SNIP>

Maybe. I really don't care. I can't believe that after
complaining about the OP, I have become the main element in
continuing the discussion.

<SNIP>

It will be years until NO ONE on the PLANET uses a dial-up
modem. I have no intention of even upgrading to a 56 from my
33.6 until the 33.6 bites it - I realize I am a member of an
EXTREME minority, of course.

But MANY people still use modems. People in highly
industrialized/advanced nations (I read once Holland has every
home wired for BB, ergo the joys of alt.binaries.boneless etc)
are not the majority.

<SNIP>

There may be such things, I am not aware of them, mine has NO
options except the IPX header compression on/off, and I don't
care.

<SNIP>

Yes, I knew most of that...

Yes, I knew that...

Well if it does, it requires a magic spell since there are no
options in any of the settings and there is no mention of it in
the manual (which I have read EVERY word of).

Maybe on some modems - not on mine ("U.S. Robotics Sportster
Voice 28.8 Faxmodem with Personal Voice Mail" / "28.8 v.34 and
33.6 Faxmodem with Speakerphone and Personal Voice Mail"). It
mentions "compression" only when mentioning what V.42 bis/MNP 5
stands for (and which it supports).

I don't know what latency is and my modem does not allow for
disabling compression. (Just checked what it is. OK. If I really
cared about speed, I *would* have BB wouldn't I? But I'm in no
hurry, for anything.)

{I once ran about 6 on-line hours of "Modem Doctor" (or
something) tests (during which it dialed my ISP at least 500
times) and it told me my modem was operating at 99.8 efficiency.
Good enough for me.}

Cheers and thanks for clarifying a few things for me.

P.S.
One of the pages on latency
(http://rescomp.stanford.edu/~cheshir...s/Latency.html) says:

"The solution is easy. You can get two telephone lines, and use
them together in parallel, giving you a total of 66kbit/sec. If
you need even more you can get ten telephone lines, giving you
330kbit/sec. Sure, it's expensive, and having ten modems in a
pile is inconvenient, and you may have to write your own
networking software to share the data evenly between the ten
lines, but if it was important enough to you, you could get it
done. It may not be cheap, but at least it's possible."

So, not that the OP is around anymore, but apparently one COULD
use one thousand modems at once. Unless the author of the page
is wrong.

Sigh.


--
Disagreements and the usual insults expected and welcomed.