Best V.34 modem for BAD Connections
Dileep Agrawal
best for poor quality international telephone connections? I need a
modem to exchange files with Nepal (in Asia) and the connections suffers
from noise, echoing, delays...
Thanks.
Dileep
nsx
You have to see what modem is on each end..
The same modem at two ends is the best.
If money is an issue, I suggest the USR sporster. If not, then buy the
best in the world, the PENRIL Alliance/ALX 33.6 for $600. This is much
better than USR Courier, hands down!
For more info vist the PENRIL site.
Paul O'Nolan
>
> Can anyone suggest me, from their experience, which V.34 modem is the
> best for poor quality international telephone connections? I need a
> modem to exchange files with Nepal (in Asia) and the connections suffers
> from noise, echoing, delays...
>
good enough. I tried to get a pair of Motorola v3400s going at better
than 2400 from Entebee (Uganda) to the Hague (Netherlands) earlier this yr.
Couldn't be done. From what I've read and discussions here the Telebit
line's proprietary protocols may well be the best on bad lines. AT&T
Comsphere and high end Motorola modems are well regarded as well as
Telebit's. USR's Sportster modems ?... <guffaw>
bi...@mix.com
> From what I've read and discussions here the Telebit
> line's proprietary protocols may well be the best on bad lines.
Definitely true, even though PEP is a half-duplex protocol.
Billy Y..
Richard L. Stuart
here. The most recent test of modems has been reported in the 15 Nov.
95 issues of Network Computing Mag. A very extensive test using TSB37a
as well as additional test. The test covered connectivity and
through-put. The clear winner, by a long shot, was Penril. The article
said: "The Penril knocked the socks off the competition" which included
most if not all of the brands discussed here (comp.dcom.modems). They
have a weg page: www.penril.com. They are also running an Internet
Special.
The text of the article can be found at: techweb.cmp.com. It is true
that both ends have to be good V.34 implementations and the connection
is a major factor. However, you must start by making sure you have the
best performing modem on your end or you will just be sorry and find
yourself having to buy a good replacement. There are also many people
expressing (here) their regrets on buying a cheap implementation. All
V.34's are not alike, it seems. You will get what you pay for. It's
like buying a cheap FM radio or an excellant FM Receiver with Dolby
Noise reduction, Surround Sound etc. They are both compatible with the
FM Transmitter, however the reception on one is far superior to the
other. Better connections, clear receptions over greater distances etc.
etc.
Good Luck
Dick Stuart
In <NEWTNews.8186675...@pip.inter.nl.net> Paul O'Nolan
John Sanger
USR TS did an analysis of the PCMCIA modem and determined that it was
working within the specifications of the design. They further
determined that the line noise was what was preventing me from staying
logged in for longer than about 3 minutes. The blers were very high and
the disconnect was almost always due to "retransmits exceeded the limit".
That latter is the data frame being retransmitted and the limit is 12
requests.
When stressing the line with an zmodem transfer of Netscape, the line
really showed an increasing number of blers and then at the limit of the
retransmit requests it disconnected.
The thing that gets me is that the USR TS stayed connected to my ISP's
POP for over 10 minutes without any problems staying connected and he
was calling from Illinois to Los Angeles.
So modifying the active profile to the suggested changes from USR TS did
not improve my ability to stay connected.
I am still not able to remain on for more than about 3 minutes.
I have reinstalled my PPI PractiCard 144 modem which has never failed me
on any line situation...... ;^)
I know that others have been able to stay connected to the same ISP's
POP with USR Courier and Sportster modems and even those cheaper modems
using the Rockwell chipset has stayed connected but this USR PC-Card
v.everything modem is for the pits right now.... 8^(
I am expecting a Courier v.everything external soon and hope I do not
have the same problem with it on the lines that it must use...... 8^)
Anyone with any ideas on what might help here, please make them known.
--
Ciao!
John S. 8^{)>
tedd...@netcom.com
__
Anthony Hill
Richard L. Stuart (r-st...@ix.netcom.com) writes:
> There have been a number of such request like the one below posted
> here. The most recent test of modems has been reported in the 15 Nov.
> 95 issues of Network Computing Mag. A very extensive test using TSB37a
> as well as additional test. The test covered connectivity and
Unfortunatly it wasn't all that extensive a test. The only used
90% of the PSTN model from TSB37A, which gives a good indications of how
the modem will perform on good lines, but doesn't say much of anything
about how the modem will perform on poor lines, and they never came close
to testing it on really horrible lines.
> through-put. The clear winner, by a long shot, was Penril. The article
> said: "The Penril knocked the socks off the competition" which included
> most if not all of the brands discussed here (comp.dcom.modems). They
Penril does seem to have a decent product (albeit rather
expansive), I'd be interested in seeing a real comprehensive test of the
modem as compared to some others.
> The text of the article can be found at: techweb.cmp.com. It is true
> that both ends have to be good V.34 implementations and the connection
> is a major factor. However, you must start by making sure you have the
> best performing modem on your end or you will just be sorry and find
> yourself having to buy a good replacement. There are also many people
Unforutnatly buying a good modem for your end doesn't always
garuntee best performance. If the other end is running a cheap crappy
Rockwell Glue 'n Go jobbie, chances are that the best modems (such as
AT&Ts, Penrils, Motorolas, etc.) may actually perform WORSE then some
cheaper modems such as USRs or even something like a Surpa. This is
because interoperability can cause a lot of problems, and in general the
best modems don't work so well with the worst modem.
Anthony
>>> Can anyone suggest me, from their experience, which V.34 modem is
> the
>>> best for poor quality international telephone connections? I need a
>
>>> modem to exchange files with Nepal (in Asia) and the connections
> suffers
>>> from noise, echoing, delays...
>>>
>>The best modems on the market will not help if the lines physically
> aren't
>>good enough. I tried to get a pair of Motorola v3400s going at better
>>than 2400 from Entebee (Uganda) to the Hague (Netherlands) earlier
> this yr.
>>Couldn't be done. From what I've read and discussions here the Telebit
>>line's proprietary protocols may well be the best on bad lines. AT&T
>>Comsphere and high end Motorola modems are well regarded as well as
>>Telebit's. USR's Sportster modems ?... <guffaw>
>>
>
--
Anthony Hill | an...@FreeNet.Carleton.CA
Ken Adler
>
>You have to see what modem is on each end..
>The same modem at two ends is the best.
>If money is an issue, I suggest the USR sporster. If not, then buy the
>best in the world, the PENRIL Alliance/ALX 33.6 for $600. This is much
>better than USR Courier, hands down!
>For more info vist the PENRIL site.
>
Dileep:
IF you want to contact Penril in Asia, please email me at
ad...@hk.super.net.
BTW..We do a brisk business in Asia and according to an upcoming IDC
survey Penril is basically tied as the 2nd largest supplier to PTT's in
China. Though I do travel too much as it is ... I would love to go do a
"site survey" in Nepal :-) (i'd sneek some trekking in... although
would probably wait until spring <:-) )
Take care.
Ken Adler
General Manager, Asia/Pacific
Penril Datability Networks
>
>
>In article <4ad6t1$k...@paladin.american.edu>, dagr...@bates.edu wrote:
>
>> Can anyone suggest me, from their experience, which V.34 modem is the
>> best for poor quality international telephone connections? I need a
>> modem to exchange files with Nepal (in Asia) and the connections suffers
>> from noise, echoing, delays...
>>
>> Thanks.
>>
>> Dileep
nsx
Your telling me that hooking up to shit modems is better than hooking up a
good one to a bad one? give me a break....
We (some sysops) have just completed some test (real world) on stupid FX
line and noisy lines & so forth with many modems...(USR courier, sporster
,Microcom, PP-LCD, Penril ALX, AT&T, supra, telebit,
motorola)...connecting to shit modem and good modems everywhere....Guess
what....The USR courier came 3rd, The PP-LCD second, the PENRIL 1st...
The USR does not react fast at all if any, if it retrain down, it will not
go up fast at all maybe after 2 minutes or more....The PENRIL changes so
reapidly that it one logon to an ISP, they will see why this modem is the
choice...even at 14.4, the PENRIL seems to be getting info(not downloads)
much faster than a 28.8...
vi...@ibm.net
In <nsx-091295...@nsx-ppp.serv.net>, n...@serv.net (nsx) writes:
>
>You have to see what modem is on each end..
>The same modem at two ends is the best.
>If money is an issue, I suggest the USR sporster. If not, then buy the
>best in the world, the PENRIL Alliance/ALX 33.6 for $600. This is much
The best in the world???...In what conditions, in what probe????...
Is the same case of the Computers Shopper's Magazine test?...
Is based in the 230.400 DTE-DCE connection?...
Is based only in the speed to high compressible files and 230.400 bps
DTE-DCE connection?...
I do not need a 230.400 bps conection, for what I have not a Hardware
thats support this, and I have not a software that support this!...
But I know that the 230.400 bps DTE-DCE speed conection is very good
only for Magazine Test!....=;))))
Now, you can say:
What Chip-Set is used in the Penril modems????...
Have any Cellular secure connection protocol???...What?...=;)))
>better than USR Courier, hands down!
>For more info vist the PENRIL site.
>
>
>
nsx
please read the Network computing NOV 15th issue about those modems..
After that & if you're willing to spend $$$ then talk about it...
vi...@ibm.net
Hi, there!...
>> Unforutnatly buying a good modem for your end doesn't always
>> garuntee best performance. If the other end is running a cheap crappy
>> Rockwell Glue 'n Go jobbie, chances are that the best modems (such as
>> AT&Ts, Penrils, Motorolas, etc.) may actually perform WORSE then some
>> cheaper modems such as USRs or even something like a Surpa. This is
>> because interoperability can cause a lot of problems, and in general the
>> best modems don't work so well with the worst modem.
Yes!...It is true. Some hi-end modems will have some problems to connect
with cheapest modems. I found this with the E-Tech BULLET 100, and the
cheap RPI 14K4 modems. Is a good probe if you have a public BBS system.
I am very happy for this. The COMSPHERE 3820 plus never have a failed
connections, even in the case of a cheap RPI modems...
I never found any problem of "inter-operability " with the "old" COMSPHERE
3810, and now, the COMSPHERE 3820 plus, but I found this in other brands
of good modems...
Best regards.
Vicente Sanchez. EB4BSQ.
AKA vi...@ibm.net
SysOp de MERCURIO BBS
Ken Adler
>Hi, there!...
>
>In <nsx-091295...@nsx-ppp.serv.net>, n...@serv.net (nsx) writes:
>>
>>You have to see what modem is on each end..
>>The same modem at two ends is the best.
>>If money is an issue, I suggest the USR sporster. If not, then buy the
>>best in the world, the PENRIL Alliance/ALX 33.6 for $600. This is much
>
>The best in the world???...In what conditions, in what probe????...
I think NSX was making his comments based on the Network World
comparison. Thier testing was done based on TSB37a and TSB38. We
(penril) have also done this testing inhouse using the complete
procedures and have come up with similar results. Basically to answer
your question, on clean lines you will see that we can move more data
than any other modem. At the other end of the scale Penril can
typlically hold onto noisy lines where other can not. I know this sounds
biased coming from me but this is what my personal experience has shown
me and what the industry standard testing has confirmed.
>
>Is the same case of the Computers Shopper's Magazine test?...
I am not familair with this article.
>Is based in the 230.400 DTE-DCE connection?...
>Is based only in the speed to high compressible files and 230.400 bps
>DTE-DCE connection?...
TSB tests specify 5 different files that range from highly compressable
to totally random (non compressable) files
>
>But I know that the 230.400 bps DTE-DCE speed conection is very good
>only for Magazine Test!....=;))))
>
no, not really. As the need for shoving graphics around becomes more and
more important we find more people putting in high speed asynch cards and
pumping data at 23.4kbps. These cards are getting more and more
affordable
>Now, you can say:
>
>What Chip-Set is used in the Penril modems????...
PEnril does not usa a chipset. We are a DSP based modem and all the
software (datapump and controller) are completely written in-house at
Penril. This is what allows us to differentiate ourselves from the
competition. IT allows us to optimise the code in the manner that we see
fit (such has for noisy telephone lines)
>
vi...@ibm.net
In <4b10bu$8...@tst.hk.super.net>, Ken Adler <ad...@hk.super.net> writes:
>vi...@ibm.net wrote:
>>Hi, there!...
>>
>>In <nsx-091295...@nsx-ppp.serv.net>, n...@serv.net (nsx) writes:
>>>
>>>You have to see what modem is on each end..
>>>The same modem at two ends is the best.
>>>If money is an issue, I suggest the USR sporster. If not, then buy the
>>>best in the world, the PENRIL Alliance/ALX 33.6 for $600. This is much
>>
>>The best in the world???...In what conditions, in what probe????...
>
>
>I think NSX was making his comments based on the Network World
>comparison. Thier testing was done based on TSB37a and TSB38. We
>(penril) have also done this testing inhouse using the complete
>procedures and have come up with similar results. Basically to answer
>your question, on clean lines you will see that we can move more data
>than any other modem. At the other end of the scale Penril can
>typlically hold onto noisy lines where other can not. I know this sounds
>biased coming from me but this is what my personal experience has shown
>me and what the industry standard testing has confirmed.
>>
Well. I need to see this article. I have very interest in any probe about
modems...I never say that the Penril's modems are bad. I know the old
line of Penril V.32bis and V.32terbo. They are a good Solid-Rock stability
and never-failed solution...I know that all hi-end modems, like Racal-Decca,
(Now Zip-Comm???...:-)...), Motorola (Codex), Paradyne, and Penril, of course,
are a feature loaded and good investiment. I ever say about the great
differences with the others brands...
>>Is the same case of the Computers Shopper's Magazine test?...
>
>I am not familair with this article.
>
>>Is based in the 230.400 DTE-DCE connection?...
>>Is based only in the speed to high compressible files and 230.400 bps
>>DTE-DCE connection?...
>
>TSB tests specify 5 different files that range from highly compressable
>to totally random (non compressable) files
Yes!...But now, with the high compression systems used in the V.34 modems,
(And even the old V.32 terbo modems)
the 115200 speed DTE-DCE connection is a "brake" in this probes. And the
advantage is for the modems connect at 230K4...This advantage, in my opinion,
is refflected in the overall statistics and is a not true speed advantage...
Even in the case of that 230400 bps DTE-DCE connection is a good
feature...Is not significant a advantage of power of processing in the
CPU-DSP or data-pump, is only the double of speed in the DTE-DCE
connection...
>>
>>But I know that the 230.400 bps DTE-DCE speed conection is very good
>>only for Magazine Test!....=;))))
>>
>
>no, not really. As the need for shoving graphics around becomes more and
>more important we find more people putting in high speed asynch cards and
>pumping data at 23.4kbps. These cards are getting more and more
>affordable
>
>
>
>>Now, you can say:
>>
>>What Chip-Set is used in the Penril modems????...
>
>PEnril does not usa a chipset. We are a DSP based modem and all the
>software (datapump and controller) are completely written in-house at
>Penril. This is what allows us to differentiate ourselves from the
>competition. IT allows us to optimise the code in the manner that we see
>fit (such has for noisy telephone lines)
>>
Well, but what DSP and what CPU. What is the composition of the data-pump,
ant the clock thats its runs???....
You trust that a two good modems, in the same *very good* line, at same
speed line connection and at same speed DTE-DCE connection, in the transfer
of the same file, all the same!...You hope great speed differencies???....
I trust that no...
Best regards.
(Maybe, I will probe a Penril in the next modem purchase)...
Chris Mauritz
: vi...@ibm.net wrote:
: >Hi, there!...
: >
: >In <nsx-091295...@nsx-ppp.serv.net>, n...@serv.net (nsx) writes:
: >>
: >>You have to see what modem is on each end..
: >>The same modem at two ends is the best.
: >>If money is an issue, I suggest the USR sporster. If not, then buy the
: >>best in the world, the PENRIL Alliance/ALX 33.6 for $600. This is much
: >
: >The best in the world???...In what conditions, in what probe????...
: I think NSX was making his comments based on the Network World
: comparison. Thier testing was done based on TSB37a and TSB38. We
: (penril) have also done this testing inhouse using the complete
: procedures and have come up with similar results. Basically to answer
: your question, on clean lines you will see that we can move more data
: than any other modem. At the other end of the scale Penril can
: typlically hold onto noisy lines where other can not. I know this sounds
: biased coming from me but this is what my personal experience has shown
: me and what the industry standard testing has confirmed.
: >
: >Is the same case of the Computers Shopper's Magazine test?...
: I am not familair with this article.
: >Is based in the 230.400 DTE-DCE connection?...
: >Is based only in the speed to high compressible files and 230.400 bps
: >DTE-DCE connection?...
: TSB tests specify 5 different files that range from highly compressable
: to totally random (non compressable) files
: >
: >But I know that the 230.400 bps DTE-DCE speed conection is very good
: >only for Magazine Test!....=;))))
: >
: no, not really. As the need for shoving graphics around becomes more and
: more important we find more people putting in high speed asynch cards and
: pumping data at 23.4kbps. These cards are getting more and more
: affordable
: >Now, you can say:
: >
: >What Chip-Set is used in the Penril modems????...
: PEnril does not usa a chipset. We are a DSP based modem and all the
: software (datapump and controller) are completely written in-house at
: Penril. This is what allows us to differentiate ourselves from the
: competition. IT allows us to optimise the code in the manner that we see
: fit (such has for noisy telephone lines)
This is all very nice. Great modems aren't much good to
people if they aren't affordable or if you can't find them
anywhere. What is the going rate for Penril products? Where
can one find them?
Regards,
Chris
--
Christopher Mauritz | For info on internet access:
ri...@mordor.com | finger/mail in...@ritz.mordor.com OR
Mordor International | http://www.mordor.com/
201/212/718 internet access | Modem: (201)433-7343,(212)843-3451
nsx
BINGO, Ken you have said what I need to say. Al-cheapo modem will always
be alcheapo....
What revision number is the latest one for PENRIl ALX 33.6K?
Dale Shuttleworth
: >Is based in the 230.400 DTE-DCE connection?...
: >Is based only in the speed to high compressible files and 230.400 bps
: >DTE-DCE connection?...
:
: TSB tests specify 5 different files that range from highly compressable
: to totally random (non compressable) files
: >But I know that the 230.400 bps DTE-DCE speed conection is very good
: >only for Magazine Test!....=;))))
: no, not really. As the need for shoving graphics around becomes more and
: more important we find more people putting in high speed asynch cards and
: pumping data at 23.4kbps. These cards are getting more and more
: affordable
Do you really believe that people are going to send lots of uncompressed
graphics files over a modem connection (particularly in Europe where you
pay for connection time)?
Whilst I agree that you have to sell modems, I am suprised that you are
unwilling to recommend a copy of PKZIP to your customers. Another case
of the triumph of commercial interests over common sense :-)
Dale.
--
******************************************************************************
* Dale Shuttleworth *
* Email: da...@giskard.demon.co.uk *
******************************************************************************
Ken Adler
>This is all very nice. Great modems aren't much good to
>people if they aren't affordable or if you can't find them
>anywhere.
As they say in New York "This should be the worst of my problems."
>What is the going rate for Penril products?
I am not sure what the street price in the USA is (as I cover asia) but the Network
computing article has us listed as US$595. Let me also point out that Penril makes
Remote access servers that have modems built in. Sort of like having a terminal
server and modem rack put together. The per port cost of this is very low when
compared to buying seperate units. It is also very low when compared to other
integrated solutions like USR Total control, etc
>Where can one find them?
We do have a sales and R&D office in the "201" area code area right across the
street from Brendon Byrne Arena. I'd recommend that you call Jim Gallagher at
1-800-473-6745 (800-4-PENRIL). I am sure that he would work a way to get them to
you at a price that you feel comfortable with .
Take care
Ken Adler
General Manager, Asia/Pacific
Penril Datability Networks
>
Ken Adler
<SNIP>
> As the need for shoving graphics around becomes more and
>: more important we find more people putting in high speed asynch cards and
>: pumping data at 230.4kbps. These cards are getting more and more
>: affordable
>
>Do you really believe that people are going to send lots of uncompressed
>graphics files over a modem connection (particularly in Europe where you
>pay for connection time)?
>
No, not really.
I think you are missing the point.
What I was referring to was not so much shipping static graphic files around
(compressed or uncompressed). What I had in mind was other bandwidth intensive
content such as Web pages, real time medical images, real time remotely sensed
imagery, audio and video streams, etc.
These are the interesting cutting edge applications that are being implemented by
serious users of datacom and internetworking equipment. This is the market
that Penril focuses on in our design.
>Whilst I agree that you have to sell modems,
..and dial on demand routers, and remote access products, and voice compression and
data muxes...etc
>I am suprised that you are
>unwilling to recommend a copy of PKZIP to your customers.
> Another case of the triumph of commercial interests over common sense :-)
Now, now... lets not jump to conclusions. Please give me the benefit of the doubt.
:-)
Pkzip is wonderful stuff but in my 12+ years of experience on the
net I haven't yet figured out a way to use it to compress real time
graphic (or other type) info which was the point of my comment. PKzip is really only
good if your data is sitting still. Life gets exciting when bits are fresh and
dynamic. :-)
John Sanger
: I am being killed by line noise or so it seems.... ;^)
Well, I got my Courier external v.everything and it is rock solid. I
have also replaced the PC-Card v.everything modem with another PC-Card
v.everything and this one also works rock solid. So it seems that the
first PC-Card was a lemon. It is being returned to USA-Flex and
hopefully they will return it to USR as defective. USR needs to delve
into this one as they checked it out via a login to it and found it
funcitoning within specs.
Christian Weisgerber
> procedures and have come up with similar results. Basically to answer
> your question, on clean lines you will see that we can move more data
> than any other modem.
This doesn't seem plausible. On a clean line, no bit errors, any V.34
modem will move 28800bps. Are you saying that only Penril implemented a
working V.34 modem?
--
Christian 'naddy' Weisgerber na...@mips.pfalz.de
See another pointless homepage at http://www.rhein-neckar.de/~mips/.
-- currently reading: Richard Matheson, The Incredible Shrinking Man --
Christian Weisgerber
> (compressed or uncompressed). What I had in mind was other bandwidth intensive
> content such as Web pages, real time medical images, real time remotely sensed
> imagery, audio and video streams, etc.
Audio streams? ADPCM, CELP, ...
Images? Video streams? MPEG-2, ...
Penril may offer components for those applications, but I doubt it's a
standard part of their modems. Especially since choosing an optimal
compression scheme for a data stream mux'ed from several applications is
somewhat difficult.
> Pkzip is wonderful stuff but in my 12+ years of experience on the
> net I haven't yet figured out a way to use it to compress real time
> graphic (or other type) info which was the point of my comment.
> PKzip is really only good if your data is sitting still.
At least in the Unix world, there are filter type compressors (among
them gzip, which uses the same compression algorithm as PKzip) that can
compress a moving data stream. Usually either the input or the output is
sitting still, but this is a limit of application rather than the
compressors themselves.
> Life gets exciting when bits are fresh and dynamic. :-)
<Munch, munch> Bit du jour. Excellent.
vi...@ibm.net
>Ken Adler <ad...@hk.super.net> writes:
Hi, there!...
>
>> procedures and have come up with similar results. Basically to answer
>> your question, on clean lines you will see that we can move more data
>> than any other modem.
>
>This doesn't seem plausible. On a clean line, no bit errors, any V.34
>modem will move 28800bps. Are you saying that only Penril implemented a
>working V.34 modem?
>
It is true...At the same good line, at the same file, all modems
connected to 33K6 perform the same, or not????....=;)
Only a 230.400 bps DTE-DCE connection with large compressible
files can perform more in the average of probe...
And it is not a true advantage!....
Ken Adler
>Ken Adler <ad...@hk.super.net> writes:
>
>> procedures and have come up with similar results. Basically to answer
>> your question, on clean lines you will see that we can move more data
>>working V.34 modem?
Thatis not at all what I am saying. The only thing that I am saying is that if you
look at the NEtwork Computing Magazine article (get the dead tree edition as it has
the pretty charts) you will see that of the 19 modems they tested, Penril was the
only one that connected over 10kcps. The closest competitor was USR that connectd
over 9k on approx 5% of the US network. According to the article, Penril was above
9k on approx 60% of the network , over 10k on 50% of the network and over 11k on 15%
of the network (as specified by TSB27a).
Also, I am not saying that Penril is the only manufacturer that has made a working
V.34 *COMPATABLE* modem. There are zillions of people making *COMPATABLE* V.34
modems. However, with V.34 there is a big difference between being V.34
COmpatable and being FULLY COMPLIANT with Recommendation V.34. There are only
a very few modems on the market that comply fully with all that is in
Recommendation V.34 (ie. to my knowledge no chipset modem is fully compliant)
Ken
Matt Fox
>
>na...@mips.pfalz.de (Christian Weisgerber) wrote:
>>Ken Adler <ad...@hk.super.net> writes:
>>
>>> procedures and have come up with similar results. Basically to answer
>>> your question, on clean lines you will see that we can move more data
>>> than any other modem. Are you saying that only Penril implemented a
>>>working V.34 modem?
>
>
>
>Thatis not at all what I am saying. The only thing that I am saying is that
if
>you
>look at the NEtwork Computing Magazine article (get the dead tree edition as
it
> has
>the pretty charts) you will see that of the 19 modems they tested, Penril was
th
>e
>only one that connected over 10kcps. The closest competitor was USR that
thats compression.
Using artifically highly compressible data
Ken Adler
Well, NC magazine call's thier labs "Real World LAb" They actually used the tsb38
files (there are 5 in total - from totally uncompressable files to highly
compressable graphic files). They also have a chart showing specifically the
thruput on a random file (i.e. not artificially compressable).
Ken
>
Ken Adler
>Hi, there!...
>
>It is true...At the same good line, at the same file, all modems
>connected to 33K6 perform the same, or not????....=;)
I would definately have to say "NOT". There are other issues that are important
such as the "horsepower" of the hardware itself, size of the dictionary, etc.
So even if you have two modems that are using the same file and are connected at the
same line speed (i.e. 33.6 or whatever) there is still a significant chance for a
difference in the actual thruoughput. In fact this is the whole rational behind the
TSB28 test proceedure. It is intended to test the effeciency of V.42bis
compression independant of the line speed. The NEtwork computing used the TSB28
files and corelated them to the TS27a tests which looks at how a modem operates over
verious line conditions
Ken
Jacques Cardinal
and 28
Happy Hollydays,
Jacques Cardinal
Jacc...@mbr.centra.ca
Jacques Cardinal
>na...@mips.pfalz.de (Christian Weisgerber) wrote:
>>Ken Adler <ad...@hk.super.net> writes:
>>
>>> procedures and have come up with similar results. Basically to answer
>>> your question, on clean lines you will see that we can move more data
>>> than any other modem. Are you saying that only Penril implemented a
>>>working V.34 modem?
>
>
>
>Thatis not at all what I am saying. The only thing that I am saying is that if you
>look at the NEtwork Computing Magazine article (get the dead tree edition as it has
>the pretty charts) you will see that of the 19 modems they tested, Penril was the
>over 9k on approx 5% of the US network. According to the article, Penril was above
>9k on approx 60% of the network , over 10k on 50% of the network and over 11k on 15%
>of the network (as specified by TSB27a).
>
>Also, I am not saying that Penril is the only manufacturer that has made a working
>V.34 *COMPATABLE* modem. There are zillions of people making *COMPATABLE* V.34
>modems. However, with V.34 there is a big difference between being V.34
>COmpatable and being FULLY COMPLIANT with Recommendation V.34. There are only
>a very few modems on the market that comply fully with all that is in
>Recommendation V.34 (ie. to my knowledge no chipset modem is fully compliant)
>
>Ken
>
>>
>>--
>>Christian 'naddy' Weisgerber na...@mips.pfalz.de
>> See another pointless homepage at http://www.rhein-neckar.de/~mips/.
>>-- currently reading: Richard Matheson, The Incredible Shrinking Man --
>
>
against TSB38 , you usualy test it in pair, same modem brand at each end.
In the real world this is not the case, the implementation of V.42bis or
MNP5 might not be identical from different manufacturer , Therefore
results will be quite different,
when you use the modem through the PSTN running real application.
Jacques Cardinal
Jacc...@mbr.centra.ca