OpenBCM V1.13 (Linux)

Packet Radio Mailbox

DB0FHN

[JN59NK Nuernberg]

 Login: GUEST





  
PA2AGA > HDDIG    12.11.99 13:58l 174 Lines 7218 Bytes #-9687 (0) @ EU
BID : HD_99_287D
Read: GUEST
Subj: HamDigitalDigest 99/287D
Path: DB0AAB<DB0KFB<DB0ZKA<DB0ABH<DB0SRS<DB0ROF<DB0ERF<DB0SHG<DB0SM<PI8DAZ<
      PI8GCB<PI8WNO<PI8HGL<PE1NMB<EA7URC<PE0MAR<PI8VNW
Sent: 991112/0900Z @:PI8VNW.#ZH2.NLD.EU #:16218 [HvHolland] FBB7.00g $:HD_99_28
From: PA2AGA@PI8VNW.#ZH2.NLD.EU
To  : HDDIG@EU

Received: from pa2aga by pi1hvh with SMTP
	id AA23035 ; Fri, 12 Nov 99 08:08:56 UTC
Received: from pa2aga by pa2aga (NET/Mac 2.3.67/7.5.3) with SMTP
	id AA00016810 ; Fri, 12 Nov 99 07:17:01 MET
Date: Fri, 12 Nov 99 07:04:09 MET
Message-Id: <hd_99_287D>
From: pa2aga
To: hd_broadcast@pa2aga
Subject: HamDigitalDigest 99/287D
X-BBS-Msg-Type: B


>> Years ago, when I still visited the yearly Frankfurt/Darmstadt packet
>> meetings, I always discussed with the locals about that.  Our standpoint
>> has always been "use PC hardware at the node.  you get free upgrades in
>> performance over the years, and you can't beat the price/performance".
>> Theirs was always "PC hardware is not reliable, consumes too much power,
>> is too large" etc.
>This is still true today I think. The problem is the availability of
hardware.
>If the Flexnet group publishes a new design today, it is likely that it
>will be obsolete some month later, because the manufacturer doesnīt ship a
>certain part any more. This makes it very difficult.

On the other hand, when using "Industry standard PC parts", the new parts
usually are sufficiently compatible.  Over the 10-year period that our
SCC card design exists now, the typical PC motherboard available for a
node has moved up from a 8088 at 4.77 MHz with 640K RAM (comparable with
the 6809 on the Flexnet boards of those days, but more memory) to a
Pentium at 1XX MHz, probably about 100 times faster.  We had to do no
development for that at all.
(the above is assuming that a HAM will donate his "previous" computer
system for use at the node, once he has bought a new system.
state-of-the-art systems are of course even faster)

Only after 10 years the problem is emerging that motherboards not always
have enough ISA slots to take enough 4-channel SCC cards to build a node.
A PCI card could overcome that hurdle.

> PC hardware *is*
>unreliable and consumes much power which is a problem if your node is
>installed at a location such as TV towers or high mountains "hill station".
>I know several nodes which can only accessed every eight weeks or such.
>You canīt put a linux box there, if it hangs in fsck every time the power
>fails.

But then, I have operated PC-based nodes with uptimes for well over a year,
and Linux boxes also don't have that problem.  Fsck can be configured to
repair everything it possibly can without user intervention.
I have no personal experience with the packet radio stuff in Linux, only
heard that it is a bit unreliable, and seen in the design that it leaves
a lot to desire.  Maybe the people that have concentrated on packet
software for dedicated cards in Germany should have a look at it.

Power failures have not been a problem for us, but when they are it would
be possible to use a UPS.
Also, using a PC does not necessarily mean using a diskdrive and/or a
read/write filesystem.  When you want to compare reliability at the same
functionality level, you should run the PC with a read-only floppy only
used to boot the system, or even use an "EPROM drive" card.
A Linux system with a read/write mounted harddisk may be more fragile,
but also is much, much more versatile than an EPROM-based 6809 board.

In the nodes that I have seen, power is usually not paid by the kWh,
and besides that it would be interesting to see how much difference there
really is between the *total* power drawn by a node running on a 486
board with 8-12 SCC channels plus a floppy disk or EPROM drive, and
a similar-functionality rack of 13 flexnet cards.  Same for the MTBF.
Don't compare a flexnet rack with a PC doubling as a BBS.

>> So maybe it is coming our way now? :-)
>This is likely (IMHO), but not because it is the technical superior
>solution.

My standpoint is that it is technically similar, and superior in
extensability and upgradability.  Especially when development resources are
limited.

>> BTW: I got an e-mail recently from someone asking about the availability of
>> a PCI SCC card (presumably because he has no ISA slots available).  Is that
>> card available commercially?
>Iīve seen it at the Weinheim/Mannheim UHF convention, I donīt know if it
>is available commercially. It is very new and there are only linux drivers
>as far as I know. Perhaps Jens is reading this and can comment himself.

Probably this is not what that person is looking for...
What he needs is a Z85230-based card that works with all the wellknown
programs that include an SCC driver (usually my driver or a derivative),
but for a PCI slot.
Of course, a much better card, with busmastering DMA etc, could be designed
for PCI.  It would be the superior solution for higher speeds and for
nodes with many ports.  That is probably what Jens has done.

Rob
-- 
+----------------------------------+--------------------------------------+
| Rob Janssen     pe1chl@amsat.org | WWWhome: http://www.pe1chl.demon.nl/ |
| AMPRnet:     rob@pe1chl.ampr.org | AX.25 BBS: PE1CHL@PI8WNO.#UTR.NLD.EU |
+----------------------------------+--------------------------------------+
>.

------------------------------

Date: Tue, 9 Nov 1999 17:49:12 -0000
From: "Good Buddha" <postmaster@nospam.demon.co.uk>
Subject: Help with RFC

Mike > wrote in message <808mcp$t5k@newton.cc.rl.ac.uk>...
>
>Steve Sampson <ssampson@usa-site.net> wrote in message
>news:s2cbanpi3e647@corp.supernews.com...
>> I think once you put the word out, that you are looking for a piece
>> of test equipment, the word gets around.  Someone may have an
>> older piece that they are thinking about upgrading.  Another way is
>> to go to government surplus sales (DRMO).  Here at Tinker we have
>> a sale every couple of weeks
>
>You really are living in a different world ! Or perhaps should not  post
>this to UK.radio.amateur.


RUBBISH! The post is about _REAL_ Ham Radio, and not the
consumerist abusive style that has come largely from the CB
fraternity.





>.

------------------------------

Date: 10 Nov 1999 04:15:26 GMT
From: andy@acsteele.demon.co.uk
Subject: Help with RFC

In uk.radio.amateur Good Buddha <postmaster@nospam.demon.co.uk> wrote:
: Mike > wrote in message <808mcp$t5k@newton.cc.rl.ac.uk>...
:>
:>Steve Sampson <ssampson@usa-site.net> wrote in message
:>news:s2cbanpi3e647@corp.supernews.com...
:>> I think once you put the word out, that you are looking for a piece
:>> of test equipment, the word gets around.  Someone may have an
:>> older piece that they are thinking about upgrading.  Another way is
:>> to go to government surplus sales (DRMO).  Here at Tinker we have
:>> a sale every couple of weeks
:>
:>You really are living in a different world ! Or perhaps should not  post
:>this to UK.radio.amateur.


: RUBBISH! The post is about _REAL_ Ham Radio, and not the
: consumerist abusive style that has come largely from the CB
: fraternity.

Do Hams have a monopoly on interests in electronics?

Is it only CBers that use boxes they dont fully understand ?

Didn't think so.

The post was about real _AMATEUR_ radio, I think we both recognise that
,but in different ways.

You can send in your contribution to this digest by
sending an e-mail to: hd-group@pa2aga.ampr.org
or (via BBS-net)  to: hdaga@pi8vnw.#zh2.nld.eu




Read previous mail | Read next mail


 22.05.2026 05:11:11lGo back Go up