Re: Equipment for receiving MSG-1

Alan Sewards <alan.sewards@...>

You have made my point - we don't *know* how much a second machine is
really needed. We have lots of armwaving but few facts. As it makes a big
difference when looked at in terms of thousands of users buying and
maintaining equipment, it might be a good thing if we got it right ahead of
time? (IMHO)

Best regards - Alan

----- Original Message -----
From: "Guy Martin" <agm@...>
To: <MSG-1@...>
Sent: Monday, September 22, 2003 10:44 AM
Subject: Re: [MSG-1] Equipment for receiving MSG-1

I don't believe the Rx machine needs to be anything special, I use an Abit
KT7 with a Duron 800 and its just cruising along consuming minimal
resources. However when data arrives it does need to deal with it instantly
and not be held up by MDM decompressing or saving, it needs a 'non maskable
interrupt' in effect and MDM provides a tick box to do just this. I use a
Gigabyte GA7VKML with an Athlon 1200 in the display machine and the
processor goes to 100% when MDM is dealing with data and uses about 322/512
Mb of RAM. I really need a faster machine for this application although
having said that it does work ok with no dropped segments. A serious spec
machine may function well as a single unit and a lower grade machine could
receive data and decompress/display alternately. It is always important to
turn off services that you don't need and which consume resources as vanilla
installs come with everything turned on. Regarding XP, I'm a Win2K man and
not yet happy with XP when u!
sed outside of MS products.

Cheers, Guy

----- Original Message -----
From: Alan Sewards
To: MSG-1@...
Sent: Monday, September 22, 2003 8:17 AM
Subject: [MSG-1] Equipment for receiving MSG-1

While John is busy collating the inputs from those who have participated
the MSG-1 dissemination trials, I thought it might be of interest to
question some of the basic assumptions. The main one of these is that
needs to have two computers, one for the reception function and one for
decoding. Because I only had one, I had to make do with it for all
functions, and in doing so I expected I would lose some of the data.
However, as time has gone on, I have begun to realise that any data that
might have been lost due to possible overload of the single computer is
insignificant compared with data lost from other sources associated with
dissemination trials. In fact, I did a test overnight where I stopped
MSG DM and then restarted it at 0500UTC, so it had about 12 hours of
captured data to process, which is does in what I call 'catch-up' mode,
is running flat out with the CPU utilization running around 80-100% on a
continuing basis. It took about two hours to process all the stored data
and, as far as I could tell by a visual inspection of the thumbnails,
were no missing segments in either the period of stored data or in data
captured during the catch-up processing. This leads me to suspect that a
single computer is all that is needed for MSG reception, provided of
that it has sufficient power. I should note here that I only collect
of the 12 HRIT channels, all the LRIT and FSD, so I do not have a full
I may try an experiment of collecting the full set and see if that makes
difference. My computer is nothing special, 1800 Athlon, with 512 MB.
Because I have had great difficulty in getting two computers running
Windows XP to network properly, I am seriously considering departing
the two-computer concept and sticking with a single one for MSG, and I
pretty confident that it will work. Anyone else had similar thoughts?
we do some serious tests to find out if two computers are really

Best regards - Alan

Alan Sewards
email: alan.sewards@...
web site:

Yahoo! Groups Sponsor

To unsubscribe from this group, send an email to:

Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

---------- - Messaging for educators.

To unsubscribe from this group, send an email to:

Your use of Yahoo! Groups is subject to

Join to automatically receive all group messages.