🚀 go-pugleaf

In article <C0Dno6.7vo@ccu.umanitoba.ca> ens@ccu.umanitoba.ca () writes:
>In <1992Dec31.212754.1@seb.se> s19184@seb.se (Michael Evans) writes:
>>I have been thinking about the same thing. I know there are Ethernet 
>>cards for various ST-models in Germany, but I think that most of them 
>>are for Megas (ie expansion bus machines).
>
>They have ACSI boards too but the prices are outrageous.
>
>>The thought that struck me the other day (it did not hurt ;-) was that 
>>there are many "pocket" LAN adapters available for portable PCs. These 
>>plug onto the parallel port and supply a Token Ring or Ethernet 
>>(Thinwire or 10BaseT) connection.

There is a problem with connecting these LAN adapters to the Atari ST.
As far as I figured it, the Atari is missing a couple of signals that
are available on the PC's printer port. So I doubt you would be able to
have them run on an ST. It might be different with the Falcon which is
said to have more signals on the printer port. Nonetheless the speed
that is achievable by the printer port (which is driven by the sound
chip) is very limited. Even if handshaking data is reduced to mininum
and if you use optimized assembler code to access this port, you will
probably only get less than 100KB/s. So using a port (such as the
ACSI-DMA port) that supplies built-in handshaking information will be
needed to do the trick.

>A better way would be to use a MAC SCSI-ethernet board with an
>ICD ACSI-SCSI connector. This would be faster, but the same
>problem with lack of software drivers exists. Now if you
>could get software only from the German companies at a reasonable
>price...
>
>Werner

A couple of friends of mine have been programming a network for the ST
for quite a while. The software is running very reliably --- in fact
this system has by now been in use for more than a year at my former
high-school and the computer science teachers are very satisfied with
it. Unfortunately there are still problems with the proprietary
hardware which will only allow for cables that are no longer than
30feet. Though this hardware is comparably cheap, which would allow
for selling the network as a low-cost alternative to products such as
PAM's Net or BioNet, my friend are on the long term looking for a
solution that is based on ethernet. As of now, none of them have been
able to find any of those SCSI<->Ethernet boards you have been talking
about. So, if you could provide me with the address of a place that is
selling those, you might, with some luck, be able to get the drivers
within the near future. Adapting their software to new hardware should
not be too difficult for it only requires replacing the transport layer;
the existing software is quite modular, so I do not expect this to take
more than a couple of days for getting a working alpha-release. BTW
there are already several alternative transport layers, one of them is
using the MIDI port (though that one is way too slow!) --- so it *is*
possible to replace these modules!

I hope, these boards are not too expensive, since buying two or three
of them for developing the software would be required...

I would apprecitate, if you mailed your reply to
srb242@math.uni-muenster.de so it does not get lost in between all the
other news :-)

Thanx in advance

Markus Gutschke

---