rminnich@udel.EDU (Ron Minnich) (02/09/88)
First impressions of DNET.
***** GEE, THIS IS GREAT! *****
hmm, i blew all that time on internet.device and then matt
comes out with this thing ... it just isn't fair.
First off, it works. Second off, it requires no kernel hacks,
yet also appears to be pretty efficient. Although there does seem
to be a slightly longer echo loop it is not real bad even at 9600,
at least i don't think. But the real amazing thing is how well file transfer
works. Beautifully. and fast, as fast as tcp/ip over SLIP. I am
impressed.
There is only one slight caveat. In moving from TCP to DNET
you are moving from a very standard protocol which runs on lots
of different OSs to one which is 4.xbsd specific. On the other
hand you could probably move dnet to many different machines.
I think i may drop internet.device and move over to dnet (but am
not sure yet). No matter what, DNET is quite a piece of work.
Makes me wonder why i bought a C compiler!
Highly recommended for a test-drive at least!
ron
--
ron (rminnich@udel.edu)sean@ms.uky.edu (Sean Casey) (02/09/88)
In article <1122@louie.udel.EDU> rminnich@udel.EDU (Ron Minnich) writes: > There is only one slight caveat. In moving from TCP to DNET >you are moving from a very standard protocol which runs on lots >of different OSs to one which is 4.xbsd specific. On the other >hand you could probably move dnet to many different machines. I don't know beans about TCP/IP, but... perhaps a server/client could be written under DNET that would make it possible. I know a MacIntosh user that would like to see a Mac DNET. Sean -- -- Sean Casey sean@ms.uky.edu, sean@ukma.bitnet -- (the Empire guy) {rutgers,uunet,cbosgd}!ukma!sean -- University of Kentucky in Lexington Kentucky, USA -- "If something can go will, it wrong."
dillon@CORY.BERKELEY.EDU (Matt Dillon) (02/10/88)
> There is only one slight caveat. In moving from TCP to DNET >you are moving from a very standard protocol which runs on lots >of different OSs to one which is 4.xbsd specific. On the other >hand you could probably move dnet to many different machines. >I think i may drop internet.device and move over to dnet (but am >not sure yet). No matter what, DNET is quite a piece of work. >Makes me wonder why i bought a C compiler! > Highly recommended for a test-drive at least! >ron >-- >ron (rminnich@udel.edu) Well, the implementation is currently AMIGA<->AMIGA, AMIGA<->4.2BSD, or 4.2BSD<->4.2BSD (the last one used mainly for testing). The protocol itself is arranged in two levels and not dependant on a particular machine. The packet level assumes only that it has a full 256 character codes / byte to play around with and a decently asynchronous mechanism for Tx/Rx. The multiplexing layer assumes the machine multitasks. I've got about two dozen people testing it now (well, a dozen now, a dozen as soon as I distribute it to the USENET alpha testers). -Matt
keithd@cadovax.UUCP (Keith Doyle) (02/11/88)
In article <8802092248.AA04303@cory.Berkeley.EDU> dillon@CORY.BERKELEY.EDU (Matt Dillon) writes: > Well, the implementation is currently AMIGA<->AMIGA, AMIGA<->4.2BSD, > or 4.2BSD<->4.2BSD (the last one used mainly for testing). > The protocol itself is arranged in two levels and not dependant on > a particular machine. The packet level assumes only that it has a > full 256 character codes / byte to play around with and a decently > asynchronous mechanism for Tx/Rx. The multiplexing layer assumes the > machine multitasks. What are you using for the transmission media, RS232? If so, sign me up! Keith Doyle # {ucbvax,decvax}!trwrb!cadovax!keithd Contel Business Systems 213-323-8170