Date: Wed, 20 Mar 1996 10:35:13 -0600 (CST) From: Joe Greco <jgreco@brasil.moneng.mei.com> To: kaveman@magna.com.au (Julian Jenkins) Cc: jkh@time.cdrom.com, terry@lambert.org, lehey.pad@sni.de, hackers@FreeBSD.org, isdn@muc.ditec.de Subject: Re: ISDN: "modem" or board? (Was: Microsoft "Get ISDN"?) Message-ID: <199603201635.KAA29512@brasil.moneng.mei.com> In-Reply-To: <Pine.BSF.3.91.960320214926.990B-100000@kavemachine.magna.com.au> from "Julian Jenkins" at Mar 20, 96 09:58:30 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> > Oh, I know the limitations of that, I was just trying to figure out > > how Greg was managing to take 64+64 and get "230K" from it. I'm well > > aware that the next step from a 115.2K UART is 230.4Kbaud since they > > generally just double the previous value, but Greg's comments seemed > > to indicate that he was also ADDING the overhead rather than > > subtracting it and coming up with a wholly new bandwidth category for > > ISDN, which would be a neat trick and worth some money were it that > > easy.. :-) > > The key point is that the ISDN line is syncronous, so a 64kb/s line can > carry 64kb/s of data, while the async line from a 16550 or whatever need > extra start/stop bits. Assuming we can get full capacity from the serial > line and we hav only one of each start and stop bits, this means that > each byte transmitted along the async line consists of 10 bits (8 bits > data, 1 start, 1 stop) so the capacity of the async line that can carry > the same amount of data as a 64kb/s ISDN line is 64/8*10 = 80kb/s. Something I haven't-tried-but-would-be-interesting: Running a single B channel at 115200 and seeing what V.120's throughput is (i.e. whether or not it is passing the data efficiently or inefficiently). If it is running just the data bits across the link, one would expect to see 64000 / 8 = 8000cps. If it isn't, then one would expect 6400cps. If the former is true, then it's just a matter of convincing ISDN TA mfrs that there is a market for terminal adapters running at 230.4Kbps with "bonded" B channels, for 16000cps. :-) Of course, I still think TA's should do compression too... I can just see the need for 921.6Kbps serial ports... ... Joe ------------------------------------------------------------------------------- Joe Greco - Systems Administrator jgreco@ns.sol.net Solaria Public Access UNIX - Milwaukee, WI 414/546-7968
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199603201635.KAA29512>