![]() |
On Fri, 16 May 2003 11:35:20 -0400 George Bell <gbell5@cox.net> wrote: > > The first parameter should be the modem buffer size, which translates > into decimal as > 2800. I presume this is in bytes? That means there's just 2.7 Kb of > modem buffer memory. Geez, why so little? If your average video card > can hold multi-megabytes then why only a few paltry kilobytes for a modem? > There are many reasons. Serial transmission, which involves the sending of a single bit of data at a time, over the telephone lines is an inherently very SLOW process compared to the capabilities of the PCI or AGP video bus. For practical reasons, a large data buffer makes no sense, except perhaps on a very old and antiquated 8-bit machine. Also, the whole technology of using modems and telephone lines is also highly antiquated. The current protocols have their origins way back when digital computers were far less capable. Modems are essentially a throwback to those ancient times. Then there is always the issue of cost. Increasing the buffer size, even for cosmetic reasons, would add a few pennies to the cost of manufacture and the corporate managers would begin to scream. But digital technology is very simple. A weekend hobbyist could very likely gather a bunch of chips from off the shelf at the local electronics store and solder together a very potent serial data transmission device with lots of buffer space. Of course, he would have to implement his own protocol, but with Linux/GNU this is also a straightforward matter. If successful, the next step would be to approach various venture capital firms for the means to effect mass production and distribution. Most of us, however, would prefer to stick with the established conventions. Frank Peters ____________________ HylaFAX(tm) Users Mailing List _______________________ To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.cgi On UNIX: mail -s unsubscribe hylafax-users-request@hylafax.org < /dev/null *To learn about commercial HylaFAX(tm) support, mail sales@hylafax.org.*