Maybe you can tweak your settings to get even better speed than y9ou are currently getting.
Even better? Wow!
Sorry, that just sounded funny, like it was already good.
FCC does not state the max speed of Dial-Up..
Not speed explicitly, but power and frequencies yes, thus speed. (considering the mechanisms of mo-dem of digital data on analog voice lines)
47CFR68.308A good but somewhat dated article on the subject:
http://www.mttlr.org/html/ShortNotes/jamison.htmlI had found a better description of how modems work before, but I can't seem to locate it now.
I have considered a second wired line and use a multilink setup to get double the speed. It would be kind of expensive though.
In my case, obviously I'm not getting anywhere near the allowable maximum, as the limitation here is the signal quality over the copper on the poles between me and the switch... Hey! maybe somebody on my road is overpowering the line!
.. It wasn't so long ago that we were on a "party line" on our road. (Anybody else old enough to know what that is?)
I've already been over all the init strings, firmware updates and most usable tweaks. As I mentioned, I do get slightly better speeds using the tighter networking stacks of QNX or lesser *nixes, thus the interest in getting OP working on Wine again.
One thing to watch is how an OS reports the connection speed (Windows included), this can be misleading if you have not actually tested it. I can get windows to report a connection speed of 115K by setting the physical serial port I'm using so, but the negotiated speed between my modem and the v.92 server at the switch house will be unaffected by how fast my PC talks to my modem.
I have even looked into developing a superior mo-dem specification, I had envisioned use of FFT (Fast Fourier Transform) to modulate and demodulate complex analog signals containing digital information. It seems to me that article I can't find described this is essentially what is already being done. Then I thought about other possibly smarter ways analogous to psychoacoustic MP3 encoding but lossless. If you think about it if you play a 4 minute song over the phone line you are transmitting considerable amounts of data (already in analog form) In digital form that would represent more data than you could download in the same time on the same voice line. I think it may possible to cram way more digital data into an analog signal, we just have to be smarter about it.
Obviously a lot of thought has already been put into it:
http://www.itu.int/rec/T-REC-vAnd now that DSL, cable and fiber have come along, it probably will not get any further attention unless someone like me finds a way to cram just a bit more in that skinny little wire.