> I agree! It seems to defeat the whole purpose of going all digital.
> When they came out... The whole idea was to have digital lines coming in
> to your facility, so you could eliminate the problem from being on your
> end. But, it seems the exact opposite has happened. With the PM2s and US
> Robotic Couriers, you gave people the software worked them thru getting
> the software setup and quit worrying about connection problems. Now it
> seems to be an endless stream of try this init string or that init string,
> always having to fiddle with modems..
(Tried to find one part to quote about, but all of it applies)
We had the PM2 with USR Couriers setup, and our customers were always
pleased that they would never get disconnected, or have slow connections,
or the other problems that the other local providers had.
Now we have PM3's, and messed around with the older modems, with slow
(high latency) connections, disconnects, some modems not connecting at
all, some modems not connecting at decent speeds. (I'm not talking about
v.fast or v.FC.)
For the most part I was able to continue to assure my customers that fixes
were coming soon. Then we got the K56Flex modems. I said to myself,
"Yes! Finally! We won't have to deal with problems anymore!" Wrong.
USR Sportster/Megahertz PCMCIA modems seem to be the worse for us. Some
seem to work just fine (to a point), others will drop connection every 5
minutes, or every 10 minutes. Some seem to only connect less and 2 out of
3 times. While the latecy issue is (almost) resolved, the new modems in
some ways have broken more than they have fixed.
> And on the same lines, why does Sherwood have a page up with different
> strings, for different modems. Don't get me wrong, I appreciate it
> very much. But the point is, why doesn't Livingston have a page up, with
The modem init strings are great. The bad thing is that we were working
just fine with all these modems before (PM2E/Courier) without needing to
change modem init strings. How many of you have explained to your
customers that they need to change a modem init string, and you know that
they're thinking, "We didn't have to do this before, why has it changed
now?" ... We have always been of the mind of "Get the new drivers and
they'll solve your problem. Stay away from software-based modems. Don't
buy cheap modems and you won't have disconnect problems." Now all those
recommendations are thrown out the window, since the brands we recommended
are now the brands that seem to rebel against the PM3. <G> It just
doesn't look good for us as a provider.
> And now we(well we don't,we are number #469) have K56Flex ver. 1.1, what
> exactly is the meaning of interoperable code?
We've had a few people able to connect at 42K or 44K ... The K56Flex v1.1
code doesn't seem to be interoperable, but just more "standard" ... As
long as everyone has v1.0 or greater, they'll have a better connect.
Perhaps the .516's and .519's were terrible at switching speeds in
between. I know that Supra can't handle line problems with their Release
Candidate of v1.0. I worked with a customer who had to force it to a
lower K56Flex speed, otherwise it would drop connection the second it
tried to change speeds. (This is just speculation. I don't *KNOW* why
the modem is disconnecting, but I'm assuming just bad code on Supra's
part - We have another customer that stays at 44K most of the time, but
I've seen him go up and down without disconnections. In fact, I think
I've even seen him drop to v.34 once or twice, but don't quote me on
that. <G>)
> Agreed again, it can't continue to be the customer's fault on modem
> compatability problems.
If the modems never worked, that's one thing. But the fact that all these
modems worked one way or another before PM3's, it's just really difficult
to blame the customer. We can continue to provide fixes and patches, but
none of this is 100% until the modem code on our end is clean enough that
a new customer can signup and be able to get online immediately without us
spending time tweaking this and fixing that.
Jaime Bozza
Nucleus Communications, Inc.