🚀 go-pugleaf

Karl,

>In a personal account format, you're talking about $59 a month!  Now *that*
>is very, very reasonable in cost.  If you don't need 250kbps of effective
>performance you can get half of that for $39 -- again, assuming personal,
>outbound-only, part-time connectivity.

$59 per month for 128 Kbps synchronous access speed, plus data compression at no extra cost, is 
about the best price for personal Internet access I have come across.  I would be interested to 
see any other offerings in this price and performance category!

Many people don't know that V.34 modems are asynchronous, which means that for every 8 bits of 
data a start and stop bit are also transmitted -- this leaves a net throughput of about 23 Kbps 
under the best line conditions.  This means an ISDN 2B line is better than five times faster, 
excluding the effects of compression.


>This is true.  I just find it incredible that people think that cutting 
>the cost of real (as in 128kbps+) access to the net by 75% in the space 
>of six months through the application of technology still isn't enough!

The fact that ISDN lines can be had, on a residential tariff basis, for about $30 per month, 
with no per-minute usage fees (for calls under 8 miles in distance), is significant in my mind. 
 And given the way that AOL, CompuServe and MSN are heading (AOL is already there), once you 
have fast IP connectivity you are set!  Just connect to the Internet, run your online client 
software, point to their gateway IP address, and you are flying along!

[...]
>An Ascend Max-4000HP (which is what we use here) runs about $25k.  That's a
>moderate configuration, not a totally decked one.  You can do ISDN MUCH 
>cheaper than that, but the problem for the customer is that performance 
>suffers severely and compression becomes unavailable.
>
>Compression is a particularly sticky issue -- without hardware assist
>you're dead as an ISP if you try to permit this.  The CPUs in your routers
>will die trying to handle real-time compression of multiple T1 sources
>(remember, ISDN at 128kbps is 1/12th of a T1 *per customer online*!)

Very pricy stuff, but as you say, there is no real option today.  Maybe one of these years 
Chicago Cable (TCI) will offer IP access over their cable network, and Ascend will drop their 
prices by 75%...


>>But ISDN success is still uncertain. It is an 80's technology which the 
>>telcos have never really embraced. But the big question is if something 
>>faster becomes available at an affordable price. Not to mention when. 
>
>I doubt it, for the reasons above.

ISDN's success is no longer uncertain.  Pick up any networking trade magazine and you will find 
lots of success stories.  The equipment manufacturers are finally jumping on the ISDN bandwagon 
in a big way.  Look at Ascend's stock price over the last year to see how Wall Street is 
reacting.  ISDN may not be the fastest and best, but when you are faced with using billions of 
dollars of embedded copper plant (twisted-wire), it's a darn good start away from analog phone 
lines.

>The killer for standard dedicated access is the backbone hardware required
>to support it properly.  Again, you can cheapen the service by offering it
>over Frame Relay, not using real backbone router hardware (ie: 2501s on
>your backbone instead of AGS+ and 7000 equipment), etc.  But it will show
>up in the performance your customers see.
>
>I don't believe in doing that just meet a price point.  MCSNet is happy
>to let the other guys duke it out for the "least buck" service offerings,
>while taking the high-performance, and slightly-higher-cost, road.

Thanks for the informative post, Karl.  The fact is, you have put your money on the line for 
ISDN, unlike other companies in the area.  As you say, it is one thing to throw together some 
async terminal servers with ISDN Terminal Adapters, and get mediocre performance improvements, 
but another to go with a professional (and expensive) solution to do the job right!

Regards,

Stuart Brainerd