ftp.whtech.com/ftp.whtech.com.tar/ftp.whtech.com/Geneve/9640news/CAT01/MEMORY1.ARK

Text File | 2006-10-19 | 8KB | 190 lines

?
               THE FACTS ABOUT TI MEMORY SYSTEMS
                         AN EDITORIAL

Over the last six months there has been a lot of noise on the
computer networks and in user group newsletters on issues
related to extended memory cards for the 99/4A. Some people
have blatantly asked people to come out and choose sides on a
very complex issue without understanding what they are
choosing. In fact, what should be a pretty objective decision
has been turned into an emotional gut-churner - a question
decided by loyalties, petty rivalries, lies and innuendo.

Frankly, this is why we are in the situation we are in today
- and why companies like Myarc and Corcomp left the
community. Everything technical debate gets turned into a
personal vendetta, thinly disguised ambition is allowed to
prevail over substance, and the community eats its young yet
again.

When I set out almost 3 years ago with a really talented
bunch of guys to put together a new kind of memory card for
the 99/4A, I had believed that the community had finally
outgrown that kind of thing. I was wrong. Seeing all of this
stuff all over again has made me seriously consider throwing
in the towel once and for all.

Why? Because everything we've done with these cards has
either been ignored, mis-represented, or labeled as too
"controversial" or "not ready for prime time".

For 8 months we've been mailing out press releases, articles
and newsletters about our memory cards that apparently no one
is reading, and user groups aren't re-printing or even
reporting on. The only reference to what we've done that I
ever see in print is usually in in an article about our
competitors, or in an editorial that simply says that there
has been a "debate" and that it has "gotten out of hand".

This is simply ridiculous.

This is the most important thing I've been involved with in
the 10 years I've been in this community, and unless the
community gives this a fair hearing, well, I guess it's
finally time to cut my losses.

Here is my last attempt to get the unvarnished facts out in
front of you, the reader.

1. The Asgard Memory System (AMS) is available NOW - it is
NOT still in "development". We announced the product the day
it was commercially available for sale, and in stock. In the
last 8 months, we've been refining the product, writing
software, and working on the next generation card. Our only
competitor announced their product over a year ago, and have
yet to release more than press notices (which all seem to be
faithfully reprinted everywhere). It is pretty hard to
compete against something that so far exists only on paper -
especially when the unreleased product gets more press than
the one that you can buy today!

2. We started AMS almost 3 years ago - long before there ever
was a "National Committee for TI Standards". This so-called
committee has never met more than once, doesn't include most
of the TI hardware or software developers in the U.S., much
less the rest of the world, and has produced a specification
for memory systems without any real debate, which endorses
our competitor's plans. Before we had a chance to object, it
was the declared "standard". Can you say "railroaded"?

3. Our memory system was designed to the only standard TI
ever made for extended memory on the 99/4A - the one used in
the TI-99/8. In fact, the guy who DESIGNED the TI-99/8 said
our design was identical to the one TI specified.

4. Because our design was built to TI's specifications, it
doesn't conflict with any other card in the P-BOX - except a
32K card. You can plug it in and your Horizon RAM-disk, Myarc
HFDC, or anything else you have will still work fine.

5. Our design uses standard, off-the-shelf components. EVERY
other extended memory design uses lots of custom ICs, and
even more custom programming (as in a big DSR). Custom parts
not only drive up the development time, they also drive up
the cost, and guarantee that the design remains proprietary.
By using off-the-shelf parts, we keep the price down, and
guarantee competition. Remember how much TI used to charge
for the 32K card when they were the only one making them?

We designed our system to the "KISS" method - "Keep It
Simple, Sam".

6. Our system is tried and tested. We use the exact same
"memory mapper" (the chip that controls the computers use of
memory) that TI used in their 99/8, their 9900 minicomputers,
and that IBM used in the very first IBM PCs. This component
has been available for 10 years - all "bugs" in it have long
been removed.

7. Everything about our system is "open". Anyone can write a
program for it or enhance it - the hardware and software
specifications are available free of charge. Heck, the 5-disk
development system we've spent the last 18 months writing is
even fairware - and posted on the bulletin boards.

8. The AMS is very fast. It can switch pages over 10 times
faster than any competitor, and with little program code
(even in Assembly). Why is speed important? If you are
sorting 512K of data, or loading 512K of pictures, you'll
notice the speed - in fact, you'll notice the other system is
less than half the speed.

9. Our system doesn't have its software in a DSR - and we are
proud of it! Why?

     A. We found that putting the operating software in a DSR
        makes it run much slower than if it was in RAM - and
        really doesn't give any benefit to the programmer or
        the user.

     B. Any DSR increases the chance for compatibility
        problems - who wants to waste time finding problems
        with Myarc cards?

     C. A DSR is "fixed". If you find a bug in it, the only
        way to correct it is to replace it. Consider all the
        pain Myarc users have gone through with EPROM
        upgrades of the HFDC and the Geneve.

     D. If programs are written to work around a DSR bug,
        they may not work when the DSR is fixed.

     E. If the software to use the card is built into each
        program, than the only thing we have to do to correct
        a bug is issue an upgrade. Old programs written for
        earlier versions of our operating system software
        would continue to work fine, and new programs could
        take advantage of new features without worrying about
        hardware compatibility problems - since the operating
        system isn't in hardware.

     F. Why do you think Microsoft and Apple load their
        operating systems from disk, and not from ROM chips?

10. We have a complete set of development tools available
NOW. Even if our competitors released their card today, it
would be a year before they had a system that was as easy as
ours is for programmers. Because our software was designed
before our hardware, we were able to design a "programmer
friendly" system that is far easier to program than any other
extended memory system. This is important - as so many people
have said, who wants a memory card there are no programs for?

In the last 8 months since we released our first AMS card
we've released 2 software packages that take advantage of the
card (including the word processor FIRST DRAFT), and software
from other people has started to appear. Around 20 AMS cards
are in the hands of developers around the world.

Is any of this news? Apparently not - I've seen few of the
facts above in print anywhere, even though we've put them in
a half-dozen articles.

The facts, on their own merit, should be compelling enough
for people to put aside their differences and really weigh
the benefits of what we've done - instead of consigning it as
some "curiosity", or ignoring it.

We wanted to put together something that was cheap enough to
build that every TI user could have one, and yet was simple
enough to write programs for that every TI programmer could
do so. I think we've done that. If the TI world isn't
interested at this point, doesn't care, or wants to keep
waiting for fantasies, well, I can take a hint.

Thank you.

Chris Bobbitt
July 2, 1993





Press <CR> !

Download complete.  Turn off Capture File.