Loadstar 128 #37/q37.d81/t.diskovery 37

Text File | 1993-01-01 | 10.6 KB | 199 lines

    D I S K O V E R Y : T H E   P A S S I N G   O F   T H E   T O R C H

                              by Fender Tucker


    I once read that 80% of all 'zines' opening editorials begin with an
apology (or excuse) for why the issue is late. Meeting a deadline,
especially in the computer field, seems to be one of the hardest things to
do. Keeping an upbeat mood in a dwindling market is also rough.

    I was feeling down when LS 128 #36 came out and so I announced that we
were no longer accepting pre-paid subscriptions for issues past #40. Most
people read this as saying that there would be no more LOADSTAR 128 after
issue #40 but I only meant that we didn't want to owe a lot of people for
future issues IF there were no more issues past #40. We wanted to have a
clean end for the magazine, unlike some of the other 8-bit magazines that
have gone out of business owing subscribers money for undelivered issues.

    But then I started thinking more clearly and realized that the market
supports LS 128 decently, it's only my mental condition that is having
trouble. Because of LOADSTAR 64, which is published 12 times a year, I was
not able to devote as much time and energy to LS 128 as I felt I should,
and I felt guilty about it. Of course the solution is not to torpedo LS
128, but to get someone else to do the work that I couldn't do. I
immediately thought of Bob Markland.

    Bob has been one of the most dependable and prolific 128 programmers of
the late 20th Century, and he was excited at the prospect of helping me
out, helping out the 128 community, and maybe even making a little money at
the same time. He can spend all three months of an issue's cycle finding
programs, editing them and turning LOADSTAR 128 into a real magazine with
features. You've probably noticed how "bare bones" the recent issues have
been lately. That's because I could only devote a week or two to an issue
of LS 128.

    So forget anything I've ever said about the future of LOADSTAR 128. Bob
is knowledgeable, energetic and dedicated and let's let HIM tell us where
we're going. I know he'll be as upfront as I was, but I have no doubt that
the news will much better coming from him. He'll be almost as accessible as
I was, so please write him with any comments or suggestions for LS 128. The
C-128 deserves a great magazine on disk, and one that appears regularly,
packed to the gills with the best 128 software around. Bob Markland is the
man who can provide us with such a magazine, especially if we all pitch in
and help him. If you've written any 128 programs that help you out around
the house, send him a copy. If you know of a good 128 programmer who
doesn't know that he can be paid hundreds of dollars for a good program,
tell him about LOADSTAR 128. If you see something online or on a PD disk
that, with a little upgrading, would be LOADSTAR 128 quality, send Bob a
copy.

    The new address of the LOADSTAR Tower West is:

    Email:  [e-mail]

    Bob Markland
    Box 1535
    Gillette WY 82717
    307-686-8123

     If you're shipping him something UPS and need a non-box address,
it's:

    Bob Markland
    6501 S. Douglas Hwy #7
    Gillette WY 82716

    I'm sure Bob has more details about his new job in his editorial that
follows this one. Let's get behind him and make LOADSTAR 128 into the best
little disk magazine in Wyoming.


 THE SUPERCPU V2
 ---------------

    Another reason why it was so wrong for me to scuttle LOADSTAR 128 is
that one of the best new devices for the C-128 is now available. Creative
Micro Design's SuperCPU V2, CMD's name for the 20 MHz accelerator which
works in both 64 and 128 mode, is on the market and I've had one for the
past couple of weeks. Now I never want to do without it.

    I haven't had time to do an exhaustive, detailed test of it -- and
probably won't, so somebody else should -- but I've used it with several of
my favorite 80-column programs. THE WRITE STUFF 128 was of course the first
program I tried once I installed the SuperCPU V2. By the way, installing
the SuperCPU in my flat-top 128 involved taking the cover off, removing the
MMU chip, inserting the chip into the pop-on board provided by CMD and
inserting this board into the MMU's former slot. The board had four wires
with pin clips on the end which needed to be clipped to pins on another
chip close by. All in all, a five-minute job from start to finish, and one
that anyone can do.

    I'm using TWS 128 right now and it all works fine at 20 MHz. The screen
is displayed faster and the keyboard is more responsive. The speedup is
really noticeable when loading from a partition (or disk) with hundreds of
files. At 2 MHz it takes about three seconds for the program to display the
first page of files when I select LOAD from the command bar at the top of
the screen. Each page takes another three seconds. At 20 MHz it's around a
second per screen.

    I just tried some sorting. I took the first section of this Diskovery
and, using the Search and Replace function of TWS I turned each SPACE into
a carriage return. That gave me a "list" of a little over 500 words,
randomly sorted. Then I "ate" them into a buffer then restored them with
CTRL-SHIFT-R, sorting them alphabetically. The sorting at 2 MHz took around
85 seconds. At 20 MHz it took 9 seconds, a speedup of almost exactly what
it should be. The accelerator really works!

    There are many other functions of TWS that are benefited by the
SuperCPU V2, I'm sure, at least anything that's done in memory without disk
access. For loading and saving, I have RAMLink and JiffyDOS so to me, disk
access time is negligible.

    The only part of TWS that the SuperCPU makes "worse" (that I know of at
this time) is reading an unknown type of file from disk and converting it
to TWS' screen code. In TWS 128 you press CTRL-4 to see the normal
directory scroll down the screen. Then you use CTRL to halt the listing,
and STOP to get an arrow to appear. Then, holding the CTRL key down, you
use the CRSR UP-DOWN keys to move the arrow to the file you want to read.
At 20 MHz this arrow is way too fast. It's very difficult (if not
impossible) to stop the arrow on the file you want.

    So what's the solution? Why, simply flip the SuperCPU SPEED switch to
Normal before listing the directory. So far I have been able to flip from
Turbo to Normal and back at anytime in TWS without a crash. Anyone who has
had a SuperCPU V1 knows that you can't just flip between the two speeds at
any old time. You have to pick a moment when nothing of importance is
happening interrupt-wise. I'm pretty sure that if I were to start flipping
my SPEED switch indiscriminately, TWS 128 would eventually crash, but to me
the good news is that TWS is solid when flipping the SPEED switch.

    My other favorite 128 program is ed bell's MAGIC 128 (last on LS 128
#35). It's a great directory editor (and more) and it seems to work okay at
20 MHz. In fact it's much more enjoyable at the faster speed because when
you have a directory with over 200 files, it re-sorts the files much
quicker. When you insert a file from way down in the directory into one of
the top positions, all of the files after the insertion point have to be
moved down, and it can take several seconds for MAGIC to do it. At 20 MHz
it happens in about a second. Since I clean up disks with 250 or so files
every month (the 1581 LOADSTAR) I really appreciate the speedup.

    One bug I've found: when editing the directory of a 1581 disk with
MAGIC at 20 MHz, certain files' sizes are messed up. A 20-block file all of
a sudden can become 310 blocks in size! It's weird. The BAM still thinks
the file is 20 blocks long so Validating doesn't fix it. The blocks free is
correct for it being a 20-block file. And when you copy the 310-block file
using JiffyDOS built-in copier, it copies as a 20-block file. If you load
and list it (or read it) it's 20-block long and correct. The only thing
wrong is that when you list the directory, certain files will have file
sizes that are three digits long instead of two.

    Since I can easily "fix" the file by copying it to another disk and
back again, I haven't worried too much about this bug. It only seems to do
it on a 1581 or FD-2000 -- not on a 5.25 inch drive. I think the position
of the file on the MAGIC edit directory screen is what determines which
file(s) will be messed up. For instance, if the directory's files take up
three screens in MAGIC, there'll be three files with too-large block sizes,
all three of the files occupying the same relative position on the screen.
Strange. If anyone cares to research this "bug" and can give me and other
MAGIC users some advice, we'll take it. ed bell is no longer in the
Commodore field so MAGIC can't be "fixed" but it's not really a MAGIC bug
or a SuperCPU bug. It's just one of those little incompatibilities that we
shouldn't be surprised about.

    I've also tried a few of my 128 games at 20 MHz and except for a few
cases where some menus move faster than is comfortable, they work fine.
Much better, actually. Notice how long it takes for my KEGLER 128 program
to draw the first screen. On my system, the Main Menu simply pops up on the
screen. You can just barely see the screen being drawn.

    For programmers, the SuperCPU has one huge bonus. The speed of LISTing
a program is just about perfect, speed-wise. Normally, I want the code
scrolling up the screen to go faster. Now it does. And have you ever tried
to insert a small line number in a 150-block BASIC program? It can take up
to 8 or 9 seconds for the cursor to come back. At 2 MHz it takes a long
time for the BASIC editor to move all of that BASIC text up in order to
squeeze in the low-numbered line. At 20 MHz the cursor comes back in less
than a second.

    Mark Fellows, Doug Cotton, Charlie Christiansen and all of the other
folks at CMD deserve a big round of thanks for inventing the SuperCPU V2.
At around $200 it's the financial equivalent of buying a scanner or some
other peripheral for an IBM system. The IBM world thrives because of its
hundreds of gadgets, and here's our chance to make our world faster and
better. True, we all got along okay at 2 MHz, but now that I'm clicking
along at 20 MHz, I don't want to go back to the slow old days. If you want
to stay up with me, consider asking for a SuperCPU V2 for Christmas. I
can't speak for everyone, but I'll take a 20 MHz accelerator over a
sweater, a couple of shirts, some socks and underwear anyday.

 P. S. If I've implied that Judi gives me only clothing for Christmas, I'm
wrong. She usually gives me books, games and other neat things. As far as
I'm concerned I could live for another 50 years and never buy any more
clothing. Clothes? Who needs 'em?

                       \\\\\ RETURN - Menu \\\\\