Classic Computer Magazine Archive CREATIVE COMPUTING VOL. 9, NO. 8 / AUGUST 1983 / PAGE 229

... Commodore's port... (column) John J. Anderson.

Commodore's Port

Before we get started with the column this time around, I would like to put out an all points bulletin to Commodore computer owners everywhere, in the form of a call for manuscripts.

Here at Creative Computing we are currently amassing material for at least two books about Commodore computers. The Creative Vic-20 and The Creative Commodore 64 will continue in the tradition of Creative Microcomputer books, providing a compendium of articles about the two hottest Commodore machines around.

If you have a contribution to make to either of these two books, we would be most happy to evaluate it for inclusion in either or both books. Because we like to include new material in the anthologies, your application, program or review may be just what we are looking for. So get it done. Make sure that it is typed, printed out, or otherwise legible, and doublespaced. Attach program listings, photos, and/or diagrams. Then get it in the mail to us.

As ever, we are Creative Computing, 39 East Hanover Ave., Morris Plains, NJ 07950. IF you want your manuscript to be returned, enclose a self-addressed, stamped envelope, correctly sized.

Who knows? Perhaps your own work will be immortalized by Creative Computing Press. So get cracking!

In this month's column, we'll take a look at a very controversial topic concerning the Commodore 64, then evaluate a new mass storage device for the Vic-20 and 64 machines.

The Great Commodore 64 Video Controversy

When the Commodore 64 first arrived on the microcomputer marketplace, it caused quite a stir. It was the first home computer with 64K standard; at an announced list price of $595, it was a real bargain.

In the business of microcomputing, six months is a generation, especially in price strategies. Now all you need to do is leaf through the mail order ads in any microcomputer magazine to find the 64 selling for $350 and less, postpaid. I confidently predict that the machine will break the $299 barrier very soon. It won't be long before a Commodore 64 system with disk drive and printer can be purchased for less than $1000.

At those prices, the Commodore 64 is very hard to beat. It offers sprite graphics, built-in Basic, and very sophisticated three-voice sound capability. Logo and CP/M have been slated for upcoming release. Add to this, announced piano keyboard peripherals and advanced gaming capabilities, and you have the perfect home computer, right?

Seeking Resolution

Well, maybe. The reputation of the 64 has been clouded by reservations about the quality of its video. We have received inquiries from all over this country and beyond, asking us to comment on the video quality of the 64. Some people tell us that the pictures they have seen on screens at their local computer stores are singularly unimpressive. Others ask us whether early models carry insufficient RF shielding, causing interference.

Most often consumers poised to purchase a low end microcomputer ask whether a problem with video should affect their decision to choose the 64. My answer is no.

Scout's honor and for the record, though it may not be the best in its class, the Commodore 64 is capable of producing a very acceptable color video signal.

Before a user passes judgment on the reception of a unit in a department store, computer store, or even his own home, he should understand some of the obstacles to good video, and the requisite steps to overcoming such problems. Commodore 64 video is definitely one of the more persnickety signals you may encounter, and without a doubt it demands kid glove treatment. You must realize that it will not look the way it did on an NEC monitor in the store when you hook it to the 20-year-old Philco in your den, though I admit that the sales-person seldom underscores this before a sale.

It has been my experience, however, that a video signal is rarely well tuned in a retail store. Think of how lousy the pictures look on the ordinary TVs in a department store, then think of the additional skill required to set up a computer. You can usually do better at home. Of course, there are limits to how well you can do with an ordinary TV, especially with a 64. Actually, the most dramatic way to improve the output of a Commodore 64 is with a video monitor.

When you hook a computer to your home TV set, you route the signal through a special modulator. Then the video and audio signals are sent through the antenna imput and tuner of the set, just as if it were this week's segment of 60 Minutes. Although this is certainly a convenient means of accessing a preexisting CRT, it has severe limitations The signal degenerates in passing through the tuner section of the set--if only the video and audio signals could bypass all of that, and go directly into the circuitry designed to get the picture on the screen and the sound into the speaker.

That is exactly what many inexpensive raster monitors are designed to do. The two we have at Creative are the Amdek Color I and and the TI Color Monitor. Both have built-in amplifiers and speakers to handle audio, and either can be purchased for under $400.

The difference between the video quality of the 64 on a regular RV and one of these monitors is very striking--the monitor video looks twice as sharp and can take much greater color saturation without "bleeding'--the extremely unfortunate smearing of borders between clashing colors, which is the foremost complaint I have heard from owners of the Commodore 64.

Apparently Commodore has taken the complaints to heart, because they are introducing their own inexpensive raster monitor.

If you have been thinking about getting a monitor, you should be aware of another potential boon: No longer need the other members of the family equate your time on the computer with their loss of regular TV. I might mention that this made quite a difference in my own life with wife and micro.

Whether or not you have a monitor, there are further steps you can take to maximize the signal of the 64. One of the first points I noticed about tuning in its video is that the monitor must be set quite differently from the levels I had thought of as "default' (based on the time that I have logged with Apples and Ataris). In fact, on the Amdek, my monitor of choice, there are detents for default settings which always seemed about right for those machines. Not so with the 64.

The most important change is to back off on the color level. This will immediately improve reception in almost every case: The Commodore chroma level is designed for less saturation than other machines, and less than broadcast TV too. Next, increase the brightness and decrease the contrast of the picture. This, combined with lessened color intensity, will improve video quality substantially.

Although colors on the 64 will be less vibrant at these settings, they still can be very bright. If they look washed out, go ahead and boost color saturation. Look for an overall balance between less than usual color and contrast, and greater than usual brightness to reach maximum video quality.

A Look At Raster Technology

And yet, even these steps are not enough to thwart all cases of color bleeding, or some of the other peculiar problems of Commodore video (we've seen them on the Vic-20, too). To understand more about these problems, we must take a closer look at raster technology in general.

In the design of any machine there are trade-offs. In the design of color computers, high-end designers may commit to RGB technology, using special, expensive monitors to create impressive resolution and clarity. On some of these systems nearly photographic results can be achieved, and color bleeding occurs only when desired, by making color borders ragged or pointallistically blended. Modulation to a conventional color TV is automatically ruled out on these systems.

In succumbing to the restrictions of raster technology so that a signal can be pumped to raster monitor or conventional television, the limitations of that technology are passed to the computer designer. The resolution of such systems is very limited, and some kinds of color bleeding can be prevented only by avoiding the color combinations that cause them. These limitations are defined by the system, and nothing inside the computer will get around them.

Let's compare. On the Apple II computer, character sets are always white on a black background, no questions asked. On the Atari, backgrounds can be any of 256 colors, but text must always be white or black, and in many cases, a lack of contrast or severe color bleeding makes certain color combinations unacceptable. On the Commodore 64, there is much greater flexibility. Character sets in any of 16 colors can appear on a background of any of the same 16 colors. The color of multiple characters can be controlled individually in a very straightforward manner with direct keyboard commands (pressing CONTROLWHITE turns the cursor and all subsequent characters white). Combined with the keyboard graphics characters, this flexibility offers a powerful graphics tool.

It also creates problems. Many color combinations render the characters completely indistinguishable. Then there are the default color settings, which are mediocre at best. Though there is no bleeding, the light blue characters on a dark blue field with a light blue border are real eye-strainers. No wonder people complain about video on the 64 with such a case of the blues. At the least, the temptation is to turn the cursor white on the default background. Alas, that causes color bleeding.

Before I could get serious about programming with the 64, I had to solve this problem. The result is presented here as Program 1. It very simply makes all background and border combinations as easily accessible as character set colors and automatically picks the best available character set color to complement a chosen background. The choices always represent a color combination causing the least possible color bleeding.

As you can see from the color photographs reproduced here, many character set/background combinations are far superior to the default set, and I think you will agree that the video quality is very acceptable.

The only backgrounds I couldn't do much with were 3 and 4, cyan and purple. Those backgrounds create smearing problems with all other colors. My advice is to avoid them except for border colors and character graphics applications (on a black background).

All 14 other colors have complements that work out very nicely. Most of the time they are light and dark versions of the same color: light green of a dark green background or dark green on a light green background. In some cases another combination works out, such as yellow, color 7, which looks good with brown text on it.

The reverse of default, using dark blue letters on a lighter blue background, is greatly superior to the default, I think. But my favorite is number 12, medium gray--very restful on the eyes while maintaining good contrast. After choosing it as a background, you can turn the cursor shade white, light gray or black, with absolutely no smearing. This could be very handy for applications that call for highlighted text.

After any set of choices, the values to be POKEd for that color combination are enacted, then listed at the top of screen. As the character set color changes after these lines are printed, you will see some great examples of severe smearing in the POKE read-out lines. To bring them in clearly, you may have to select the same color combination a second time. Then they will be displayed in a suitable color combination. You may note the figures to include wherever that color combination is desired in other programs.

The Program

The program is short and self-explanatory. It displays the locations and values needed to create each color combination. I encourage you to use it as a starting point. Other complementary color combinations I did not include are surely available--you are by no means locked into the automatic choices here. But they do provide a basis on which to build.

There is another strange idiosyncrasy of 64 video, which seems, unfortunately, impossible to remedy. Even on a monitor, we see light, parallel vertical lines on all backgrounds except black, gray, and white. The only thing we can do is lessen contrast to minimize them. Though they are by no means severe, they are the kind of thing that just does not happen on an Apple or Atari. I do not know whence they emanate, but my guess is that they are not RFI related.

Well, I hope I have made the issues at stake and the image of the 64 a bit clearer for you. Of course the only way to really get the picture is to resolve it for yourself.

The Wafertape Alternative

If you own a Vic-20 or Commodore 64, you know that up until now your media for mass storage have been limited to two parochial choices. You could make an investment in the speed, convenience, and reliability of a disk drive. The only problem with this approach is cost, which in some cases exceeds that of the computer itself.

But if you chose not to part with $400 for a drive, you were stuck with the very dreary prospect of cassette storage, using the inexpensive Datasette recorder. The snag: Although audio cassettes are great for storing the superlative audio of Kate Bush, they are not so good at storing digital data. They are necessarily slow and unreliable and do not offer the possibility of random access.

But now there is another form of storage for data from your Vic or 64. It is the wafertape or stringy floppy.

A wafertape? What's a wafertape? Well, a wafertape is something like a cassette, only quite a bit smaller, actually about the size of a credit card and not much thicker. It is wound in a continous loop and can be of various lengths (20 feet is standard). The tape is driven at about eight inches per second, and a 20 footer can hold about 16K of data. The splice connecting the tape into a loop is silver and can be detected optically by the drive. By counting out footage from the splice, the stringy floppy can approximate random access, though it may take substantially longer than random disk access.

Stringy floppy? Well, yes, the term "stringy floppy' is probably a misnomer because the medium is neither stringy nor floppy. The intent is to evoke the image of a tape medium capable of many of the same features as a disk.

Though the wafer is smaller than cassette, it is a much more reliable computer storage medium. The tape speed of 8 ips (as opposed to 1 1/2 ips for audio cassette) speeds data transfer and greatly increases the reliability of that transfer. Tape drop out and transport problems are minimized. In addition, wafers can be write protected in a manner akin to floppy disks.

The hardware unit is extremely compact and relatively free of mechanical parts in its transport mechanism (see photos). Your Vic-20 or Commodore 64 addresses stringy unit 0 as device 20. A second drive will be device 21. They attach in the same manner and via the same DIN pinout as a disk drive. The Datasette drive can remain attached and operational along with a wafertape device, which can also be daisy-chained via DIN to other devices, including disk drives and printers.

At about $200, the Exatron Stringy Floppy delivers performance placed about half way between a cassette and a disk drive. As a potential buyer you must ask the question whether waiting for savings to increase and disk drive prices to fall is preferable to committing to a wafertape system.

There are many factors to consider. Wafertapes are relatively expensive (about $3.00 each--not much less than a disk), and are available from only a single source: Exatron. Don't look for preprogrammed wafers, at least for the foreseeable future.

Couple these factors with the recommendation in the rather lackluster documentation accompanying the unit that you should store only one program per wafer until you gain "further experience' with it, and you will realize quickly that wafers are a more expensive storage medium than disk.

Why only one program per wafer? That is a question I asked, too. The documentation equivocates. As I discern it, the answer points out the main disadvantage of the wafertape drive. To store multiple files on a single wafer, you must advance to a clear area of the tape by loading through existing files, and then use a special save command. The unit will return an error condition if a single save file is longer than wafer capacity, but it will write quite blithely over existing files with no checks whatsoever. It is up to you to estimate how many files of what sizes may share a single wafer.

Although its performance was good, it seems to me that the ESF 20/64 is an unfinished product in many wasy. The potential is there for a new and reliable storage medium, but that potential is not fully reached. The unit is just too dumb, falling in the lamentable category of products that call upon the user's brain memory as buffer to take care of tedious details. Written files can be verified, but it is just too easy to blow an existing file while writing a new one. The stringy floppy I tested was not only exceedingly dumb, but didn't even have an on/off switch. This means that you have to get on all fours to plug and unplug the unit to use it. That is really annoying.

Whether or not the wafertape has any real future in the microcomputer industry is for Exatron to decide. If it takes the time to finish its product, that certainly will be a start.

I think that despite its problems, the wafer does have some real advantages as a medium. It is smaller, more powerful, and more reliable than conventional cassettes. It could be made portable because its low-voltage motor could be battery powered. And for single file storage, it is relatively easy to use. Apparently, Coleco agrees, because it will market a read-only stringy unit for single file preprogrammed Colecovision games. Compared to ROM storage, the wafertape is a very viable alternative.

But the price must drop substantially if the ESF 20/64 stringy floppy is to survive. At the price of the Datasette, the stringy floppy begins to look like a much more attractive product. At $200, however, it is simply too much for too little. The thought of two wafertape drives on a system at the same cost as a single disk drive is laughable (although the Commodore disk drive has its problems). It is simply not an option any educated consumer would consider for long.

The ESF 20/64 is covered by a very attractive warranty however, and should provide reliable, though limited service.

Next month: graphics and sound on the Commodore 64.

Table: Program 1.

Photo: A sampling of complementary color combinations for the 64.

Photo: The Exatron Stringy Floppy.

Photo: The wafertape is compact and continuous.

Photo: A look inside the unit.