Some of you may have seen the following item in a recent Ebayzilla auction:
"The Half-Speed VECTREX was my solution to evaluate new games at higher difficulty levels. It was also used to resolve internal component speed problems. This actual machine was used to resolve the "NARZOD" raster problem that only occurred at very high game levels. It was also used to qualify the new "Web Wars" and "Polar Rescue" games."
I got curious about Ebayzilla's machine and figured I'd try to make my own Half-Speed Vectrex. It seemed like all I'd have to do is change the clock speed, which could be easily accomplished by replacing the 6 MHz crystal.
Considering the original clock speed I would've guessed that Ebayzilla replaced the Vectrex's clock crystal with one that ran at 3.579545 MHz. They're pretty common crystals that are used a lot in color TV and whatnot, and chances are if you had a couple crystals sitting around, that would be one of 'em. If you do the math that actually winds up being closer to 60% of the normal speed, and Ebayzilla's listing said that the machine "actually runs at 58% of normal speed." The Vectrex is clocked at 6 MHz, so 58% of that would be 3.48 MHz. I looked in a couple of catalogs and I couldn't find any crystals that ran at that speed, so I'm not sure if Ebayzilla's claim about the actual speed of the machine is quite correct, or if I'm the one who's off. :)
I didn't have any 3.579545 MHz crystals sitting around... so much for my claims about spare crystals! I used a 1.8432 MHz one instead as it was the only thing I had that was under 6 MHz. The machine would then be running at just under 31% of the original clock speed, just about one-third speed. I got out my soldering iron and went to work; I had my "third-speed Vectrex" ready for its maiden voyage in a matter of minutes.
I was curious as to whether the display or sound would be affected by the modification. I'd assumed that both were tied to the machine's clock, but Ebayzilla's listing said that the only difference was that "the game runs much slower and there is more flicker". My questions were soon answered when I flipped the machine on and the CRT warmed up.
After what seemed like forever, I heard a tone, then another. I realized I was listening to the Vectrex 'startup music', but that it was being played slower and at a much lower frequency than normal; the notes themselves were more drawn out than usual. I also realized that I'd been hearing the familiar Vectrex 'buzz', but it too was at a much lower frequency than normal.
The display was off, too. Where I'd normally see text that said "VECTREX" and "GCE - ENTERTAINING NEW IDEAS", I saw only the word "VECTREX", and it was HUGE! The flashing marquee around the title screen was absent, too. When Minestorm finally booted, I realized that I was only seeing a portion of the screen and that it was greatly magnified compared to normal. The mine-laying ship was gigantic, for instance.
At this point something in my head went "Danger! Will Robinson!" I remembered a passage from an Owen Rubin article about Vector arcade games:
"We discovered a problem a bit too late which caused the displays to blow out. It seemed, to correctly draw vectors with proper clipping, you needed to be able to draw a distance off the actual screen. This allowed scrolling and lines to disappear gracefully without the need for expensive clipping hardware. Unfortunately, if you drew too far off the screen, you could blow out the deflection amps in the monitor. Effectively, you were trying to draw on the back of the cabinet, and the hardware would let you try! We thought we had some hardware checks to prevent this, but they didnt always work."
I shut the thing off pretty quick and took a look at the schematics. The Vectrex uses an LM379 op amp to drive the yoke, rather than a pair of power transistors. I realize that the output of the Vectrex's deflection circuitry is trifling compared to that of an arcade machine - the screen is less than half the size. Still, I don't know how hardy the deflection circuits are in the Vectrex or if there's any 'hardware checks' to prevent you from drawing stuff too far offscreen, but I decided to play it safe rather than risk blowing anything out.
I located the "X SIZE" and "Y SIZE" adjustments and turned the machine back on. When the display came up again I tweaked the two screen size controls. I was able to shrink the display, but even with the smallest screen size I could get the display was still too big. I realized the crystal was too slow.
I popped open a small color TV and grabbed the 3.579545 MHz crystal out of it. I'd installed a socket in my Vectrex when I removed the original crystal, so it was a snap to replace the too-slow 1.8432 MHz crystal with the new 3.579545 MHz one. I powered up the Vectrex again and twiddled the two screen size controls. This time I was able to get a display that looked reasonable. I grabbed my DIP-switch multicart and set it for the "Rev. 4 Test" program. I was able to get the machine recalibrated in a matter of seconds; the screen was now back to its original size.
I removed the multicart and started playing Minestorm. By the second or third game, I was seeing levels I'd never played before!
The Half-Trex... is it actually a piece of precision test equipment with a joystick, or just a home video game with training wheels? You be the judge!
If you wanna make one of these beasts for yourself, here's some tips:
If you want, you can then use the "Rev. 4 Test" program and some of the procedures outlined in the service manual to re-align the display after you'd made your modification. If you don't have access to Rev. 4, eyeballing it'll get you close enough.
Copyright © 1999, Larry
Dworsky. All rights reserved.
Last revised: December 03, 1999.