Jump to content


New Members
  • Content Count

  • Joined

  • Last visited

Community Reputation

33 Excellent

About gtoal

  • Rank
    Space Invader

Recent Profile Visitors

1,522 profile views
  1. Tony Holcomb and Helmut Mueller also have templates that they'll share directly with developers if you PM them - they look better designed than mine but require a larger size of card. My stuff has moved to http://gtoal.com/vectrex/DIY-Vectrex-Cartridge-Box-Kit/ because I had trouble uploading the final files to the FB file area. Bill of Materials ================= silvered card 11x17 or larger https://www.amazon.com/gp/product/B01AYBS5JW/ white card for insert (or something a little stiffer) 11x17 or larger https://www.amazon.com/gp/product/B06XCDBD2C/ Krylon Crystal Clear https://www.amazon.com/Krylon-K01303007-Acrylic-Crystal-11-Ounce/dp/B00023JE7K/ Krylon Easy Tack (very optional) https://www.amazon.com/Krylon-K07020007-10-25-Ounce-Repositionable-Adhesive/dp/B000A8AYO4/ Masking tape (eg Lowes) https://www.lowes.com/pd/ScotchBlue-1-88-in-Painters-Tape/1002792518 A3 printer https://www.adorama.com/icaix6820.html https://www.ebay.com/sch/i.html?_from=R40&_nkw=hp%20photosmart%208750%20printer&_sacat=0&rt=nc&_udlo=100&_udhi=260 A3 vinyl cutter (or 12in x 24in vinyl cutter) https://www.iloveknk.com/current-specials-2/ --------------- Before making and assembling the box, you might want to practice on the easier tray insert - also it makes it easier to fold and glue the box if you have the tray inside to provide a little extra solidity to push against when gluing the flaps. Cut the file vectrex-tray.mtc and fold & glue it. --------------- When I set this up I was using 11x17 card which was the largest size I could find at the time which would fit both my printer and my vinyl cutter. Later I discovered that 12x18 paper is available. The unfolded box just fits in 11x17 but the tuck-in flap needs to extend right to the edge of the paper in the area that would be in the non-printable margin in most printers. Also cutting right to the edge of the paper can sometimes be problematic, so we just don't cut right to the edge. So the cut file I've made does not cut all the way around the box - there are two areas where you need to release the shape from the enclosing card by cutting it away with scissors. I'll mark up a photo to show exactly where in case it is not obvious. The image is sized at 300dpi and should print without hassles caused by exceeding the printable area into the margin (you must print with the image centered). You don't need to select borderless printing, which isn't supported on all printers. (The old version of Photoshop 5.5 that I bought many years ago is now available for download on the net at https://winworldpc.com/product/adobe-photoshop/5x - printing the psd file from Photoshop would be the most reliable way to print accurately) Turn on 'registration marks' and 'corner crop marks' in the windows printer driver. Be sure to set paper size to 11x17in portrait, and for paper type if there is nothing better that matches, try 'other specialty paper' which *should* slow down the printing to a speed less likely to smudge or have alignment problems. Make sure the "SCORE/CUT LAYER" is off. It's not accurate and is only there as a visual guideline. The PSD file contains 3 different versions of the central image. Make sure only one of them is selected before printing. Print the file tailgunner-box.psd centered, with 'other specialty paper' for slow careful print at 300dpi When you first try this you might want to turn off the image to save ink, and print in draft mode - it may take a few iterations around printing and cutting before you get everything perfectly aligned. The card you print to assumes 11x17 inches. If using a different size you'll need to make small accomodations when cutting. You will cut using vectrex-box.mtc Align your cutter's origin with right edge of the card, level with bottom alignment line. When you cut, there will be an L-shaped alignment mark '|_' which should be as close as possible to the right-hand edge of paper There's a printed registration mark '---' on both the left and the right of the print, and that should be aligned to the cutter's axis/zero point. I designed the cut file for use with "Make The Cut". If you are using another application, you'll need to convert the cut files to your format. I've included as many exported formats as I can but you should expect to have to put in some effort to convert to your cutter format. Please feed back any files you create and I'll add them to this archive. The MTC file includes an alignment layer which you can use as a rough position check when you first start cutting. It'll score central spine an extra time. Expect to do a few cuts to get the position right while you're getting set up. A common printing and cutting technique is to tack the material to the cutting mat. You probably don't want to do that here. The design is such that it will cut OK as long as the near and far edges are taped down. Tape near edge liberally (up to 1/2 in of overlap) and the far edge more conservatively (eg 1/10 in) Tape the printed paper over the path of any rollers that run over the print area. (when you print on the silvered card - you'll skip this if practicing on cheap plain card) Clean the rollers with lens wipes and let them dry. The cut file has a few layers. They should all be turned off and then enabled one at a time - so enable the scoring layer first and run the cutter on very low pressure. Next, make the actual cuts - possibly in several passes with increasing pressure. Maybe a maximum of 3 passes to preserve alignment throughout. 25 pressure @ 50 speed, 30 @ 100, and 45 @ 150 worked for me. But these parameters all depend on your specific vinyl cutter and you'll have to work out what's best for you by trial and error... When it's done and removed from the vinyl cutter, cut the 4 places on the edge of the card where the cut lines do not extend to edge... then remove the cut-off edging. If this is a final box, spray the flat card with fixative and wait 12 hrs before folding. Fold and glue. Not too much glue if that gets the card wet. Clear 'mucilage' might be better than the white school glue. When assembled and tucked in, you might notice a very slight bowing of the front of the box - if so, you can fix this by spraying the tuck-in flap *only* with a very light coating of easy-tack repositionable spray, so if this is wanted, mask off the rest of the box exposing only the tuck in flap, and pass a VERY LIGHT spray of repositionable Krylon Easy-Tack over it from a distance. When you put the tray insert in the box, be careful not to have the tray stick to any wet glue on the box! Please do take a photo of your box and post it on our Facebook group! And now that you've mastered building one from this design, edit the psd file and create a new box for your own roms! You'll need the fonts in "Poppins.zip" which are an approximation to ITC Avant Garde (LT or Pro?) Demi, which is my guess as to the family used on the original boxes. Graham
  2. gtoal


    Well, this actually shows promise. Admittedly I made a bit of a rough job of it and there's some visible glue on this test scrap, but it is holding really well and I think with a little practice at application this could be a winner.
  3. gtoal


    Good hints! I had been thinking maybe you were using LOCA (liquid optically clear adhesive - the UV curing stuff) so I got one to try - will report back when I do. I'm lucky I bought a bunch of gels last year that I haven't used yet - wasn't aware of the QA problem. Typical of Chinese suppliers.
  4. gtoal

    Basic question

    Apart from being smaller, it's very like a vector arcade experience, and the flicker is dependent on the quality of the programming - the vast majority of games have no noticeable flicker. A loadable flash cart is an essential. There are around 250 freely and legitimately downloadable binaries that you can find. Actually many more if you include scene demos rather than playable games.
  5. Actually that was also a static binary translation, not an emulation. In fact it started off as the same one discussed above; and the asteroids was initially the one that David Welch produced, but I later heard from Thomas that he had rewritten the translators himself so I believe the final released versions used Thomas's translators (which haven't been released). Since there's a lot of interest in this stuff again I've taken up a background task to produce a full suite of binary translators - 6502, 6809, z80 and CCPU. Maybe also 68K if I get those others finished satisfactorily first. Work in progress is at http://gtoal.com/SBTPROJECT/ - I'll move them to github only after all four are working to some extent... If you want to browse the code to get a feel for the techniques involved, the 6809 one is probably the nearest to completion at the moment. I did a write up a few years ago which you can still find at http://gtoal.com/sbt/ PS Tailgunner and asteroids and others will be available for the Vectrex on our new "PiTrex" cartridge which uses a Pi Zero WH as its CPU to drive the Vectrex, with the 6809 being halted under control of the cartridge. Described elsewhere.
  6. gtoal


    ChopperTheDog: I didn't notice when you posted this that you were doing colour inlays in your overlays. I'd considered that but thought (without actually testing it) that they would be likely to pop out under minimal wear. What are you using to glue them in place robustly? Or are you sandwiching them under a clear layer, in which case you would need a glue that invisibly attaches the layers together? I've only used my cutter for monochrome overlays, with a paper printed layer on top for the opaque border. I'ld quite like to try some dual colour ones if you could share your secrets on how they're put together! Graham
  7. gtoal

    Misaligned vectors

    The VIA is the 6522 which is effectively the vector generator on this machine. different models may have marginally different timing, and Clean Sweep is such a hack that it is extremely sensitive to timing differences.
  8. gtoal


    I posted a design for a laser-cut version at:
  9. gtoal

    Misaligned vectors

    by any chance have you replaced either the VIA or the CPU from the factory-installed ones?
  10. you might look for Malbans "weird vectrex" test rom. (http://vide.malban.de/forums/topic/peer-testing-notes-findings-suggestion-etc) probably the cause is that those are cartridges which implement bank switching, and your Via is non standard. G
  11. https://arxiv.org/ftp/arxiv/papers/1811/1811.02035.pdf also a rewrite in C is linked to at
  12. Pretty sure one of the schoolkids on Scratch has posted an algorithm like that. It'll be in this collection of maze generators: https://scratch.mit.edu/studios/403013/ Look for "Ellers algorithm" http://weblog.jamisbuck.org/2010/12/29/maze-generation-eller-s-algorithm Here's another one... http://people.cs.ksu.edu/~ashley78/wiki.ashleycoleman.me/index.php/Sidewinder_Maze.html (there are others that work similarly, can't remember the names right now. You can read about maze algorithms in general at https://www.astrolog.org/labyrnth/algrithm.htm ) Just keep scrolling down, the page refreshes with more when you hit the bottom. (All those Scratch programs have source available, just select it and click "See inside")
  13. It's no great secret - was written about here some time ago: http://www.digitpress.com/library/interviews/interview_paul_allen_newell.html G
  14. Yes indeed 🙂 ( https://www.ebay.com/itm/40pin-Male-to-Female-Ribbon-GPIO-Cable-for-Connection-Raspberry-Pi-3-2-Model-B/303202023930 although a little pricey. I have some parts on order to make my own. Beware IDE cables have one connector on the wrong side.)
  15. Because of the hassles of porting complex programs to a bare metal environment, my current PiTrex development environment is actually a Pi 3 connected by a ribbon cable, rather than an embedded Pi Zero. This is fine, but it introduces a caveat when I say that it appears that Mame (advancemame to be precise) is fast enough on its own and that we won't require multiple individual heavily customised static binary translations to make arcade games work. (The caveat being that Pi 3 cores are a bit faster than a Pi Zero core). On the Pi 3 I can dedicate one of the four cores to the PiTrex and run it with most of the OS overhead turned off. However we haven't turned off *all* OS overhead, so this environment can still produce the occasional screen glitch, and isn't as completely rock-solid as when we run just the application alone on a Pi Zero in 'bare metal' mode. Anyway what I'm getting round to saying is that I expect most of the vector games that AdvanceMame can emulate will run on the PiTrex. I.e. for now we don't think we require the more difficult step of producing individual binaries through static binary translation. On the other hand some of those games do put a *lot* of vectors on the screen at once, so it's quite possible that the emulation could be fast enough but the display hardware incapable of keeping up. One last point to set appropriate expectations: the PiTrex seems to bring out the worst in the Vectrex regarding the infamous buzz... It's not continuous any more, but it's loud and in direct proportion to what is being drawn. (If the screen is blank, the Vectrex is silent). Haven't tested with a no-buzz kit yet but I'm assuming it will work the same here too. Graham
  • Create New...