wierd_w

It's an artifact of the printing method. It's clearly a lithograph print. Basically, it's a dot/line matrix embossed stamping technique that uses sequential stamps. If the registration of the stamp or the paper is off on subsequent operations, the image will have registration issues between colors. (Such as between the yellow and blue above. It also is what is responsible for the "ghosting" around the white lettering above it; the stamp is flexible, because it is a continuous process roller. As such, the stamp does not lay flat in areas of high contrast, because part is much thicker and part is much thinner, and the stretch/deformation is not uniform. A low res scan that blurs these imperfections, scaled up with a neural-net up-scaler (Like Waifu2x and co.) will give you vastly superior results than will a high resolution scan.

Dosbox-X has a config option to load a binary blob into the adapter rom area and execute it, at least according to its reference conf file. https://github.com/joncampbell123/dosbox-x/blob/master/dosbox.reference.conf It has another option for a "Run once" to load another, DIFFERENT blob, that executes before the ROM bios. A little assembler-fu to jump execution with a loaded adapter rom (containing the whole image, not just cart basic), and you are all set.

Nothing gets hardcore program nerds red-faced faster than arguing the teensy details of performance enhancements in language interpreters or compilers. Some will assert "Look, that long integer division with floating point on a processor that lacks a dedicated FPU is gonna be hella expensive, M'kay? That's why I used integer math!" and another will assert "But you did it WRONG!!" or "it breaks my compression program!", or "there's a bug in it!" (because it naturally lacks the same precision as the floating point math routine). Then there is the arguments over using one language over another.. Whoo.. that gets ugly when it gets started. Just let them dish it out to each other, it will end up OK in the end. You just gotta remember that labors of love result in people being attached to their work, and thus when it gets criticized, it can feel like personal attacks. Throw into that the often co-morbid social interaction deficits that hardcore nerddom has, and what appears on one side as raging flame, is really "No, but really-- it's not factually correct!" on the other. This is what I am picking up here. We're adults, we can handle it.

See Brutman's page about the dumper. https://www.brutman.com/PCjr/pcjr_cart_dumping.html This site claims to already have a dump of the system bios. http://retrograde.inf.ua/pcjr.htm From the sounds of it, you need the system to run the bios, which scans the cart space for the signature bytes, and then sets the correct entry points for basica. DosboxX, a "more advanced" version of dosbox, might do what you want. I have to head to work now, but can find out later. Edit: From this GIT commit from last year it appears booting a system bios is possible in dosboxX. https://github.com/joncampbell123/dosbox-x/issues/774 Not sure how mature it is now though. Worth looking into.

but not all divisions will be for graphics that align to real discretized elements (like pixels, cols, rows, etc..) though. Some will be for things you really do want decimals for, like say-- the handful of "Home finance" programs, which would need several significant decimals to handle things like compound interest correctly. The question was, given the full set of all existing basic programs that use FP division/multiplication, what is the maximally needed precision-- and, once you know that, what is the magic number to achieve that level of precision with integer math using a Carmac like approach. This would allow those programs to still run correctly, without invoking the complex set of crazy that is floating point math on a processor without a dedicated set of ALUs for that process (Eg, No FPU).

This reminds me of the basic compiler, which *ALSO* substitutes integer math for floating point math (for the same reason.) Perhaps a revisiting of Carmac's "magic number" based approach would be a good compromise? https://blog.dave.io/0x5f3759df-a-true-magic-number/ Still integer math, but returns values approximating floating point in complexity. A better question to ask oneself-- How many times does FP math end up getting a rounded result anyway, and what is the largest significant decimal depth used? Armed with that, a "not fully accurate, but good enough because the program rounds it off anyway" integer math approach should be possible, that would not break most software.

Ahhhh.. Those polymer clay kids... https://www.craftster.org/forum/index.php?topic=291004.0 They have tested the silicone based parchment paper idea. They suggest 100% transfer is possible onto the polymer clay using the parchment paper. (I suspected; since the silicone treatment will prevent the polypropylene (the bulk material of laser toner) from fusing to the paper fibers directly, and so it should just "peel" off the surface once misted with the solvent.) It works very well with polymer clay (and many plastics) because the toner fuses to the polymer material (being another solvent softening plastic itself!), but it should stick to metal and glass as well. It just might have a habit of flaking off. Perhaps a fabric craft mini-iron would be useful... Horrible youtube video of polymer clay transfer demo. Polymer clay contains a liquid plasticiser, which is what is softening the toner. (You dont have to apply anything, this is raw polymer clay; eg, Premo or Sculpey) In our case, we would want to use something like a hairspray pump bottle, and mist the solvent onto the image, turn it over onto the cleaned metal surface, press, iron with the mini-iron, then quickly rip the paper back. Should theoretically transfer near 100% of the image.

Hmmm... Now I have additional thoughts... I remember seeing awhile back, about people wanting to know about ways to screen on new lettering onto the metal of these shells. I am reminded of the "laser toner + wintergreen oil" image transfer trick, and now I wonder if silicone powder impregnated parchment paper would work well as the intermediate transfer film. (It should not stick to the fuser in the printer, AND should release the fused toner easily...) I had this thought, because if you send shells in for Titanium coating services, you will have to re-apply logos and such. (the surface has to be completely clean before coating will stick, and that means removal of all old printing.) I dont know if my old laserjet 4V still works or not... I could cut some parchment paper and test I suppose.

No, exactly what is on this cutlery.

Upon further reading, I seem to find that pretty much every PicoATX (even the 20 pin versions) do not provide the -5v rail, because they consider it to be a legacy feature that nobody needs. That means that a breakout board for one will need one of these baked into it to provide the missing -5v rail. Should still be inexpensive though.

Mia Culpa. Just sayin, if you want metallic gold colored metal objects, TiN is inexpensive AND durable. That is all. The real trick is getting your hands on the shells to have processed in the first place, which is what the thread is SUPPOSED to be about. (and if you dont want shiny gold, similar processes can give all kinds of pretty rainbow colors instead!)

this place claims to be in CA, and gives free quotes. http://www.pvdamerica.com/get-a-free-quote/ if you can source a bunch of Ti aluminium shells, it never hurts to ask, especially when the quote is free.

aluminum reacts strongly with everything, and getting the gold to strongly adhere is a challenge. TiN coating could probably be done as a batch job reasonably cheaply if you can get enough units to process, and had buyers lined up in advance. It looks quite nice, and is obscenely durable. I mean, look how lovely it is... its meant for metal cutting tools for cnc mills. the normal wear and tear of a computer case is childs play for the stuff.

Obtainable with successive washes of metal salts. Wash 1) Cupric sulfate. This will cause a replacement reaction of the aluminum with copper. Only dip it just long enough to make the surface turn reddish, remove, rinse in ammonia and distilled water, then polish with polishing cloth. Wash 2) Silver nitrate. This will rapidly replace the copper with silver. Once the surface turns lustrous white, remove from solution, neutralize with ammonia and distilled water, then gently polish with soft cloth. Wash 3) Auric chloride. This will replace the silver with metallic gold. The coating will be very very thin. Cover with a nice clear coat. It WILL however, be a true metallic gold coating. More interesting and probably more durable-- Vacuum electrostatic coating with titanium nitride ceramic. (TiN) Not as lustrous, but very nice gold color, and VERY durable. Would not need a surface protectant.

disney soundsource for the win? but then you miss out on the FM synthesis...

You can totally dump the rom basic on a real jr. It shows up as a normal option rom. The same cart dumper used to grab the cart Roma can grab it too, because the cartridge space and the system bios memory areas overlap. You can completely replace the Jr bios with a cartridge in the slot.

Considering the Ti has a Centronics parallel port addon, an LPTOPL3 is possible as well. (It's a Yamaha FM synth hooked to an 8bit LPT port, for vintage laptops without sound hardware.) You would have to write software to control it, but it could be tacked onto a Ti, at least in theory. With clever use of its tone generators and volume control, you CAN do 6bit mono PCM. Some old dos games that suppored adlib cards did that. The LPTOPL3 would basically be an adlib card, just tacked on the Ti.

I was gonna measure the battery, and make a holder for 2430. It's the closest match to stock part. I have some nylon loaded in my fdm, which should be fine. Sourcing conductor is what has me scratching my head. Probably gonna have to do surgery on a pop can, then sand the aluminium. Cans have a coating on them.) I would rather use copper sheet, but apparently that's hard to get on my area.

No, I only recently mentioned it in the "PSU replacement?" Thread. I cited what pins in the 20 pin header you need, and what they go to. I am kinda surprised that the community designs custom VDUs, but there's no PicoATX breakout. It should be a drop in replacement with such a board. (Shrug)

You have a hydro block press laying around? A 3axis mill to cut the two halves of the press dye? Perhaps a cmm probe to make sure your run is good, or have a good point cloud scanner, and know somebody that can reverse the cloud into a proper digital model? A water jet to cut the flat patterns? Because that is the kit you will need to make a few hundred stamped parts. (Used to work in aerospace.) I "might" be able to do the reverse engineer work, but I would do a substandard job designing the press dies. I was a draftsman, not a tool designer. That's black magic voodoo shit. The die needs to be the right draft angle for the exact aluminum formulation, so the springback results in the correct geometry after it releases. (Or you have to press at a specific temperature to get the stiffness parameters of that formulation when pressed) There's industry secrets, hard earned experience, and formula work involved. Then there's heat treat you have to conside after pressing. What condition do you want the metal to be taken to after forming? Any surface treatments? This is not at all straight forward.

Dosbox does Jr emulation too. IIRC, you can also define cartridge roms. Let me check. Yup. https://www.dosbox.com/wiki/Configuration:DOSBox

Looks like it uses Hercules monochrome. Does it have an LPT port? An old iomega zip100 could probably live there, or perhaps an old LPT IDE hard drive (good luck getting something new to live inside though.) The people at Brutman's PCjr site were able to get a version of guest.exe (the dos TSR for a zip100 parallel, like linked) running on an 8088 based Jr, so I think your Beauty PC should work too, if it has an LPT port. Not sure how its upper memory is laid out though. IIRC, Guest.exe is a memory hogging beast. Looks like there is a smaller driver called palmzip.sys that works on 8088s, and works with dos 2.1 http://leute.server.de/peichl/palmzipe.htm Just a thought. I happen to have a beat up zip 100plus floating around. I don't even know if it still works or not. Edit the LGR Oddware for this beast says it is a Toshiba T1000 with a different screen. A T1000 has an LPT port, so you should be able to use such a thing with it. Claims it is CGA capable, but that display is clearly monochrome. I would stick with hercules aware games and run in that mode.

External ATX is unelegant, but is a good showing that the signals will work fine. PicoATX is a mass produced consumer product, and a breakout board would be cheap to manufacture.

I understand that the cover and the game itself are not anywhere close to the same thing... But that level of quality was simply not possible. Like OLD CS1, I viewed them as what the game makers envisioned when they made them, and how you are supposed to 'think' when you play them. Mechanically, many of those old games make absolutely no sense unless you read the manuals, and understand the story elements they provide. Take for instance, the 2600 Yar's Revenge. It came with fancy cover art, but also a fancy mini-comic inside as part of the manual. https://www.atariage.com/comics/comic_thumbs.php?MagazineID=48 Lovely, and beautifully made. It showed that the game's designers intended there to be much more than just a buzzing noise, some static, and lots of bloops an nibbling sounds. There's only so much you can do with the 2600, but what they could not put in the game itself, they put in the box with it. These days there's this view that the makers of games aren't out to tell dramatic stories, just out to fleece your wallet with microtransactions and in-game eyecandy. It was a completely different aesthetic.

Excellent! This means a simple breakout board for a picoATX (20pin!! NO 24 pin!!) would work great with a Ti 99/A4. That would fit neatly inside the case. (PicoATX is smaller than the original power board.)