alex_79

If you search "zaxxon atari 2600" on youtube you'll find a lot of gameplay footage taken from real TVs (both CRTs and LCDs) or capture cards. They all bounce to various degrees.

Remember that the left difficulty is used to freeze the game when in "A" position (which is "B" position on the Gemini, as they labeled the switches backwards!) That's a bug in the game itself and not related to the Gemini. The game does not initialize the decimal flag correctly. The status of the CPU flags and registers at power on is undetermined and the game can randomly glitch like you experienced with any 2600 console. Just keep power cycling until it starts correctly (briefly trying another cart might help).

No, you can't swap the chips, as their pinout is different and the PAL TIA requires a separate 4.433618 clock for color generation. You'll have to modify the board by cutting and rewiring traces, add the components for the extra oscillator, and replace the 3.579545 MHz quartz with the 3.546894 MHz one needed for PAL. CX2600_NTSC_schem.pdf CX2600_PAL_schem.pdf

I've been asked via PM about the console that came with the red joystick I posted above. I just reply here in case someone else is interested. The console is labeled "Diablo 128" and, apart for the weird color scheme, is no different from the other 4-switch clones mentioned in this thread. It has 128 games built in. It was sold in Italy, and both the box and the console control panel has Italian text. The following pictures are from on-line auctions, as I only own the joysticks and not the console itself.

Not surprising. They surely didn't require tight tolerances. If it (sort of) worked, then it was good enough!

Actually, the shaft is shorter and has a different shape compared to a cx40. So you'd need to modify it quite a lot to make it fit. Also note that there's no rubber boot here, as it's all part of the top half of the shell, made of (cheap) hard plastic. Even the feet are part of the mold! This one is from another clone console and while similar in construction, it's better made. Not good, but it's at least usable.

The exact reasons are beyond my basic knowledge on this. It could be due to differences in timing, or maybe the CPU slightly pulls some of the pins high or low even when it's not actively driving them. Normally, that weak pullup/pulldown effect would be overridden by a chip actively driving the bus, but in case of the undriven lower bits during a TIA read, it's enough to change their logic level (which normally persists for a few microseconds from the previous operation because of the capacitance of the lines) before the CPU actually reads it, and thus reveals the bug. Yes, even when games with such bugs do work, they do that "by chance". The condition that allows them to work is unstable and unreliable and can be perturbed by many factors.

I'm pretty sure that those screenshots (and the number of built-in games) is just false advertising. Those are 2600 clones, with a number of pirated games installed (typically 128, 256 or 512), mostly badly converted to PAL from the NTSC versions and therefore showing wrong colors. They were sold with different brands and names (probably from the same few manufacturers). Here in Italy, I remember seeing them for cheap in the toy aisles of supermarkets and malls in the mid to late '90s, before being replaced by famiclones. I have a couple of them, boxed, which I bought locally for cheap (less than 10 EUR) a few years ago. They're not really worth more than that, IMO. But, if you happen to get one for cheap, it is worth checking the installed games, as you might get lucky and find some interesting one not yet dumped. Just like the consoles, those joystick are cheap and low quality. Good luck trying to hit diagonals with any precision using them. Look at those "switches"!

Thanks for the confidence, but I wouldn't define myself an hardware guy. Just a random guy posting on the internet. And as such not really trustworthy... With that in mind, my understanding is that anything can expose the bug in those games, from the chips and components inside the console, to the hardware in the cartridge itself, and even interferences from nearby external appliances. The TIA is not driving 6 out of the 8 data bits (because there's no circuitry in it to do so), which means that anything else can introduce slight perturbations and alter the result. And that includes the CPU.

Can you take a picture of the connector at the end of the console cable and the adapter you're talking about? What you have is a SECAM console, for the french market. The RF cable should already have the right connector to plug directly into the antenna socket of any European TV, so no adapter should be necessary. You might need to manually set your TV to "System-L" and "SECAM" color format when searching for analog channels, assuming that it is compatible with the SECAM-L standard. At least with the LCD TVs I have, an automatic scan won't find the signal from my SECAM console. I have to set it manually (on channel UHF 36) and specify System-L and SECAM in the options.

I don't know if they fixed that in the manual, but from previous threads it appears that the jumper configuration was determined for NTSC 6-switch models, and assumed to be the same for all other 2600 boards using a CD4050 (e.g. Juniors and PAL variants), which is not the case, and caused issues (out of order "LUMA" values) for people who did the install following the instructions. Another good reason to trace the actual board you're working on like you did, and find convenient places to tap the signals from. Also, I think the CD4050 is supposed to be left on the board in case of the 6-switch model too, with the mod board soldered on top of it (piggybacked). Else the joystick triggers (which are buffered by the CD4050 on a 6-switch) wouldn't work anymore.

It's the same for any 2600 variant from UK, including 4-switch and Junior ones.

No, the difficulty switches are not connected to the CD4050 (check the schematics in the 2600 service manual). They are connected directly to 2 pins of the RIOT (6532) chip. They connect those pins to GND when in "B" position, and leave them disconnected in "A" position. Oxidation/dirt inside the switches themselves might result in them not making contact when closed, and so the games would see them always in "A" position. So, if you're sure that they are mechanically good, and there aren't cold solder joints or broken traces on the board, then the culprit is likely the RIOT chip. BTW, you posted in the wrong sub-forum. This is for discussing about programming games for the 2600. You should have posted this question in either the main Atari 2600 or the Hardware sub-forums instead.😉

No, the PAL TIA has a different pinout and also requires an additional oscillator. They're not interchangeable without modifying the boards (replacing the crystal quartz, cutting and rewirng traces).

Yup. And the same is true for Kool-Aid Man! Read the manuals! 😃

It is very important to specify where you are located, as different countries used different TV systems and connections. After a quick search, I found a match for that model and it appears that your TV is PAL, so I assume you're in Europe. The front buttons are presets. You can tune each of them to a specific channel using the trimmers and switches on the side of the TV. The white switches below each trimmer select the band. If you're in mainland Europe, your console should be in "PAL-B" standard and output on channel VHF 3 or 4. Set the white switch to band "I" and tune the trimmer (with the console powered on and a cartridge inserted) corresponding to the preset you're using until you find the channel. If you're in UK or Ireland, then your console should be in "PAL-I" standard and outputs on channel UHF 36. In that case set the white switch to "UHF" band.

Stella release page says that for Windows you need the "Visual C++ Redistributable for Visual Studio 2017". Did you install that?

On the 7800 console I had, the metal strip was all scratched and dented too, and even partly detached and bent on one side. I replaced with a piece of "woodgrain effect" adhesive vinyl I had on hand. The bottom half of the case was all cracked (with missing parts too), so keeping the originality of that particular console wasn't really a concern. What about a custom laminated sticker with an "Artie the Atari" strip? Looking forward to seeing the final reassembly. You really gave dignity back to that poor console.

Some info can be found here:

You have four ports in total, so the max is four devices at the same time. E.g: three controllers + one savekey. Anyway, the supported configuration (number/type of controllers and other peripherals, and which port each of them needs to be connected to) depends on the specific game.

Seconded. Without a cartridge, the 2600 doesn't output a standard video signal (there's no vertical synchronization, which is handled by the game code), so the TV might not find it even in case it supports analog channels. If the TV finds the channel without a cartridge, the result could be a black screen, or a bunch of colored vertical bars with or without some continuous sound.

The matrix is 3x4. The columns are connected to pin 5,6 and 9 of the controller port (TIA inputs for paddles and joystick button), while the rows are connected to pins 1,2,3 and 4 (RIOT I/O). The only two pins left are 7 and 8, which are power and ground. There isn't an additional I/O line to increase the size of the matrix. Also, multiple key pressed at the same time can cause "ghost" key press to be detected, or prevent some of the key to be detected, because of interactions between the rows and columns.

Do you have several driving controllers and none of them works?! What are the symptoms exactly? (Remember that the ONLY original game that will work with a driving controller is Indy 500. There are a handful of hacks and homebrews that can use it too). The mechanical encoder is very simple and it's uncommon for it to have issues. If the controller is malfunctioning I'd suspect the cable.

Updating the firmware from the SD card is only possible if the existing firmware is working. (Similarly, upgrading off SD card/wifi for the Unocart/Pluscart requires a working bios. If the cart is unresponsive for some reason, you need to use the usb port, if present, or a ST-link programmer) In some circumstances the firmware can be corrupted, and then the only option is to use the usb cable and the PC software to revive the cartridge. Moreover, the PC software + usb cable is the only way to flash the cart in "single game" mode (useful for developers) and to revert back to multicart mode. So the information in the manual is still relevant. I suppose that when a new firmware will be released, the instructions and the loader for upgrading from SD will be posted again, like it happened with the 1.06 beta release. Anyway, the cart comes already flashed with the latest bios and ready to use. You don't need to program it. The Harmony subforum here on AtariAge is the where you'll find the most up-to-date info.

Test the trak-ball with the switch in "JS" position. It emulates a joystick in that way and it should work with most 2600 and 7800 games. The experience won't be as good as when playing one of the hacks in real Trak-ball mode (switch in "TB" position), but at least you can verify if the controller is in working order or not.