TailChao

320A is only going to net you eight colors over 320B's seven, and then you're stuck with 1bpp draws - but this nearly doubles your fillrate. This was the mode I wanted to use if we were going to shoot for a second title on the hardware, but it's not a performance panacea. If you think you can stylize the graphics well then it's worth a try. Really the biggest restriction with 320B is how many of Sally's cycles will be absolutely decimated by Maria's fetches, so if you have anything with complex physics (for this hardware class anyway) it'll get a little rough. At the end of the day any use of the high resolution modes is going to be more difficult than 160A. But to me the crisp visuals are worth it. Again, still convinced 8-way scrolling in 320B can be done without too much extra junk in the trunk.

None of the currently available Composite or S-Video modifications bother with an audio amplifier, at least to my knowledge. It's not so much that this decision makes them "not designed for cartridge audio" but moreso less robust than desired - including my own board. For something mass produced which is could be hooked up to who-knows-what in a termination and leveling sense, you want a buffer or amplifier there.

Glad to hear the console (and game) made it back safely. I'm convinced this is because of the A/V modification's design - specifically the lack of any amplifier for the sound. Both the TIA and Cartridge's Audio are tapped directly and run to the audio jacks. Redesigning and installing a new mod is a bit excessive, but if you have a small amplifier around (even a stereo) try running the audio through that before it hits the TV. In the long run though, I hope the UAV is revamped with a proper audio path. Especially now that we're seeing more and more 7800 titles with enhanced sound. The quality varies from console to console, and is very sensitive to noise. It's also unlikely to look okay on anything other than an old CRT. I do want to add that the 7800's board layout (and analog section) could be significantly improved, not just for proper A/V but even for RF. Better signal isolation, yadda yadda. Again, I like the hardware but this aspect of it always felt really cheap.

We're now down to the last 25 Copies of the 7800 version.

The little expansion connector for sound boards is a great idea - and nice work cramming the YM2151 and all its supporting components onto the one above. My only recommendation is to add some sort of cartridge audio mixing knob (at least a pot) if there isn't one already. This'll get rid of a lot of headache with the different leveling in both the commercial machines and aftermarket mods.

As my inner fillyjonk continues to terrorize dust and clutter, it's time to make some excess equipment available... I have a handful of DEV R1.0 FoxBoards available for sale, originally used to test Rikki & Vikki. They're programmatically identical to the production boards but feature a Socketed DIP Flash, Socketed CPLD, and FTDI Serial <-> USB Bridge for communicating with the BupChip's debug shell. If you're interested in working with the SOUPER mapper, this and BupSystem are the only way to do so (at least as of writing). It can also be a good platform to develop your own mapper or softsynth. I'll toss in one of our prototype shells so you'll have somewhere to keep it. Because of their scarcity and original expense to make - the price is $100.00 + Shipping but I'm open to negotiations or trading. If you're interested send over a PM.

Egad, was I programming for this thing too?! Welp, not anymore - now it can be your problem! Up for sale or trade is an NTSC Atari 2600 Jr. Janice / Unicorn / 1Chip board. It includes the bottom half of the 2600 Jr. case and button membrane (not pictured above, shown below). No rework has been done on the board aside from adding a socket to use the built-in game feature, the board is fairly clean and the passive components seem okay. So does it work? Yep. Since this is just a board, I'd prefer to trade it for something of equivalent value. Here's some examples... Unrefurbished / not recapped Game Gear, I don't care if it works Atari 5200 Controller, preferably functional ...but if you'd like to make a cash offer (seriously, I'll let it go for like ten bucks + shipping) send over a PM! I'd rather this be sent to someone who will use it rather than into the trash - and will ship anywhere in North America.

No, while I have fond memories of working on Zaku and the Lynx I have moved on from both. Brandon's post from the previous page (and year) should also be noted, so I've duplicated it below... ...this will not change. I apologize for being terse, but just so everyone understands my situation... Please do not contact me about licensing Zaku, porting Zaku to other platforms, or developing software for you without any sort of realistic funding. The enthusiasm is appreciated but I cannot help with any of these things.

The newts have been sold, onto the next highly useless collectable item.

It's 1,000 across three runs - 300, 300, and 400.

It's the Newton's defining feature. Amazingly, the 2100 only took two tries to recognize "Eat up Martha" and that's because my first was in cursive. The 110 on the other hand couldn't recognize over twenty attempts at "Egg Freckles" and I had to give up and use the on-screen keyboard. The software really did get better, but woof - was it rough.

Continuing along the great e-waste adventure, here's some newts... Up for sale or trade are an Apple Newton MessagePad 110 and 2100. The MessagePad 110 includes the box and (as far as I know) all the paper materials and info-video originally packed in it. The 2100 includes an AC Adapter and 4MB Flash Card, but is missing the stylus. You may notice the 2100's battery is wrapped in plastic - the original cells started to leak so I don't want to leave it inside the Newton. These can be removed, the shell cleaned, and replaced with new NiMHs. Both Newtons are in decent shape, a few scuffs here and there and the rubberization is getting a little gummy with age, but they're still serviceable. The AC Adapter works with either MessagePad as does the stylus, which is great since you've got to take the whole lot or nothing at all! Let's give them a test! The 110 is on the left and it's quite plain how much the handwriting recognition improved as the platform matured. I'm looking for $100.00 + Shipping for all of this. As with the Atari 800 that started this fantastic junk trek - cash is great, but I also like to trade! Here's some items I'm looking for which could possibly reduce or remove the price... TurboGrafx-16, PC-Engine, or PC-FX stuff Game Gears, if not recapped Watara SuperVision or TV-Link, especially these since I'm reverse engineering the hardware I'm happy to ship this anywhere in North America. If you're interested, send over a PM!

Noticed an interesting Maria flavor while repairing a late revision NTSC 7800, specifically the variety which supports two SRAM footprints. Huh, RP2C14... did Ricoh fab this one? I didn't notice any functional differences - the timing is identical or close enough for the Maria fetch detection in SOUPER to work. Although I didn't have time to investigate this in detail as the console had to be returned home. One observation is that she seems to run way cooler, so the color drift isn't quite as severe as from the other fabs.

This has been a rough time for eBay 7800s. I had a big problem with inserting my early prototype boards backwards and they all survived, so you should be fine. This is also why I had Tiido add the big LABEL SIDE and BLANK SIDE dummy silkscreen to the production boards. The fact that this is occurring with the built-in game strongly hints something is wrong with the console itself. Yeah, this looks like Maria pulling duplicate entries from the Display List List. The actual problem might not be in her or the internal SRAM, but they're definitely suspect. Rikki & Vikki only has one NMI for the frame tick, actually. I didn't want to start twiddling rasters with our schedule and budget. That was supposed to be for later stuff using the PMC1 and it's fancier line counter. Having to "warm up" usually hints there's some sort of failure in a passive component (like a capacitor) or the voltage regulator. So a few things to investigate if you can't return it to the eBay seller... Voltage Regulator (7805 in the back) Capacitors (only three you really need to replace anyway) SRAM (2x 2KB) Maria

Ah, cool - I don't own a Péritel 7800 so good to know there's an easy workaround. My assumption has been the SCART daughterboard tapped the audio before it was attenuated and mixed through the resistors, is that actually the case here?

Like Commando, Rikki & Vikki uses cartridge expansion audio for music and the TIA for sound effects. Unlike Commando, you can use the left difficulty switch and volume knob in the cartridge to adjust the balancing to your taste. When discussing the "correct" mixing circuit, it's usually referring to the NTSC one - where the TIA and cartridge audio go through a 18KOhm and 6.8KOhm resistor, respectively. The PAL circuit uses 6.8KOhm for both, and while this is annoying - it's still official. Many of the older modification guides use the PAL circuit - and I've seen more than one installation where no resistors were used at all, the cartridge audio left disconnected, and numerous other permutations of nope. Addendum : There's also the SCART / Péritel 7800, which is an official nope.

That's my guess, yeah. Whoever reworked this replaced the regulator with a correct part - but it wasn't able to maintain anywhere near its rated current draw. The shorts were all caused by small solder flecks / whiskers near the rework areas, so they just needed to be wiped off along with the excess flux. Everything seems to be stable now, but as mentioned in the email thread I want to keep it under observation for a bit to make sure. At least as of writing, I'm not aware of any Atari 7800 A/V or S-Video modifications (available as a kit) which actually include a proper audio amplifier. They all tap the signals directly. This may attenuate the audio depending upon the length of the cable and what it's plugged into. No worries, this console at least had a happy ending. As a nice bonus, this was the first time I got to meet the late (probably the last) NTSC board revision. Oh, right... 28 Copies left.

There's a palette caching issue in v0.9.6.3 - changing the console region between PAL and NTSC won't force a palette recalculation, as observed by @Turbo Laser Lynx. A temporary workaround is to open the Color Dialog by navigating to Options -> Maria (Video) -> Adjust Color... (F4) and clicking [Default] or twiddling with the sliders. Restarting the program will also force the palette to be recalculated.

It shouldn't be... ...and upon investigation - the palette isn't getting recalculated upon a region switch. You can force this by navigating to Options -> Maria (Video) -> Adjust Color... (F4) and clicking "Default" or restarting the program. My bad, I'll cross post this to the BupSystem topic.

If you're talking about the 78S05, yeah they're pretty great. I frequently use them in my development consoles. The thing is that the original 7805s should still be able to push 5V @ 1.5A - which is a fair amount of wiggle room when the 7800's components shouldn't yank more than 500mA on the 5V rail. Rikki & Vikki's boards don't usually require over 150mA. Anyway, the regulator was the issue here. Whoever modded the console replaced it with a properly spec'd 7805, but I guess the soldering job was so bad that they cooked its guts or messed it up some other way that it couldn't push over 500mA without the 5V rail sagging.

OH NO - Boring Hardware PSA! So @A Black Falcon sent in both his 7800 and Rikki & Vikki cartridge for inspection after the game failed to start up. The issue ended up being in the console itself - in particular the 5V rail was sagging horrifically. Running a ROM-Only 7800 game already pegged it down to 4.9V, Ballblazer was down to 4.8V, and Rikki & Vikki dropped it to 3.5V (!). Yikes, that's way out of spec! This was a UAV modded console bought off of eBay. In case anyone else runs into this, the console was exhibiting the following issues in other software... Extreme color drift, especially on any POKEY or EXRAM cartridges like Ballblazer, Commando, Jinks, etc. S-Video or Composite may lose sync, but RF still works (if your set's AGC is any good). Colors can drift by attaching more controllers or when pressing their buttons. Some games or flash cartridges may exhibit instability or fail to boot. I'm tracing back the issue now and already found a few shorts. Whoever modded this used solid core wire... it looks like telephone wire and the whole board is a mess - solder flecks everywhere, big chungus blob joints, blech. So please, be careful when buying premodded stuff.

The Atari 800 has now been sold and and shipped out. I'll be listing more things as I slowly sort and test everything in my closet of e-waste. Might be something else good!

I did, yes - and now I've replied to it. My apologies for the delay, things have been a little off the rails for me schedule-wise.

So quick heads up since I actually got 5 - 6 PMs about the 800 its friends : This is open for another day or two in case anyone has the more desirable items on my trade list, otherwise it's going to whoever asked first. I'm packing up everything this evening so I can actually give shipping estimates tomorrow onward.

I tried to draw the graphics to look okay with some variation in the palette, since the color generation is wonky anyway. Generally the shots from my Linytron are what I'd expect from composite - but in any case you'd want the colors over RF replicated as closely as possible. Here's some shots from the Rock Cavern to compare, one of which is old enough that it was before we started tweaking the stage layouts after focus testing... So the aqua is what I'd want. I was trying to figure out how to reliably duplicate this in a new composite mod design but never got around to it. What I can recall is lots of the existing mods just kinda smash Maria and Stella's chroma together with two resistors and this can lead to color drift - if you digitally combine them there'll be less crosstalk. Also, Tiido pointed out that C2 near Q2 in the NTSC board schematic is kinda questionable, and I agree... it looks like they were going to decouple Q2 and messed up - removing this reduced color variation as well.