gdement

This reminds me of something I was wondering about. PAL 7800's have a socketed Maria, don't they? And they have a socketed boot ROM. So suppose you were to plug in an NTSC Maria in a PAL console. Would that make it usable with NTSC? They sell Maria chips at Best Electronics. Depending how much needs to be done with the supporting video circuitry, then I wonder if it might be easier than modding a socket-less NTSC console.

If the PAL version is anything like the NTSC version, then there's some tradeoff between the 2600 and 7800 color adjustments. If you're only going to use your 7800 for 7800 games, and you don't care about the image quality with 2600, then you might get a better picture by turning the 2600 color pot to max impedance (essentially shutting it off). There's some interference between the 2600 and 7800 color circuits, and those knobs are used to control the tradeoff between the signal strength of each. So blocking off the 2600 signal might help a bit with the 7800 image. I've never tried doing this though, so I have no idea how much difference it would make.

I think it's important to point out that "model 1" Genesis really just refers to the plastic shell. There's several different mainboard revisions in those things, some better than others. I do believe that mid-life revisions of the model 1 are probably very good. Although if you get to the late model-1, then those have sound quality problems. Mine, which started having problems at only 1-2 yrs old, was an earlier revision (VA3). I've only owned one personally but from photos online I think other VA3's are just like mine. It doesn't look like good build quality to me: The VA3 apparently didn't even work right as designed - Sega did rework on them, including soldering parts to places where no proper solder pad exists. One is literally soldered into a hole that was dug in the PCB, there's a cap on the back of the board, another soldered to the CD edge connector, etc. I find it odd that it seems not to have through-hole vias, and they instead soldered the components separately on the top and bottom of the board. Maybe this was normal at the time, I have no idea. My board had 1 very bad solder joint, which would be obvious to anybody who looked at it. They also used Chhsi capacitors, which is a notoriously bad/cheap brand. I don't know if this was known in the 80's though, and those caps didn't cause a major problem on my Genesis until just a few years ago. Basically, I think this board was hung together with bailing wire and shipped to meet early demand. That's not an uncommon problem with new consoles. Later revisions are probably a lot better.

Those RAM chips are in SOJ packaging. I've never built a RAM cart but I assume you would want DIP style, not surface mount.

My early Genesis hasn't really worked right since it was about 1-2yrs old. Recently I fixed a video problem, which turned out to be caused by bad caps. The bad caps are Chhsi, which are a notoriously bad brand. Every single one I removed tested way out of spec, while the other brands in the same Genesis were still good. If you see a Genesis with Chhsi caps in it, you can assume they're bad. They're also a pain to remove properly. I'm not sure if the board even has a through-hole via or if it's just 2 separate solder pads on top/bottom. I suspect the latter because Sega soldered the tops separately. I had a hard time not damaging them. It's much easier to do the same job on modern PC motherboards. I also noticed a very bad solder joint under one of the video RAM chips. The remaining issues are lockups/sound glitches (Z80 problem?) and controller glitches. I noticed a particular component that triggers instability when I nudge it, so maybe I've found the problem.

I think what you're missing is what will he realistically sell in a 60 day period? 1000 units? 2000 units? So let's say he sells 4000 units in 60 days. He has to deal with packing and shipping them, etc. A lot of bother for $2000 dollars additional profit on a $250,000 order. Just selling everything off to the scrap dealer is a clean and easy transaction. What happens if in those 60 days the scrap dealer changes his mind? Or the price of gold goes down? Then he's gambling a quarter of a million dollars on the hope that he can make an extra two grand. It just doesn't make sense. Bill Unless I'm mistaken, he started with something like 2-3M carts, so he sold quite a large number before raising the price. Selling another few 100K isn't out of the question. I agree that small orders shouldn't be worth the time, that's why I'm mystified by their marketing strategy. They've been doing this for years. He makes no effort to encourage big orders, no quantity pricing on his site, and people who called to ask for it were flatly refused. If he doesn't want the bother of handling small orders, then I don't understand why he specifically targeted those type of orders for so long. He jumped from $5/cart to scrap, without trying anything else in between. At $1 in large quantities, he'd get a surge of business from resellers, and at least a couple thousand more from homebrew converters. At minimum, he should have offered deals on case quantities, which are something like 72 games already packed. Just slap a label on it and ship. Any normal distributor would offer a steep discount on those. Very steep if they were overstocked and anxious to unload them. The only thing that makes sense to me is if he got panicked about the price of gold. That explains his recent decision, but not why he went for years without quantity pricing. They apparently didn't know what hardware was in the carts until they were told by people asking about them. It's nice that the POKEY chips will be saved for a while longer, but they won't sell at the prices he expects. The Jinks carts just have an 8KB SRAM in them. That's nothing special but maybe he'll get a better price on them from a recycler that can desolder the chips. I doubt there's much value in those chips though. If they were trying to hold back stock that they can sell to collectors, they should have kept a small amount of each title, in proportion to their past sell rate. Not 40k each of just 2 of them. So I assume they'll be recycled in a separate deal.

I guess I'm preaching to the choir here, but: It's funny to me that they jumped from selling singles @ $5/each, to selling 1M carts as scrap. I wonder why he never offered quantity pricing on larger lots, especially on cases that wouldn't need to be repacked. His whole marketing approach was geared towards selling singles, like some retail store. Totally backwards from what he needed to be doing. From what others have said, apparently there was a time he sold them for <$1 each, but changed his mind when people were buying large lots and reselling them for profit. With the volume he had on hand, he should have been openly working with resellers, not hating them. Did he think that everyone in the world looking for Atari carts was going to find and buy them through his obscure web site? Resellers would have helped him sell more carts. Raising the price on singles doesn't seem unreasonable, given the time spent on an order like that. But if he had offered case quantities at $1/game, they'd probably have all been sold by now. Even I'd be interested at that price, and I don't have any immediate use for them.

339 boards are no real loss as the 565 board with SRAM on them are a better donor board anyhow. Jinks is a mix of 339 and 565 actually. True. The only special/interesting thing to me about the 339 is that it has a solder point connected to the Maria HALT pin. A long time ago somebody (I think supercat) mentioned the idea of using that for separate graphics/code banks. The 565 doesn't have a point connected to this. The benefit would be that this way, you essentially double the addressable range. While Maria is active, it could address 32KB of all graphics, but when the 6502 takes over it lands in a different address space. This would only be useful if 64KB EPROMs are 28-pins... I think they are. 128KB and above won't fit though. I assume the 339 Jinks carts are older builds, and that the later production that O'Sheas has are probably all 565's. I've never actually bought any of them though. As you probably know, the 339 had some issues with RAM so I think in later production it was only used for non-RAM games.

This is disappointing. It would have been nice to be able to get like 1000 copies of something for homebrew conversion, if and when the need arises. It sounds like 2 board types will still be available, but they probably won't be cheap enough to be worth using. I can see some potential use for clean/unused C100339 boards, but that one apparently won't survive. That's a pretty common board though.

These games aren't rare. Even if they all melted down I don't think it would be a big deal. The utility of these cartridges is reflected by their market value - which isn't very high. I'd like to have a cheap supply of homebrew PCBs but frankly a custom board would be easier to work with. Retrofitting an Atari board is only worthwhile if they're cheap. The POKEY would be nice but that board has other limitations. These games have yet to enter circulation. Even if most of them get scrapped, I don't think it would affect near term value or his ability to sell what's left. They'd continue to sell at about the same rate as they do now. Scrapping the excess just removes carts that won't move in the next several years anyway.

Sorry. I think I was one of the first people to mention the POKEY carts in this thread. I should have PM'd it or not brought it up. So now they're worth $15/each? Where does he get these prices from? I guess he wants to ship singles for the next 100 years.

Try a different TV. What you're describing sounds like "input lag" or "output lag" or whatever people prefer to call it. Most modern TVs don't display the image as soon as it's received, it goes through some buffering first. Some TVs have a more noticeable lag than others. Sometimes you can make it better by turning off some image enhancement settings on the TV. A traditional CRT won't have lag, it displays what's received instantly. I didn't notice the lag on an LCD I was using with my NES, except I discovered I suddenly was very bad at the SMB2 slot machine bonus game. I think it must have been very slight lag throwing off my timing. But next time I hook it up to a CRT I may find out my reflexes are just lousy.

Interesting. So if it only reacts in 1/30 sec intervals, then I guess scan timing would be impossible on the zapper. Using the 7800 as an example (just because I'm more familiar with it), there's 80 6502 clock cycles in the visible range of each scanline. (going by the "7800 software guide" PDF) I got curious so I typed some untested, probably broken pseudocode for polling a light gun (see below). I don't actually know how the 7800 gun is programmed, I'm just guessing here. It looks like it could be sampled at least 5 times per scanline, or 6 if you don't mind them being unevenly spaced. The uneven spacing is consistent so the code can still resolve it accurately. Either way, 5-6 samples isn't very many. However, you could stagger the starting point on each scanline to get more useful resolution. That would give 80-96 horizontal positions checked within a vertical tolerance of 16 scanlines, or 40-48 per 8 scanlines, whatever. That might be reasonable, but I don't know if they really programmed anything that way. You might be right about the gun having assisting circuitry. I didn't think so, but the 6502 isn't as fast at this as I thought it would be. <Maria is disabled while scanning> <set scanline counter to 0> NextLine: ldx #5 ;sample counter. We can do up to 6 samples per scanline <use scanline counter to calculate an appropriate stagger value for this line> <use WSYNC to sync with start of scanline> <use nop's or some other method to stagger the start of sampling> ScanLoop: lda GUNSIGNAL ;6 ;(RIOT access will slow it down by 50%) I'll imagine it returns 0/1 depending on result beq SAWLIGHT ;2 (when not taken) dex ;2 ;unroll the loop here to squeeze in 6 samples instead of 5 lda GUNSIGNAL ;6 beq SAWLIGHT ;2 dex ;2 bpl ScanLoop ;3 (taken until done with sampling) ;23 cycles per 2 samples (we have 80 available after WSYNC) <increment scanline counter> bne NextLine (always taken) SawLight: <decode the counter and decide where we hit the screen, etc> <X tells us which sample on the current scanline was a hit, and the scanline counter tells us what line> Unrolling the loop above does mean the samples aren't evenly spaced. To make them even, roll it back up and get 5 samples instead of 6.

Yeah, I've had trouble with mine also (NTSC). It works okay on a conventional CRT television, but I thought something was broken when I hooked it up to my TV tuner card in my PC. I could hardly see anything. The signal probably isn't quite "standard", and modern TVs/tuners might be more picky about it.

Every LCD has input lag (not the same as refresh rate), so it wouldn't work without reprogramming the game to be aware of that issue. At minimum, there's the lag of waiting for the entire frame to be transmitted to the screen before it can be displayed. This means at least the top portion of the screen will be almost 1 frame behind, even in the best possible case. In that best case, some NES games might be tolerant enough to work. Anyway, I've only heard of very few LCDs that might be this fast. Nearly all of them have more significant lag. I think new games could be programmed with a calibration screen, which would measure the lag of your display and keep it in mind during gameplay. This would require the lag to be consistent, and it might not be with some displays. They'd also be forced to use the light/dark method obviously. Is it really any different than other light guns? I thought the difference is just in the programming. I don't see why a NES game couldn't be programmed to use scanline timing. It apparently wasn't convenient though, probably due to the way that system handles it's graphics.

If they're both the same model then the parts are probably interchangeable. The adjustment knob would be a potentiometer (variable resistor). If you knew the range of the pot then you could replace it with anything that has similar specs.

I think my most valuable games are actually computer games, not console: Ultima: The Second Trilogy - Limited release version with all cloth maps and a sticker on the box. I'm a bit confused about this one, because an article online says that this version was only available by preorder. I got this as a gift, and I could have sworn it was already on the shelf by then, I never had any impression my parents preordered it. Until recently I didn't know that most of these had paper maps. I'm glad I lucked out on that. Ultima: Martian Dreams - not as big a deal to me as the 2nd trilogy, but apparently this is somewhat rare. I couldn't find anyone selling it on ebay except 1 high-priced store listing, so maybe it's value compares with 2nd trilogy. My console games are mostly cheap. I don't know which of these would be highest: Michael Jackson's Moonwalker (CIB) - This went crazy a few months ago Phantasy Star (CIB) Pac Man Jr (7800 homebrew) It gets cheaper from there.

I don't know the person you mentioned, but I think many non-programmers have a misunderstanding that stems from the word "port". Many seem to believe you can just plug in the ROM for arcade Pac-Man into some sort of "conversion" process, and get a Coleco/Atari/whatever ROM out the other end. They don't realize that the game has to be written from scratch, 100% your own code, and is thus just as difficult as any original game would be. Because it looks like an existing game, they think you copied something. Many people also don't appreciate the technical challenge involved with writing a computer program. There were several washouts in my earlier programming classes, and I'm pretty sure none of them realized what they were getting themselves into. Writing a complex program is something I'd compare to writing a novel. It's sort of like art for engineers.

I continue to die at the house of troll things or whatever they are. Last time I tried just jumping over them but that didn't work out so well. I'd really like to get past level 2 but I'm going to give up. I used to do pretty well at this game back in the olden days. 42,300

37,700 Those bonuses in the first level make a big difference. I hate the way the guy gets stuck when you duck. Very irritating.

You could sell them in bulk lots for other resellers to chew on. Homebrewers may have interest in buying lots of 7800 Ballblazer, just because those all contain a POKEY chip. Not that anybody wants to desolder large quantities of those chips by hand.

I would agree if you said early 90's. I don't have a big sample to draw from, so maybe it's just me. But I think the 1989 Genesis, though awesome, is crap for build quality. The ones from 1990 onward are good though, I've never known of those to have anything wrong with them. They went through lots of board revisions, and the early boards have lots of rework on them. Mine has had problems come and go ever since it was about 1 1/2 years old. All the hack-job rework on the board tells me it must have had design problems. At first I thought mine was just an unlucky unit that had to be modified to pass QC, but then I saw photos online of that same rework on other people's boards, so it was apparently "standard" for the early revisions. I bought it in mid-1990, and it was already having problems by early 92, so I didn't think much of it quality-wise - my NES was more reliable if you can believe that. A friend bought one a couple months after me, his was always perfect and given the timing I suspect it had a later/better board. Mine was already bought awfully late to turn out to have 89 dates in it. Most recently I fixed a video problem, but it's having a recurrence of controller glitches (which first happened in 1992) and apparent Z80 instability. I may have narrowed down the problem though so I'm hopeful. I recognize that maybe I'm the only one to have problems with it, but all the board rework tells me there were genuine issues with the early Genesis.

Horizontal can be either 160 or 320. Both modes are possible. Also, the resolution can change mid-screen. Many games have 320 resolution in a status bar area (for text), then drop to 160 resolution for the gameplay area (160 mode is faster and more colorful). I imagine emulator screenshots would all be rendered as 320 pixels though. Actually, since the 7800 pixels aren't square, the emulators might upscale them to look correct on a computer monitor, so getting 1:1 pixel screenshots might be a problem. The vertical resolution is technically 243 lines (from the "7800 Software Guide"). According to something I found in some code, I think the PAL output is 293 lines. This exceeds what will actually fit on a CRT television, so only part of it is normally used. Exactly how many lines get used would be up to the programming. That's why vertical might be different with almost every game. Unused scanlines would be blank, I don't know if the emulators actually detect this and crop them out of the image, but maybe so. So basically, the theoretical resolution is 320x243 NTSC, 320x293 PAL. In practice, most scanlines in most games will be rendered with double-wide pixels, and only about 192-200 or so (NTSC) scanlines are reliably visible on all televisions. [edit: added PAL info]

Sounds like with the way you had it hooked up, the VCR is what was tuning into the RF signal and providing the audio output to the speakers. Since it was better when you tried using the TV's tuner, the VCR might be the culprit. If you happen to have any other VCRs handy I'd try a different one. There could also just be a volume setting on the VCR that was set too low, but I don't think I've seen a volume control on a VCR before.

You may have already checked this, but make sure the cable you're using is actually good. I found a big difference between cables with my 7800. I'd also try a different TV if possible. My 7800 apparently has pretty bad output. I had a horrible time getting a decent signal to my computer's TV card, then when I hooked up to a regular TV it was much better. So the quality of the tuner in different TVs might also make a difference. Is the background noise consistent with graphics appearing on the screen, or is it just constant? If it seems affected by the screen image then it's probably interference between the video and audio signals, possibly originating in the console itself or it could be the TV's fault. How are you getting the signal to the external speakers? Is the TV outputting line level output - or is it outputting according to the volume setting? If the volume on the TV is too low, and then you're amplifying it at the speaker, that might cause hissing. Do you get good audio from the TV speakers? If so, then that narrows it down to a problem between there and the external speakers.