flip

Hi, The memory diagnostics write to the color ram - on a B/W, this doesn't do anything, but on a color one, it produces a nice test pattern. It only shows towards the end of the test, but running it again has all the pretty colors! Attached is the st2 of my diagnostics - press 2 to run the memory test in a Victory machine and after a while you'll see colors. Background doesn't change, but I guess that could be added... FliP test.st2.zip

sorry, missed the question here... Leaving the multicart in should be fine - in fact, given the age of the connector, it's probably best to pulg and unplug as little as possible. in addition, leaving a cart in protects the connector from dirt and what have getting in. FliP

Can't see any obvious changes in the main logic: all the chips are the same (unlike the change from D->E) - haven't looked at the RF circuit yet... FliP

Here's some photos of the #155 console. The serial number on the outside matches the one written on the board. More importantly, it IS a revision C board as suspected by TLD1985! The warranty sticker (still sealed until this afternoon!) mentions the year 1976, as do some date codes on some of the chips: the CPU suggest week 45 in 1976... The CPU, ROMs and video chip are white ceramics, which looks really quite fragile. The ROMs result in the same checksum and part numbers as the other boards, meaning that the on board firmware wasn't changed or updated along the way, unlike the hardware design. FliP

Interesting... inserting the multicart (or any other cart) disables 2 of the 4 roms in the console. The lower ROM with the interpreter and interrupt routines remains active between $000 and $3FF. It can easily be disabled however by connecting the ROMDIS signal to all four ROMS. The multicart has a jumper, which maps the range between $000-$7FF and $C00-$FFF to the multicart. You could put your own code in one of the slots and run a custom version of CHIP-8 I think, as long as you correctly map to the 512 bytes of RAM (between $800 and $9FF, which is hard coded). At a push, a cart could be developed that has up to 64k available, though you'd still be stuck with a limited amount of RAM: the /MWR signal is not available on the cartridge slot, so you're stuck with read only memory on the cartridge... Some of the bigger games use video page caching to speed things up - that would probably be impossible to do on the RCA Studio II. FliP

While the sounds wasn't going to impress anyone, I doubt they were concerned about the sanity of store clerks. The sound can be switched off via one of the switches on the bottom, in case it would drive someone in the shop nuts... More likely they had to economise space on the cart: the title screen alone takes 1/4 of the available space. And there's 6 text lines, which would be at least another 1/4. Looking at the dump, they had about 320 bytes to code the rest... The rest seems to be graphical data. FliP

Yeah - my bad. I had loaded the victory rom in the Studio II emulator, so that doesn't work properly... Nothing to do with the Demo Cart (duh!) FliP

Hi all, Thanks for the efforts of someone who wishes to remain anonymous and stupus, we now have a dump of the RCA Studio II Demonstration Cartridge! The updated image for the multicart can be found here. The demo cart is in slot 0-F (so the left dial on 0 and the right dial on F) The .st2 file for the emma02 emulator is attached... One thing i've noticed is that the cart doesn't seem to work correctly with the emulated clones (MPT-02, Victory, ...). I need to try this on a real machine, but haven't gotten around to that. Since most people won't be able to re-flash the chip on their multicart, several people have offered to help out with that. ekeefe and KylJoy are both in North America, which should speed things up a bit... If there's more, then feel free to let people know. Rather than sending the entire cart around, people should be able to gently pry the square chip (marked 39sf040) out of the socket and post only that... We could also set up a sort of carousel: if I send a few people new chips, they could send the old one in so they can be updated and sent to others? Enjoy, FliP RCA_demo.st2.zip

Here's pictures of the opened cartridge. FliP

I think we can leave the shell/label as original as possible, but have the PCB different enough to distinguish them either in color (the current set is red which was never used for real carts). We could also add some silkscreen message to indicate a repro on the part sticking out. That would make it harder to cheat while looking the part when it's in the console... also: all the repro ideas are nice, but I guess there's a limited number of shells... Unless we find a new source (i.e. commission a new production run), there's not going to be more than a few dozen? It would be a shame to sacrifice working original carts for repros. FliP

Hi, The first batch of multicarts (about 40, including some complementary ones) is all gone. I need to re-order the PCBs, which I will do in bulk as soon as I've done some testing with an updated design. I post here when they are available again, but it will probably be mid-February... FliP

Quick and dirty attempt: https://goo.gl/forms/yiYXzYqoryasF5A73 To be clear: the most interesting would be the serial number written/stuck on the PCB - meaning the console needs to be opened. While the numbers on the console shell/box/PCB seem to match for early and late production systems, that doesn't seem to be the case for most Studio IIs I've come across... The results are here: https://docs.google.com/spreadsheets/d/1XBAsnVxHJ_Q2bkE02U74nG3c8W2KhculCUdPljC2B0M/edit?usp=sharing If you prefer not to open your console, no worries... Main aim is to get an idea of how many of the different revisions were made... FliP

Good find! Hadn't noticed this before... The earliest board I've seen is #1345, which is a revision D. So that suggests the switch from C to D was made very early on in the production. Comparing the iFixit board, I can't see any obvious electronic differences to the revision D in the main logic (at least not like the difference between D & E). The highest D I've come across is #10,929. This appears to be a replacement PCB for technical interventions (i.e. it didn't come in a case). It's guesswork, but it could mean that 1000 revision Cs were made, then 10000 revision Ds and the rest were Es (up into the 150,000s). Perhaps we need a serial number DB? If there's interest, I can set one in in google sheets/forms? FliP

Just to make sure: the jumper is on the multicart - not inside the console... It has a yellow square on the attached photo. Just move the jumper so the left two pins are connected instead of the two pins on the right... FliP

Sorry - I was away for a few days and traveling all day today. Revision Ds have a low serial number, probably lower than 10,000, but I don't think the board revision should matter on an actual Studio II, though it doesn't hurt to try. Move J-2 to the 1-2 position and see whether this makes any difference... It's very odd that cleaning the slot would offer a temporary solution, let alone that some games seem to work. Every time I've seen such behaviour, i've traced it to one or more RAM chips behaving badly. You could try running the RAM test several times to see whether it trips up... Failing that, It might be a timing issue, whereby the oscillator that generates the clock is no longer tuned correctly. That *could* throw the TPA circuit on the cartridge off - but that a long shot, especially if you're getting a good image on the screen... FliP

The 4 copper rivets need to be drilled out. Don't think there's any other way to take them out. If you take a drill that slightly wider than the rivet hole, it's not too hard and there shouldn't be any damage to the shell itself. Once they're out, take out the old PCB, clean off the stickers and put the new ones on. Then you'll need to drill two holes for the dials and put the shell together again... In the shells that I modified, I re-used two rivets. I glued these in the top holes. For the bottom ones, I used a copper screw thread insert (see photo) that just fits the hole left by the rivet and a matching screw. That way, the cart can be opened without needing holes in the stickers... FliP

Here you go - page order is a bit funny, for printing reasons, so watch the page numbers on the bottom to make sense of it... FliP multi_manual.pdf

Phew - you had me worried there for a while On the Studio II, the cart works with the default setting. If you have an earlier board (Rev D), you can set J2 to position 1-2 (instead of the default 2-3). This bypasses Ed Keefe's TPA circuit that re-creates the full TPA signal on the cart. Without this circuit, the revision E boards seem to mirror the higher address ranges to the lower ones. This is described in Paul Robson technical pages. I can't think of any reason why RCA designers would have done this: it effectively limits the usable address space to $400-$7FF... On the clones, I suspect that the timing is slightly different, so that the TPA circuit struggles with keeping things in sync. I've not looked with a logic analyser, but since bypassing the circuit works, it's not something to worry about. In short: RCA Studio II, revision D: J2 1-2 or 2-3 shouldn't matter RCA Studio II, revision E: J2 must be in position 2-3 (as is on delivery) Clones, including Conic and MPT: J2 must be in position 1-2 FliP

Great to hear you got yours already! For the Conic and the other PAL clones, try to set jumper 2 (the one on the right when looking at the component side) to the other position: short pins 1 & 2. I think the TPA signal is not crippled like it is on the Studio II, so the circuit can be bypassed... The rotary switches: they do have a distinct top that points at what is selected, but it's hard to see... The EEPROM only has 128 slots, so selecting 8-F on the left dial is the same as using 0-F Quite a few games, old and new, require you to press something before they start... FliP

I can guarantee that that won't work. The video circuit of the color/PAL clones is very different from the RCA Studio II. You can get a monochrome output easily enough (even without any composite mod) but to mix in the colors is trickier. I have a basic circuit working, but it's rather tricky to install as the RF board needs to come out to reach a trace on the underside. I think I might use the reference design on the datasheet of the CDP1864 to design an independent composite or S-video signal, but no concrete plans yet... An added complication for those consoles is that there's no onboard speaker: the sound is also mixed into the RF signal... It's not a biggie, but a second connection is needed to connect the sound to a line in/amplifier... FliP

Hi all, Attached is a manual on how to install the composite out mod in your Studio II. Please have a look and let me know if there's any questions... Most carts are on their way, or will be on their way tomorrow. I had to repackage some as there's restrictions on how tick the package can be - Anything above 3cm would cost about 3 times more in shipping: I had to take the PCBs out of the shells to get the the padded envelope under 3 cm The shells have been modified a bit: to open them, the rivets had to be drilled out. The bottom two have been replaced with screws. On top, 2 half rivets were used to keep the shell together. That way, the labels don't need to be damaged if ever you need to re-open the shell... When opening them, make sure you pull it open 'evenly' so you don't bend the rivets. If it's stuck, try prying it open with a knife or other flat tool, making sure you open it evenly on all side... Let me know when you receive your cart! FliP videoMod_manual.pdf

Hey Bill, I can't send a private message to you for some reason... Can you look into that or try to send me a PM? Cheers, FliP

I am out of shells, but I can get you a cart, manual, stickers and a composite board... FliP

Hiya, No bother - things were a bit slower than anticipated: I wasn't planning to make labels & manuals, but I got lured into doing just that... Manuals arrived on wednesday and I got the front labels today. Stupus did a great job designing these, so all praise to him! The only slight concern is that the colors are a bit off: I used a different printing service, and they couldn't reproduce the 'radioactive' green that Stupus had chosen... So the greens are a bit different, but I'll make sure that the front and the back label are at least the same.... We also have the final ROM, so the carts themselves are ready. I still need to build around 10 composite boards, which oddly enough take nearly as long as the entire cartridge... I'll probably start sending out PMs this weekend, so watch you're inbox... FliP

Stupus has kindly offered to take on the artwork... He should have something to show soon (though I need to send him some measurements).