Jump to content


  • Content Count

  • Joined

  • Last visited

Community Reputation

103 Excellent

About the_crayon_king

  • Rank

Recent Profile Visitors

6,470 profile views
  1. Of what I tried of restoring narrow equalization pulses or inverted hsync pulses it would make it work on one video scaler but stop working on another. (because it wasn't 100% correct). However, all scalers seem to work with just having no sync data at times: You could get 100% correct sync. I just don't think it will be worth the effort. Vsync is turned on by "11001" in my files and I did not retime it.
  2. @soufian the German site mentions adjusting a potentiometer on the motherboard., which may be the step you are missing
  3. So I got the Grundig Superplay 4000 and apparently it already puts out RGB (Pictured Left). It also doesn't use the O/S for anything other than inverting the colors. (goes through XOR with RGB bits). Simply put the Grundig (and probably others) are limited to 8 colors even though it has bits enough for 16 colors. I put it through a the Baked Potato kit I have and you can see the difference. Not that I am settled on colors but having the road be grey instead of white is cool. Obviously, since it is running through a kit you can toggle the 8 or 16 color pallet at any time. The picture looks bad because so far I cannot get this to run on anything other than the XRGB.
  4. So minor update, I have the model 2, 8 pin / 9 pin breakouts. I don't know if this is the one I shared here, so I uploaded yet another one. This one is reversible, so one side is 8 pin mini din, and the other is 9 pin mini din. Pictures: Notice that the baked potato board sits on top of this. So this both gives you a place to mount a RGB output and the kit itself. Steps: Solder the connector you plan to use (make sure correct side). Solder the header pins for the board (if you plan to mount it here) to the RF replacement board (do not solder the kit down yet that is last). Now add the header pins in-between the motherboard and the RF replacement (stick the long side out the bottom of the MB). Add solder to the top side of the header pins (the RF replacement side). Bend the header pins on the bottom side (the bottom of the mother board) toward a ground plane. Adjust the centering so that the din connector is centered (I didnt do this step). Solder the pins down. Now add the kit to the top most header pins and solder it down., (take note of the 2 ground pads being used for additional support). The rest of the connections are like normal. I cannot tell if the din connector being slightly to the right is an error in PCB design or something else. but manually tweaking it when putting this together will fix it. Just make sure the bottom of the din connector isn't shorting to the ground plane (this shouldn't happen with normal sized 2.54mm header pins). You may have to shave the hole to fit your cable. I didn't have any issues even with the slightly off-center placement. I have the bare PCB I can add into orders going forward. On request. You would just have to find either the MD-80SM, or MD-90SM and the header pins (basic 2.54mm header pins). I could sell the whole thing as a kit but IDK how popular that would be. Gerber_PCB_RF replacement 8 PIN.zip
  5. Link to original thread: https://atariage.com/forums/topic/266026-magnavox-odyssey-2-rgb-mod-anyone-care-to-try-this/ I didn't want to quadruple post on a necroed thread. Uh, so I have been selling these for a while on Ebay here: https://www.ebay.com/itm/284314005124 Long story short I broke the code for those over the last few months. Emails have been sent but this is to let everyone know that the fault was mine. Luckily, these did not sell in great numbers. Let me know if this affected you so you can get a replacement. Beyond the RGB/Component board (which I have now fixed) I also now have the "Orange Peel" boards working with the O2. For those that do not know the OP boards put out Composite, S-Video, and RGB. Located here: https://www.ebay.com/itm/284777471645 I do not know how composite from this kit stacks up to normal composite kits. I have 3 pallets ATM Tea1002, Rene's (I think), and a third to be determined pallet. If anyone has a source for O2 pallets let me know. TLDR; I guess the point of this post was to let people know about the prior issue and the new O2 mod.
  6. I hate to double post but just to put this out there I have the code done now. EXT not jumped: Pallet 1 (COLSEL not jumped) TEA1001 - OBJ/SCR logic 1 inverts colors. Pallet 2 (COLSEL jumped) TEA1001 - OBJ/SCR logic 1 makes colors darker (like the Aquarius). EXT jumped: Pallet 3 (COLSEL not jumped) BruceAbbots pallet for Aquarius. Pallet 4 (COLSEL jumped) My personal colors I called min/mid/max saturation I could replace my personal pallet with a "non-invert" which would just ignore the state of the OBJ/SCR pin. I = OBJ/SCR R = C1 Red G = C2 Green B = C3 Blue Truth Table: IRGB : Color 0000 : Black? 0001 : Blue 0010 : Green 0011 : Cyan 0100 : Red 0101 : Magenta 0110 : Yellow 0111 : White 1000 : White/Grey 1001 : Dark Yellow 1010 : Dark Magenta 1011 : Dark Red 1100 : Dark Cyan 1101 : Dark Green 1110 : Dark Blue 1111 : Black? One of the "Black?" could be a dark grey if needed.
  7. I ordered a console; so assuming it works when it gets here I should have a RGB mod ready soon after. (this kit would be the same kits I have for Intellivison just flashed differently) I have found a pallet to use here. I found this text (in reference to the OBJ/SCR pin of the 2636): "A further complication is that some consoles such as the Interton VC4000 use these signals to alter the brightness of the colour, while others use it to invert the colour." However, the data I have seen shows it always being used to invert colors not add additional colors. If someone could clarify that point or otherwise find instance of it being used to add colors let me know. Regardless, assuming OBJ/SCR is used at all (unlike on the Arcadia 2001?) then I can have the "COLSEL" pin act to switch the function of OBJ/SCR on the colors. As in have it be toggled to switch between inverting RGB or to adding additional colors. NOTES: The color adding (instead of inverting) using the TEA1002 data would make the same color pallet as the Mattel Aquarius. This would also mean the console would have 16 possible colors instead of 8. I am aware that RGB could be created much simpler., and I still may do that and provide a PCB design for that. However, I do not plan to produce a simpler kit because the demand probably wouldn't be sufficient. If someone is interested enough I may offer these up in advance of getting my personal console to test on. TLDR; I will make a RGB kit for this console (and it's variations). If you have any input for that let me know.
  8. Rolling means you aren't getting good sync. Paired with the dark/black screen it would mean your whole signal is too low., making sync too low to be detected. If I could get a hold of one of these I could make that thing I was talking about earlier in like an afternoon.
  9. I think both Orange Peel and Baked Potato are pretty much done save for identifying/solving a problem. I still don't have a great theory as to what the issue could have been. Wonkey encoder (bh7236AF). Bad 220uf output capacitor. Clock issue. Grounding issue. I always have a signal path to ground when installing. So the ground feeding the kit and then the kits ground feeding the outputs ground. Having ground fed to the kit and then another ground (from the mobo) feeding the output connections (not through the kit) creates noise [ground bounce?]. Now if you have ground feeding the kit and then going out to a common point where you have your outputs then it will probably not cause an issue (but I have not tested that). In testing this had to potential to cause jailbars. Anyway I am scratching my head over the issue. Hopefully, it was a fluke.
  10. @walldog1 @-^CrossBow^- I would think composite rolling is specific to the TV used. A CRT/BVM/PVM should not have this issue but newer TVs could. I think it would come down to minute timing differences from what the Intellivision puts out Vs what the TV expects. Else it would come down to sync again. There were additional corrections that COULD be made (narrow equalization pulses and inverted hsync pulses inside of vsync) but there is just not enough room on a CPLD. Walldog asked me about this earlier and I think the above two scenarios makes sense to me. So that would be down to needing a FPGA and that is not something I think I can make by hand ATM. I would expect CRTs to be able to handle the signals here. If you are using a newer TV then you should probably use RGB/Component through a video scaler (XRGB, OSSC, or Retrotink 5X [there are others but those 3 I recommend]). Even offbrand RGB SCART scalers should give better video on a newer TV than using Composite directly on those same TVs.
  11. This is what I have been working on for the past day or two. Uses Spartan 3A and the ADV7125 video DAC. This is not done yet. It is my hope that people who understand the code of these consoles (but not board design) will use this to create new video solutions. This should be able to work on almost any console to make RGB or component (when it is provided the code). Currently, I am just probing for interest or input. I do not think this will be something I can sell in significant numbers. I will probably make a small batch and sell at cost to people who want to develop off of this. I will also probably make a smaller installable version in case I or others actually develop some code for this. This is currently 3.5x3.5 inches. 31 bidirectional I/Os (5v to 3.3V logic translation). Output could be programed for RGB or YPBPR. 4 RCAs for sync YPbPr. One MD-80SM (8 pin mini din) for RGB. 4 jumpers to set various settings.
  12. It depends on what you are using for sync. If you are using Y (from S-Video) or Composite for sync then if you send that to a TV as well that is going to cut your video P-P in half. There are enough options for sync that you should be able to configure about anything to meet your needs (in regards to Orange Peel that is). 1. Orange Peel is cheaper for no other reason than it is easier for me to make (by a lot). It actually costs a little bit more to make Orange Peel over Baked Potato. Hand soldering LQFP is no joke. 2. The size was one of a few trade offs from one to the other. 3. I am not really sure why you would be having audio issues on the model 1. The bypass solution I made for the model 2 might work for you. I haven't implemented any sort of audio circuitry in the RGB boards because it is better to keep that noise separate. Anyway, I'm glad you like the board -------------------------------------------------------- I added a 2.54mm header version of that breakout posted before. More practical. RF REPLACE 2.54MM HEADER.zip
  13. I haven't confirmed it but I would think there is internal resistors on the audio pins (inside the audio IC). Which would be why the audio lowers when adding a termination to the three mixed audio lines. Other than that IDK anything about audio. --------------------------------------------- I have attached several different RF replacement board gerber files. I found the dimensions in the astec modulator datasheet (I was off 0.3mm and 0.25mm) from my measurements. Sometime in the future I will print these (as a PCB) to confirm they work. From printing on paper it seems to fit. I assume the pinouts and everything are correct. These can work with any kit and any model Intellivision (AFAIK). Has S-Video, 9 pin, 8 pin, and RCA replacements. If it has "ALT" in the name that means it has 3 points of connection but is smaller. The default has four and is larger. These will make the connectors 'upside down' on the model 1. The only thing I am missing is the spacing of the spacer (that would sit below this PCB) and the metal pins that would be soldered to mount this. You could make do with random things you have laying around but I will try to find something more specific. If anyone has ideas for spacers/header pins that will work here let me know. The holes should be 4.5X1.5mm so the header pins would have to be smaller than that. RF REPLACEMENT.zip
  14. @-^CrossBow^- I don't think there is anyway around the model 1 being upside down unless you use the existing 8 pin din connector and drill a hole or solder the md-80/90SM directly to the ground plane. The size of the PCB should perfectly match the current RF mod's holes (or that is the goal anyway). I just made smaller ones because it may have some benefit. I think primarily this would be for people who rather not cut a hole; and I think many people would come around to the idea if it were easier to do. I took many measurements with a caliper but I could use some other data points. If anyone has an astec RF modulator datasheet that gives the distance center to center of the pins that would be a great help. This is what I have so far: I will have to tweak the PCB and print out the design (on paper) and maybe error check the design that way.
  • Create New...