RGB problems with consolized NeoGeo MV1c

Discussion in 'Arcade and Supergun' started by erkan, Sep 7, 2013.

  1. synrgy87

    synrgy87 Well Known Member

    Joined:
    Jul 12, 2012
    Messages:
    1,769
    Likes Received:
    20
    get 5v from whatever you are using to power the MVS.

    do you want to use RGB or RCA / S-Video ? or all 3? getting confusing now lol.
     
  2. Bad_Ad84

    Bad_Ad84 The Tick

    Joined:
    May 26, 2011
    Messages:
    8,619
    Likes Received:
    1,387
  3. Calpis

    Calpis Champion of the Forum

    Joined:
    Mar 13, 2004
    Messages:
    5,906
    Likes Received:
    21
    Even using a premade video amplifier you'll unfortunately need an attenuation circuit (and level shift circuit if the amplifier isn't "rail-to-rail" or doesn't have a level shift internally, or isn't AC coupled). After attenuation, you'll then need to make sure the amplifier has high impedance inputs; many/most video amplifiers and encoders will probably have 75 ohm terminated inputs, so it'll have to be modified to get rid of those.

    You want:

    high impedance Neo Geo DAC output -> attenuation and level shift (2-3 resistors/color) -> very high input impedance amplifier -> 75 ohm series terminated output

    The Neo's DAC output is 3 Vpp, this needs to be attenuated to 1.43 Vpp if you use an amp with a gain of 1 (better noise figures), or 0.715 with a gain of 2 (most video amps).

    The Neo AES uses voltage divider of 6.8k series, 2.2k shunt to GND = 0.733 Vpp. This would be suitable to send directly into one of TI's THS amps with internal level shift, just make sure there aren't any input biasing components.

    If you use an old RGB encoder (Jrok I think), it'll probably have internal biasing components and specify a small AC coupling capacitor (like 0.1 uF) on the input which effectively does the level shift.
     
    Bearking likes this.
  4. kuze

    kuze Peppy Member

    Joined:
    Sep 28, 2011
    Messages:
    371
    Likes Received:
    6
    Neobitz is another alternative video encoder besides jrok
     
  5. retro

    retro Resigned from mod duty 15 March 2018

    Joined:
    Mar 13, 2004
    Messages:
    10,354
    Likes Received:
    823
    It's well documented, it's simplicity itself and you're somewhat misunderstanding it. Here....

    http://en.wikipedia.org/wiki/SCART

    http://www.hardwarebook.info/SCART

    Pin 16 requires between 1-3 volts to switch into RGB mode. The television is waiting for a voltage to be applied to that pin. How's he going to give it 1-3 volts any other way than using the +5V line with a resistor? Likewise, pin 8 is sometimes used as a switch.

    If you crack open any official manufacturer's RGB SCART lead, it'll wire the console's 5 volt supply to those pins via a resistor.

    Does the JROK (and Neobitz) output attenuated RGB, then?
     
  6. erkan

    erkan Spirited Member

    Joined:
    Oct 2, 2010
    Messages:
    143
    Likes Received:
    1
    I would like to have RGB, but if that means it will only work on specific TVs then f*ck it, I would rather have RCA and have it working on 100% TVs :)

    Thanks, I will look up jrok/Neobitz. The journey continues :)
     
    Last edited: Sep 9, 2013
  7. retro

    retro Resigned from mod duty 15 March 2018

    Joined:
    Mar 13, 2004
    Messages:
    10,354
    Likes Received:
    823
    You could always run both.. RGB when it plays ball and S-Video when it doesn't. JROK do component boards, too.

    I've not used a JROK, but it might be possible that it would output RGB after attenuation. Going straight from MVS to SCART via RGB has worked for me, though. Either way, it could be worth going down the JROK route.
     
  8. Calpis

    Calpis Champion of the Forum

    Joined:
    Mar 13, 2004
    Messages:
    5,906
    Likes Received:
    21
    I'm not misunderstanding what has been asserted, I had a hard time believing it because it's asinine to impose a 12V rail on equipment for mode switching. And I disagree that the standard is well documented, I don't see a formal specification anywhere accessible.

    A better system as I said would have the TV pull pin 8 up to 12V rail, and the equipment could create a voltage divider using resistors to manually switch, or add a transistor to auto-switch. It's kind of dumb to use a ternary system at all.

    That wouldn't necessarily make it correct. Consoles take liberties, sometimes a lot, so how can they be trusted to consult CENELEC? Some consoles such as the SMS and SNES IIRC output high impedance RGB which is unsuitable for a TV load or even driving a coax cable.

    What? I don't see potentiometers so it's unlikely either performs the attenuation themselves. With a 0.7 Vpp input they'll output 1.4 Vpp. You have to pre-attenuate so the output doesn't clip with the gain.
     
  9. erkan

    erkan Spirited Member

    Joined:
    Oct 2, 2010
    Messages:
    143
    Likes Received:
    1
    Neobitz-S ordered (prebuilt). Let you know how it went when I got it.


    Great idea! :)
     
    Last edited: Sep 9, 2013
  10. synrgy87

    synrgy87 Well Known Member

    Joined:
    Jul 12, 2012
    Messages:
    1,769
    Likes Received:
    20
    heres a Painted diagram probably not required but I was bored

    mvsscart.jpeg
     
  11. TriMesh

    TriMesh Site Supporter 2013-2017

    Joined:
    Jul 3, 2008
    Messages:
    2,373
    Likes Received:
    785
    The original Peritel spec didn't use ternary signaling - it just defined two states, one with voltage applied to the mode pin (nominally 12V) and the other with no voltage applied. What happened was that when widescreen TVs started coming out they wanted to have some way of communicating the screen format to the TV, so someone came up with the idea of using the existing mode pin (since it could be expected to be wired in the cables) and used a ternary format for this - with 4:3 being 12V so as not to break backwards compatibility. I have no idea why they chose 12V in the first place.

    Strictly speaking, 5V is not correct unless you want 16:9 mode, although a lot of old 4:3 TVs will happily accept it. About the only people that ever seemed to care about this were Sega - the extra pin on the PAL Saturn PSU is there to supply the voltage for the mode switch pin (which is also why PAL Saturn's don't have a C-Sync output).

    Basically, it's one of those legacy standards that was originally designed to meet a specific need (connecting screenless Minitel terminals to a TV as a display) but has managed to hang on far past the point it should have died.
     
  12. retro

    retro Resigned from mod duty 15 March 2018

    Joined:
    Mar 13, 2004
    Messages:
    10,354
    Likes Received:
    823
    Both the links basically quote the standard, which is a tiny document you have to pay a ridiculous amount for. If I can find my copy, I'll PM you.

    Yeah, well they pretty much went for a logic switch. As for pin 8, I'm pretty sure at least the widescreen implementation came as a later addition although I don't remember what it was like before, perhaps again a low voltage switch. I don't know that it's any better a solution - don't forget it could be any device connected to any device, not necessarily a television, so all devices would still have to have a 12 volt rail regardless.
     
  13. Calpis

    Calpis Champion of the Forum

    Joined:
    Mar 13, 2004
    Messages:
    5,906
    Likes Received:
    21
    The 75 ohm resistors in your scheme make no sense. You are attenuating the signal with them, which is a good thing, but the use of 75 ohms alludes to a transmission line that isn't there. If you're going to do it this cheap way, ignoring all rules, a better value is a 120 ohm R which will attenuate the RGB signals almost exactly to the correct 0.714 Vpp level.

    Technically the Neo Geo's DAC has variable impedance depending on the pixel state, this is why an amplifier is really necessary for proper RGB outputs--an amplifier can transform the DAC's impedance to a constant 75 ohms.

    But they had the data channel for negotiation by then, right? Surely it would make more sense to use that with new widescreen equipment.
    The pinout site information appears to be reverse engineered, and the Wikipedia gathered from similarly unofficial sources. Some things can be taken for granted, such as that the RGB signals are nominally 100 IRE units, but even that is not specified anywhere.

    Things other than televisions had SCART inputs?

    The specification seems backwards because it is insignificant to provide a 12V rail in the TV which can always use it for amplification, or under a pull-up scheme just use another voltage rail without consequence. A 12V rail in a computer that would otherwise not need it is a big burden. Consoles probably should have used their unregulated DC input for switching.
     
    Last edited: Sep 9, 2013
  14. synrgy87

    synrgy87 Well Known Member

    Joined:
    Jul 12, 2012
    Messages:
    1,769
    Likes Received:
    20
    televisions, VCRs cable boxes dvd players, and other av equipment had SCART it was very widely used more so than S-Video

    the 75ohm resistors are probably not necessary on all devices but i have them on my cable as they resulted in the best compatability and picture across multiple TVs and monitors that i have also with the GBS 8220, i'm not overly concerned with what is the correct way to do things, more concerned with what actually works :) higher value resistors on the RGB lines = dimmer picture and on the comp sync = losing sync rolling picture
     
  15. Calpis

    Calpis Champion of the Forum

    Joined:
    Mar 13, 2004
    Messages:
    5,906
    Likes Received:
    21
    Said devices have SCART outputs, not inputs apart from maybe VCRs, though I suspect they only used composite video.

    It's a mathematical fact that 120 ohms will get you closer to the specified input level, which means that it will still "actually work", but better. I do hope you never sell such modifications.
     
  16. synrgy87

    synrgy87 Well Known Member

    Joined:
    Jul 12, 2012
    Messages:
    1,769
    Likes Received:
    20
    i don't sell and it's not a modification it's a custom cable that works for me :) as i said i got the best results from that setup i also tried with variable resistors. "mathimatical facts" and real world performance are not always the same.

    VCRs had RGB scart input as did dvd recorders and DVRs, some even had a physical switch on the back to switch between composite and RGB and anything that also worked as a passthrough
     
  17. retro

    retro Resigned from mod duty 15 March 2018

    Joined:
    Mar 13, 2004
    Messages:
    10,354
    Likes Received:
    823
    VCRs most definitely used inputs, and RGB. Monitors have SCART inputs, too. And DVD recorders. The latter most certainly have models that use RGB on the SCART input. It's irrelevant when it comes to switching, though.

    Oh, and there are auto switching SCART blocks:

    http://www.tlc-direct.co.uk/Products/MX520R.html

    The text used for specifications on those pages is very similar to the text in the standard. Of course, the copyright is very specific in prohibiting reproduction of excerpts of the standard, so those with access probably simplified / reworded it a bit to err on the side of caution.
     
  18. TriMesh

    TriMesh Site Supporter 2013-2017

    Joined:
    Jul 3, 2008
    Messages:
    2,373
    Likes Received:
    785
    I think the first widescreen TVs started coming out quite a while before the format on that data link was standardized, and at that point most of the cables simply didn't connect up the data lines because doing so was more likely to result in strange behavior than anything useful - but they all had pin 8 wired, because there were a number of TVs that you simply couldn't put into SCART input mode any other way.

    Sure, a lot of the decisions made might seem strange from a pure engineering standpoint, but when examined in the context of trying to minimize the amount of hassle you get from the millions of people who had legacy products, it makes a lot of sense.
     
  19. erkan

    erkan Spirited Member

    Joined:
    Oct 2, 2010
    Messages:
    143
    Likes Received:
    1
    Finally got the Neobitz and I still have the exactly same problems with it mounted.

    I have connected the Neobitz to MV1-C jamma as the following:

    Ground - Pin 2
    5v - Pin 4
    Red - Pin 12
    Blue - Pin 13
    Green - N
    Sync - P

    (I also have power to pin 3, 5v, and pin 1, ground, removed all other cables on jamma edge slot)



    I have only connected the RCA composite from the Neobitz, I connect:

    Yellow wire from Neobitz to on RCA center
    Black wire (ground) from Neobitz on RCA shield

    Any ideas? Is the MV1-C toasted?
     
  20. kuze

    kuze Peppy Member

    Joined:
    Sep 28, 2011
    Messages:
    371
    Likes Received:
    6
    You might want to try cleaning the cartridge slot. Some debris had gotten in the second slot of my two slot and caused strange artifacts like your screenshot. Once I cleaned it out, it started working fine like my primary slot.
     

Share This Page