As some of you might know, I'm working on the next iteration of my retro gaming AV Switch Box. It's an extension on the design of my previous switch box, which I've documented on my Youtube channel: That design used a pin on the MCU for every input, placing a hard limit on the potential number of selectable inputs. It was also limited to turning an entire bus switch on or off at a time. Furthermore, due to the monolithic design, it was not expandable in any way, and changing any of the information for the LCD display means re-compiling the firmware and re-flasing with a USB AVR programmer. Despite this, it works well, and suits the number of consoles I have at the moment. I certainly don't plan on getting any more, but I can't rule it out either. With that in mind, the idea for my next switch box has started taking shape: ATMega32u4 Microcontroller - Plug the thing into your computer and use a terminal to make changes without re-compiling the firmware. Single USB (B? Mini? Micro?) port for programming and connecting to power Modular Stack up to 16 input modules, each of which could be single SCART RGB, single DE15 RGB, single Y/Pb/Pr Component, twin composite, or twin S-video inputs, all with stereo sound. Choose SCART, BNC, or DE15 conector module for RGB output Modules contain a small EEPROM containing identification and label data, so if modules are swapped around, they retain their label. Onboard selectable LM1881 sync separator 20x4 character LCD for showing console names and stuff as an optional addon Going forward, I've been toying with the idea of doing a crowd-funded production run. My question to you, Assemblerians, is what do YOU want to see in a retro gaming AV switcher? Some things I've heard so far: 31kHz video switching - Physically and electrically it's doable, 8 switched lines is enough, but I don't know if my PCB-fu is up to the task of carrying those signals with good integrity. I'll look in to it, but I imagine that in most use cases, folk have a different monitor for their 31kHz stuff. HDMI switching - Doubtful. HDMI is high frequency, high bandwidth, and I KNOW my PCB-fu isn't up to that. Possibly this is outside the scope of a retro gaming switch anyway. And some things I'm not sure how I'd approach: Case. Perhaps this could be left to people's imaginations? Maybe provide some designs for a 3D printable case? Suggestions? I don't even know how it's going to be laid out quite yet. Assembly. I'm trying to only use components that could be hand-soldered, so the potential exists to provide this as a kit. Yea? Nay? Having it assembled in China could get quite pricey for something that's probably going to be a short run. So I'm opening up the discussion. Thoughts? Ideas? Am I stupid? Keep in mind that I'm not an electronics engineer, I'm just putting some parts together based on their datasheet descriptions.
This is a really interesting project. I would probably be interested in buying one, if the price isn't too much. I have some suggestions: I would personally rather have the device as a complete product than a kit. I don't have the skills necessary to assemble such a system. I wouldn't mind paying a little extra for a pre-assembled device. A bit of a long shot, but would it be possible to add separate audio inputs? In my current setup, most of my consoles use SCART for audio, but some systems, like my Genesis and my MVS, use separate leads for audio. So perhaps you could have RCA inputs (for audio) next to the SCART inputs, and you could either have a device that would automatically switch to the RCA inputs if a signal is detected, or you could manually choose the input in the software. I dunno, just a suggestion. Does it really need that many inputs? I like that it has a lot of inputs, but having that many surely adds some bulk to the device. Just from my own experience I would think 5 or 6 would be sufficient, and it would bring it down to a more reasonable size. As for a case, I think a simple black box like the one you used would be fine, but the color should be more even - you could see some of the brushstrokes on your prototype. I understand that it's a work in progress though, so I'm not too worried about that. Those are just my ideas, do with them whatever you want.
I think the case/enclosure should be the same width of a standard AV equipment so it can stack nicely on top of an AV receiver. It doesn't have to be deep, just deep enough to house the PCB. If you are doing a short production run, consider formed sheet metal for enclosure. 3D printed case is not economical for this purpose. If you need sheet metal design I may be able to help. Great video by the way.
Great feedback! "I would personally rather have the device as a complete product than a kit. I don't have the skills necessary to assemble such a system. I wouldn't mind paying a little extra for a pre-assembled device." -- Yeah, I'm thinking the kit form would would be an option if you're comfortable with it and don't mind some assembly to save a few bucks. "A bit of a long shot, but would it be possible to add separate audio inputs?" - Certainly doable. Nothing's set in stone yet. "Does it really need that many inputs?" -- It can be as few or as many as you like, hence the modular approach. For me, my current custom box has 12, and I've run out. I started out designing this for myself, the idea of making it available to others came up much later! "As for a case, I think a simple black box like the one you used would be fine, but the color should be more even - you could see some of the brushstrokes on your prototype." -- The one in the video isn't even a prototype, that's just the one I made for my personal use, hence the recycled computer case and hand-painted-ness, I'm basically the only one that ever sees it. "Those are just my ideas, do with them whatever you want." -- The feedback is good! Obviously the more people that buy one the better, so broader appeal is a good thing. "If you are doing a short production run, consider formed sheet metal for enclosure. 3D printed case is not economical for this purpose. If you need sheet metal design I may be able to help. Great video by the way." -- Interesting, when I get to the point of considering casing, I might get back to you. Also, thanks!
Right, but if you start producing it, even if it's a short run, you're going to need to decide on a number of inputs - correct? Each one isn't going to be custom made, is it? And the size of the case (if you end up including one) also depends on the number of inputs. Or am I misunderstanding how the thing works?
The way I imagined it, first there's an enclosure that just has the microcontroller and some power supply related stuff, and empty spaces where the modules snap in. The modules are a separate item, of which you can buy as many or as few as you want. It means the unit can be expanded later if you want. 5 might be enough now, you might want to go to 8 after picking up some more systems. That was the intention anyway. Now that I'm saying it out loud, that's probably a dumb way to do it, commercially speaking. It'd mean I'd need to keep stock of the modules, there'd be ongoing work, blah, yeah, good point.
This is a simplified rendition. The front and back can be laser/milled/die-cut aluminum blanks. Top cover and bottom tray can be formed steel sheets.
I actually think the modular design is a good idea, especially as people could buy the configuration they want/need. The only issue would be with the case; if you need to change it to add modules later. Unless the modules are thin enough that the standard case can accommodate a large amount of inputs, like 15. In any case, the modular design is a good idea economically for you, as the pcb prices will probably go down drastically for 10 modules versus 1 big pcb.
im with la-li-lu-le-lo on the assembled option what would be awesome is a hdmi output , my tv has 4 but only one av set
I'd love to buy one of these too. I really need a quality 21pin RGB switch box (Japanese standard of course). That could output via HDMI or D-Terminal
Given I loved your original project, this certainly sounds interesting. I've put my monster switchbox on the backburner until I have the time to build it and enough money to grab the components. For reference, this is how mine's going to shape up in terms of inputs: - 10x composite video with stereo audio for each input - 10x s-video with stereo audio for each input - 10x component video with stereo audio and SPDIF for each input - 10x RGB SCART (signals L-audio, R-audio, R video, G video, B video, Sync, Blank, Switch) with separate stereo audio for each input - Integrated RGB to Component video transcoder on the SCART output Maybe that could provide an idea of possible configurations for your project? Mine's probably a little extreme, but I thought it might come in handy! For me, the circuit's virtually done thanks to help from yourself, Calpis and others who chimed in on that old thread. At the moment my design isn't using microcontrollers, but it sounds like it's a great way to go forward (I'm not confident in my programming skills, but since I'm in no rush I might go down that path). Good luck!
very interested in something like this, being that I have been rebuilding my game room I can not find a switcher that will provide all my needs.
Are you still working on this? The other one that superg is working on looks cool too, but it's fully automatic which is a turn-off for me. I like having the ability to manually switch between inputs, which gives yours an advantage in my mind. I also like the idea of having an LCD to display names for the inputs. Anyway, just wanted to know what the status of the project was. By the way, what would it use for power? Would it be 230V or 110V, or would you be able to switch between the two?
I'm still working on it in small bits and pieces, I don't have a lot of free time for it. I'll probably get stuck in to it over the Easter long weekend, actually. WRT casing, I like the pressed metal idea, hopefully I can make it work with the layout I have in mind. To account for the stupid angle of SCART plugs, I was thinking of having the input panel lying flat, so the cables can enter straight in from the back. It would mean the unit isn't stackable of course, so it's something I'm still thinking about.
That's an interesting idea. I think a lot of commercial SCART switches are flat like you're describing. I kind of prefer the idea of having them in the back of the unit, like the design that BuffaloWing showed. I think that would make it look more like a piece of AV equipment (which it is), like a amplifier or a CD player. If you have the inputs on the top of the unit, then you would probably need to hide it somewhere - it wouldn't look good just leaving the wires hanging out on a shelf. So the advantage of having the inputs in the back is, like you said, that it's easily stackable, but also that it would look good on a shelf next to other AV equipment. Just something to think about. One possible solution to the issue of the SCART cable angles: you could alternate the directions of the SCART cables for each row of connectors. That is, have a row of SCART connectors all sitting in the same direction, then to the side of those you would put SCART connectors facing the opposite direction, and then alternate back and forth. You'd need to have gaps in between the places where the connectors are facing towards each other. But in that situation you'd only need gaps between every 2 rows of connectors, whereas if they were all facing the same direction, you'd need gaps for every row. Maybe I'm getting rows and columns mixed up: what I mean is the vertical set of connectors. Something else I just thought of: it might be difficult to implement, but it would be cool if it had an IR receiver that you could use with a remote. I don't know if that's something you'd be interested in, but just an idea.
