Extend any Game Controller 150ft - Kickoff Thread - Feedback

[ShmupJunkie]

Hi. My name is Ilya (aka Movietimepro) and I've been a long time reader/lurker here and only recent poster. I'm an AV Professional and an amateur gaming hobbyist as all are reading this forum. I've been working diligently since last year on a pretty unique gaming setup in my own home and have reached some significant milestones that I wanted to start sharing with the community.

To be clear, this is not a new piece of hardware or product. That being said, I'm not aware of this having been done before, or on this scale, which is why I had to spend a lot of time finding the right combinations. Instead of rehashing everything I've already written, at the end of this, I will link to the couple blogs I already put together on this.

I am not sure of how much interest this is to everyone, as I don't know how many people have the desire to extend their game controllers long distance. Or in my more extreme case, centralize all of my consoles in a single room and then distribute them out to all of my displays.

Aside from my Gameroom which has a CRT and a Flatscreen, I also have a Cinema Room, a Living Room downstairs, and also my Bedroom. I went through many (many many) various products and combinations, and the below is the only combination I have found to work to my standards across all systems.

The plan is to put together some videos showing it in action and how it was done for those that are more visual. It took a while just to get to the point of distributing the video, audio and control from so many systems across 5 unique displays.

Challenge:
The What: How can I extend control of my game consoles over a long distance? To other room locations far away, up to 150ft in this case.

The Why: To centralize all of my systems, and distribute them all to multiple rooms and displays that they normally could not reach.

The How: By using a very specific USB extender, along with specific mechanical USB switches and RJ45/Data Switches, I have been able to achieve this goal for over 20 game consoles across multiple displays in my home.


The Extender:
Calrad 72-132 USB Extender. Passive no power required. I have found it works to extend any controller over standard CAT5e/6 cabling up to 150ft away. Used with a more modern USB based console (PS3/4, Xbox360, Switch, etc) no other
parts are needed. Plug and Play. https://www.amazon.com/Calrad-Electroni ... way&sr=8-1

Older Consoles: NES, SNES, AES/MVS, Genesis, Saturn, Gamecube, etc. need a quality converter. Convert from proprietary connector to USB. Each console is different. Brook adapters have all worked perfect so far in testing. Any console where no Brook adapter exists, a basic Tototek converter to Playstation2 format has worked perfectly in between. Example: Tototek SNES to PS2 adapter -> Brook PS2 to PS3/4 adapter

Why Brook: The Brook adapters all work with a wide variety of controllers and arcade sticks. For all modern systems, you can use the same USB controller without any adapter at all.

USB Switching: Mechanical USB switches only. No hubs, amps, etc. Very specifically this one. Dirt cheap and works perfectly. I have daisy chained up to 5 of them, with an extender in between, and a second extender to 150ft in a remote room, and it still worked flawlessly. https://www.amazon.com/Updated-Sharing- ... ay&sr=8-14

Room Switching: Mechanical RJ45 switch. If you have more than one remote room to send to (or multiple arcade cabs), and want to switch between where your control signal is going, this product is needed.
https://www.monoprice.com/product?p_id=1374

To summarize, in my gameroom, I have over 20 consoles connected to a flatscreen and CRT. I now have the ability to route any one of them, to any of 3 additional rooms/displays in the home, along with multiplayer controller support. I have tested it for some time with every console and it just works. It took a lot of trial and error, but the results were excellent and I wanted to share with the community in hopes others have been wanting to do the same. I went through a lot of USB extenders before I found one that works, as well as methods of USB switching. This is a very general overview and if you want specific details they’re contained in the blog.

I am hoping to get suggestions, feedback, and anything else I haven't anticipated yet that could help the project improve, in addition to being a resource to anyone looking to do something similar.

Link to room/setup overview (a bit commercial, it is my company site) : http://www.movietimepro.com/?page_id=66

Link to first technical blog post: http://www.movietimepro.com/?page_id=632

[SavagePencil]

jesus that was a lot of unnecessary words

[thebigcheese]

Interesting idea, but I feel like there are two problems that kinda negate the usefulness:
- What if I don't want to play the game that's currently in the console? You'd need Everdrives/ODEs for every console and even then you can't always return to the menu without physically interacting with the console or cartridge.
- Are you just going to leave every console turned on at all times? Not only bad for the consoles, but also hell on your electricity bill.

I guess my point is, given those two constraints, you're probably going to have to go to your game room to interact with the consoles in some manner every time you want to play, so why not just set up your game room so that it's the place you always want to play?

[ShmupJunkie]

jesus that was a lot of unnecessary words

You’re right. I appreciate the feedback and could have edited down a lot of unnecessary exposition. I went back and cut out anything that seemed unnecessary to the overall concept.

[ldeveraux]

jesus that was a lot of unnecessary words

...and an unnecessary comment. If you don't like it, move on. I found it interesting!

[Syntax]

jesus that was a lot of unnecessary words

...and an unnecessary comment. If you don't like it, move on. I found it interesting!

I tuned out half way through, so I think the comment is fairly constructive.

[ShmupJunkie]

Interesting idea, but I feel like there are two problems that kinda negate the usefulness:
- What if I don't want to play the game that's currently in the console? You'd need Everdrives/ODEs for every console and even then you can't always return to the menu without physically interacting with the console or cartridge.
- Are you just going to leave every console turned on at all times? Not only bad for the consoles, but also hell on your electricity bill.

I guess my point is, given those two constraints, you're probably going to have to go to your game room to interact with the consoles in some manner every time you want to play, so why not just set up your game room so that it's the place you always want to play?

I agree. The setup isn’t for everyone. To play a console you still need to interact with it initially to turn it on and later turn it off. That can actually be automated via IP controlled outlets for most consoles but I doubt most would find it worth the expense. Definitely would never just leave consoles on. And if you are going to swap games without an everdrive you need to be with the console regardless. So the concept wasn’t to never have to interact with consoles. Honestly that’s part of the experience in many cases. The idea was to allow playing them in other rooms if desired without having to unplug everything and move it. I do prefer playing in my game room most of the time, but I also like the ability to play in the theater, in the main living room with friends during a party as it’s a much larger space, and sometimes casually with my wife in the bedroom. I personally do have a flash cart or similar for almost every system for that reason, so I only really need to interact with the console when turning it on or off. It’s honestly been fantastic, but I realize it’s not an application for everyone.

I can imagine an application for a commercial game lounge that wants to securely centralize their consoles, maintain control of them, and simply extend control out to the displays.

For some it may simply be wanting to extend their single controller past it’s existing limitation in a single use case and not any elaborate switching scenario. Some people have a thing for hiding their equipment out of the way for aesthetic reasons and this could be a means to do that if they don’t mind the trade off of going to that location to interact with the console.

Your points are completely valid and I honestly don’t see any work around to interacting with the console when needed. To be fair, my goal was never to eliminate that aspect. Only to find a way to extend controllers long distance if desired. That absolutely requires a trip to the console initially, to change games manually without a flash cart, and when done playing. Only IP outlet control and flash carts would negate that unless I’m missing another alternative.

[ShmupJunkie]

Tested Components per Console confirmed working. Links included. I will update this section and note any significant edit dates. If I list the same product link across multiple consoles, it's because I specifically tested that console with that product.

Note that although Brook converters are branded as PS3/4 converters, they are compatible with a pretty diverse number of other controllers and arcade sticks. Each of their product pages list all other tested controllers.

Phase 2 will be testing arcade boards and converters. We have an MC Chthulhu on the way I am looking forward to testing with as well.

I list them in order from source to final USB connection if used in this way

NES/Analogue NT Mini-
A. Raphnet Famicom to NES adapter: https://www.raphnet-tech.com/products/f ... /index.php
B. Tototek Famicom to PSX/PS2 http://www.tototek.com/store/index.php? ... ucts_id=55
C. Brook PS3/4 to PS2 http://www.brookaccessory.com/detail/44237976/

SNES-
A. Tototek SNES to PS2 http://www.tototek.com/store/index.php? ... ucts_id=53
B. Brook PS3/4 to PS2 http://www.brookaccessory.com/detail/44237976/

PC Engine Duo-R-
A. Brook PC Engine/Mega Drive to PS3/4 Converter http://www.brookaccessory.com/detail/67670908/
Note: Also works using a Tototek Turbografx or PCEngine to PS2 converter, along with a Brook PS3/4 to PS2 adapter. No longer necessary but also tested.

Genesis-
A. Brook PC Engine/Mega Drive to PS3/4 Converter http://www.brookaccessory.com/detail/67670908/
Note: Also works using a Tototek Genesis/Mega Drive to PS2 converter, along with a Brook PS3/4 to PS2 adapter. No longer necessary but also tested.

MVS Single Slot-
A. Brook Neo Geo to PS3/4 converter http://www.brookaccessory.com/detail/95441291/
Note: Also may work with other arcade boards/adapters that use the same MVS style connector, but have not yet tested. Testing various arcade boards is part of the next phase

N64- Pending adapter, not yet tested

Saturn-
A. Tototek Saturn to PS2 adapter http://www.tototek.com/store/index.php? ... ucts_id=52
B. Brook PS3/4 to PS2 http://www.brookaccessory.com/detail/44237976/

Dreamcast-
A. Dreamcast to PS3/4 converter http://www.brookaccessory.com/detail/47613739/

Playstation 2 Fat (used for PS1 and 2 games)-
A. Brook PS3/4 to PS2 http://www.brookaccessory.com/detail/44237976/


Playstation3 - USB direct no adapters needed

Xbox360- USB direct no adapters needed

WiiU- USB direct no adapters needed

Switch- USB direct no adapters needed

PS4- USB direct no adapters needed

XboxOne- Not yet tested


System working but pending more testing-
Gamecube- still evaluating converter
Xbox- tested with multiple converters but not completely satisfied with results yet, still pending other adapters to test
Wii- still evalutating converter, would like to find one to also control motion menu and extend sensor bar
N64- pending adapter/converter
Older consoles: Atari 2600, Colecovision, Intellivision, etc not yet part of setup or tested
Less common consoles: 3DO, Jaguar, etc not yet part of setup or tested

[BuckoA51]

That USB to cat 5 extender has to be adding a fair bit of lag I would have thought?

I thought about something like this briefly a few years ago but abandoned the idea. In the end I simply purchased an extra PSU and video cable set for the systems we wanted to play in both rooms and if we planned a house party or anything it was then easy to disconnect them and reconnect them in the other room.

Typically for parties we would be playing games like Rock Band, Wario Ware, Wii Sports or Beat the Beat so low lag is important there... the other favourite was Bomberman and I doubt you'd want to extend this to cover 8 or more controllers from the Saturn

I think this could have practical uses for places like museums etc though.

[gray117]

I think this could have practical uses for places like museums etc though.

To be fair his house kind of is a bit of a museum (kudos on that) ...

Would be real interested to get the lag measured somehow - not clear on what actually has to happen/cost on usb to ethernet (and vica versa) conversion, I presume the OP has mitigated some of the larger pitfalls involving ethernet switching hub delays etc., but no real clear idea on what that leaves as the 'cost'.

[ShmupJunkie]

That USB to cat 5 extender has to be adding a fair bit of lag I would have thought?

I thought about something like this briefly a few years ago but abandoned the idea. In the end I simply purchased an extra PSU and video cable set for the systems we wanted to play in both rooms and if we planned a house party or anything it was then easy to disconnect them and reconnect them in the other room.

Typically for parties we would be playing games like Rock Band, Wario Ware, Wii Sports or Beat the Beat so low lag is important there... the other favourite was Bomberman and I doubt you'd want to extend this to cover 8 or more controllers from the Saturn

I think this could have practical uses for places like museums etc though.

I've been racking my brain on an objective way to measure the controller input lag and have combed some old threads in hopes of duplicating their measurements using my setup. The delay on Brook and similar adapters was already measured in a previous thread and isn't news, so my goal is to duplicate that testing both with the adapter again and through the USB extender as well. I initially thought the same thing and expected it to be unplayable for many games, and didn't seriously consider the idea, until I tested it for myself and was surprised by the results. I played through a handful of my go-to games that I have mastered and require timing and precision to 1CC. In my case the original NES Ninja Gaiden, Lords of Thunder on it's Hard setting, and Blazing Lasers. BL has 4 ship speed level settings and I always switch between level 3-4 depending on the stage. It's a pretty simple and not very difficult game (at least until the final stage where it ramps up a bit), but what I've always loved about it is the razor sharp instant response on a d-pad when playing at speed 4, almost like it's wired to your brain. I'll purposefully let the screen fill with bullets to see how much I can avoid using the highest speed level. It's just fun. On a CRT it feels sublime and there are some other games that are similar. So when I play it again through any display that adds any more than a frame of delay, I can immediately tell and it throws off my game. I have to adjust to the slight delay in response. Using my Vizio in Game Mode at 14ms of delay plus the 6ms added by the DVDO scaler (20ms total approx I measured it using a Leo Bodnar), I can feel the slight difference and have to adjust. Granted 20ms is very reasonable and playable and most would find it indistinguishable. To your point, once I add a Brook converter with it's own inherent lag, I notice the additional delay.

What I did not expect is to notice no additional delay from that point forward when adding the Calrad USB extender into the chain. Although any USB conversion has to add some form of delay, minor or otherwise, I couldn't discern any difference between using it and not. It could be that the signal format conversion occurring is so simple that the additional delay is measured in single digits. That wouldn't be unprecedented as if you look at some of the Raphnet-Tech adapters, their latency according to his site are in the neighborhood of 3-6ms. In fact, considering that it is a transmission device only and not doing any kind of controller polling (which is what the Brook adapter is likely doing), it's feasible the USB adapter itself is adding 1-2ms of delay max.

I then verified this by bypassing the Brook adapter entirely by using a Nintendo switch with native USB. Ie I can plug the USB extender directly into the Switch, and a Pro controller into the receive end. As expected, what I found is near instant response playing Smash Bros. Again, this is not objective outside of the fact that my son is a competition level player (he normally gets to top 32 in the events, not bad for a 16 year old), and he confirmed that without the Brook adapter and through the USB extender, he couldn't tell any difference whatsoever. That would also make sense if the extender is truly adding only single digit ms in delay as I suspect. It's not objective by any stretch, but if someone at competitive level can't discern a difference, it's certainly promising.

I am big proponent of objective measurement, and I painstakingly measured video lag during this whole process. Unfortunately controller input lag is so much more difficult to do. So please take my above findings with a grain of salt and as subjective. That being said, I do trust my seat of the pants subjective measurements with games that I can 1CC and have played ad nauseam, and I feel I am on the right track. From what I can discern, and I'd like to objectively measure this, that the USB extender is adding an imperceptible amount of lag (low single digits), whereas the real lag point is the adapter/converter like the Brook. I can immediately notice with that in the chain on specific games vs not at all using the USB extender only. If there is anyone in the Los Angeles area with some expertise in measuring controller input lag, I'd be more than happy to host them and see what we find. In the meantime I'll be trying to put together something of my own to measure it objectively.

You're correct about the massive multiplayer games! I have no intention of running 8 player out to every room lol. If that's the case I would just go to the gameroom. That does usually fall into the rare category in most cases. 2 player games are the norm around here and I'd guess in most applications. Thank you by the way for the constructive input. These are the things I've been working to document and measuring the controller input lag objectively has been at the top of my list. That being said, I've been happily playing a lot of my games in the theater and downstairs, and for casual gaming it's been fantastic.

[orange808]

I bet that USB extender is a complete USB device that actively resolves the signal before sending it along to the next hop.

Any repeater that fully processes the signal will probably use standard USB polling: 8ms (1/2 frame delay). After that, add the time it takes for the signal to be resolved, processed, and sent.

It's counterintuitive, but I think it would be easier to grab some "off the shelf" parts and add a complete node (with all the delay and unnecessary complexity that introduces) to the USB chain than it would be to design a board to properly adjust the voltages on the individual wires--for longer runs. Designing the board to properly boost the signal would require fresh original engineering and knowledge--and we're talking about cheap Chinese mass produced gear.

[ShmupJunkie]

I had some free time this morning and decided to modify a Sony RAP4 arcade stick I had, as I wanted some objective measurement on input lag with and without the USB extender. It makes sense that it would add some level of lag, whatever that may be. Without beating around the bush, I'll say that so far I am seeing near zero measurable difference between controller direct and using this USB extender. I posted my quick recordings below so anyone can download and play them back frame by frame. I created 2 sets of recordings. First set is 60fps. Because I was getting no measured difference in that set, I then recorded a second set at 240fps. That should theoretically show delay down to 4ms per frame approx. Still, I am seeing little to no measurable difference and wanted to get some feedback here. Maybe I am missing something, but it seems pretty straight forward.

I wired a small LED strip powered by a 9V battery directly into the X button on the stick. It's a direct contact closure. Pushing the button closes the contact and lights the strip. The strip likely lights faster than even the internal board responds to the button press. So I can record a video and from the moment the LED strip lights to the moment the reaction happens on screen, the video frames can be counted. There may be more technically adept ways at doing this, but it seemed an interesting quick test. If this is a flawed method, I'd love to know the reasoning for my knowledge if anyone wants to pitch in.

As a note, I am not measuring anything but a DIFFERENCE in delay. I realize there is a delay inside the controller itself, yet another delay between the controller and the PS3, and yet another video delay for the display itself (which can be measured and subtracted). The game itself may also have a built-in delay. Measuring actual input lag between controllers can have many variables and that wasn't the purpose of this. For this video, with all other factors being equal, I am only measuring this DIFFERENCE in delay between having the arcade stick wired directly into the PS3 versus running it through the extender. In that scenario, counting frames from each video, I should be able to see a difference in response between when the LED lights and the response on screen. I did several presses for each pass, so an average can be taken. It was very consistent across all passes. I believe in this setup, the LED lights faster than the board actually receives the signal, as it only needs a slight contact and not a complete press to complete the circuit. So the actual real life delay between press and response is likely less. I am counting frames from the moment I just begin to see the LED light to on screen response. I realize the RAP4 isn't the greatest arcade stick for lag, but I'd rather not mess up a nicer controller. And for this test, the total lag is not relevant, only the difference between direct and through the extender.

That being said, see the rest for yourself. Feel free to download these links if interested and play them back frame by frame. Quicktime Player is pretty easy since you can simply pause it, then press the forward or back arrow to advance one frame at a time. Or hold the arrows down to move quickly to the next press.

60fps videos
Video 1: Direct connection (no extender) Street Fighter II HD Remix for PS3 - https://www.dropbox.com/s/wytstirae02ii ... t.mov?dl=0
Result: 7-8 frames LED to response

Video 2: Through USB Extender Street Fighter II HD Remix for PS3 - https://www.dropbox.com/s/fhkmeh0n9tf9s ... r.mov?dl=0
Result: 7-8 frames LED to response

240fps videos. Look for the Options text to start appearing in the upper left hand corner after approx 16-17 frames

Video 1: Direct connection (no extender)- Thexder - https://www.dropbox.com/s/m11crp8q3qcnh ... t.mov?dl=0
Results: 16-19 frames LED to response (note 240 fps so approx 4ms per frame)

Video 2: Through USB extender - Thexder - https://www.dropbox.com/s/xhgcgeeb2u3fp ... r.mov?dl=0
Results: 17-20 frames LED to response (note 240 fps so approx 4ms per frame)

So I am seeing, possibly, a very minor measurable difference of 4-8ms, and sometimes none at all. Let me know what any of you see in these videos and if I'm correct here in my measurement comparison.

[gray117]

Pft that seems to suggest project success to me...

Your baseline seems a tad high for crt squinter types - but I'd hazard to say that's a fairly typical - I'd mention it though only because you'd ideally cut down on pontential chained/culminative variance by reducing that as much as possible... still the negligible measureable difference is what we're interested in.

Just gave it a google - I think hd remix on ps3 specifically is not your friend if anyone's looking at the baseline and thinking it's a little high - probably getting on for half the total time (port/system delays would ideally/realistically be half of remix's delay... so with a different test perhaps you could cut a quarter-ish of your delay total... yaadaa yaadaa ... I'll shut up because this will be another crt/pcb project of it's own, that probably won't show anything further )