Hi @david!
It is still under development. I’ll keep updating here on my progress as well as when it’s ready. I’ll post all the necessary files and instructions then.
The right version of the Dpad looks really good.
Alot of people have access to 3D printers these days, so you could just make a version of the buttons in a step file with no “icons” on. And people who wants to can 3D print their own versions of the buttons.
Can wait to make this remote for my self 
Best Regards
1 Like
Thanks! I also prefer that one and it is easier to make good idea with the buttons being blank!
The right version looks much more elegant! SO excited to see where this project goes.
1 Like
Put me down as another vote for the d-pad on the right. That looks pretty much perfect to me. I wouldn’t complain about having up/down arrows on the side buttons either, but if we can just print our own it’s not a big deal either way.
1 Like
Hi @marton; definitely interested here, and i can even help with drivers and/or feature requests given the fact that i am currently a beta tester for the Neeo remote (which will no longer be supported). I see a lot of similarities and i hope this remote can replace my Neeo remote. Ping me if you want more info so we can sync up.
In addition, i’ve written some drivers for the Ziggo Horizon box to control the current box over IP (NodeJS application) and i am currently working on a driver for the NVidia Shield TV (bluetooth driver)
I love this project. Now that Harmony remotes are less viable this seems like an amazing thing to replace (and massively exceed) that functionality.
I want one.
Great! Thank you. Any help is appreciated!
Meanwhile I came across this chip:
Something that could work for this use case.
Hi,
Just discover your amazing project. I would ask : Is it possible (not too late) to add an accelerometer and maybe even a gyroscope?
An accelerometer would be really great for the many opportunities it open for the future, and it could detect when you take it to re enable wifi faster and turn on the screen, in the future could have gesture, detect when you don’t use it to turn off faster to preserve battery. And many other things, for instance just turning the remote facing up or down could activate some scene, like if I put screen down i put the house in not disturbing mode and cut notification and other disturbing stuff.
Other thing, depending on the buttons you implement the remote could act as a android TV remote and that could be the killer app. I would be so happy to buy you product to replace my current remote by a better one that can control my tv and my house
Hope you’ll do a kickstarter !! but please don’t rush it, kickstarter is hard and we want you to succeed.
You got a nice design, please keep a nice look and design.
Don’t put too many buttons and try to avoid bad looking button like old rubber one. keep it nice and design
wish you the best and will be happy to get my hand on it 
PS: For the accelerometer this one is the one use by xiaomi in its vibration sensor because very little current http://www.analog.com/media/en/technical-documentation/data-sheets/ADXL362.pdf
Hi! Thank you and for all the suggestions!
To be honest, I am running out of space I’ve added 13 buttons that take up much of the real estate on the PCB. However all the features you’ve just mentioned are possible to implement with the proximity/gesture/ambient light sensor.
Right now it’s setup in a way that if it detects proximity (meaning I am reaching for the remote) it turns on the screen and wifi.
Facing down could be also detected with the proximity distance being almost 0.
Software wise I think anything is possible. It’s just time
I don’t want to rush it, I am working on the hardware part now. Finishing touches on the schematics and PCB layout before routing it. At the same time also designing the new button assembly, doing silicone pouring tests for the button under layer and making the case ready for test production. My focus is now to produce a couple of hardware ready remotes so I can test them. It is much easier to write the software when you have the actual hardware in your hand.
I have some plans in place how to optimise the battery life even more. I realised that right now I am not putting the display to proper standby mode as well as the other ICs, so I hope in the next version that will improve the battery life. Also I am thinking to use bluetooth instead of wifi communication with the HA server (meaning raspberry pi in my case) to save power. And only use wifi for software updates.
I hope that with these 8-10 hours would be possible. Unfortunately, I don’t think that more is possible with a raspberry pi zero. Using another chip, for example the one I have posted above would help, but designing a circuitry around such a chip is out of my capabilities.
When the hardware is ready and I have tested it, I will upload the files so anyone can download it and build their own remote. At this point a very basic software will also be available that will allow you to control lights, blinds, climate entities and media players.
This is my plan, kind of If you guys want to buy a kit or assembled remote, I’d consider doing a kickstarter project, so I could actually fund the order of XX amount of displays, cases, make professional silicone underlayers and plastic buttons, etc. But then it would still be a kind of DIY project maybe with the option that I assemble some.
Any ideas, suggestions are welcome. I’d like to get the most out of this project.
Marton
2 Likes
That’s would be a deal breaker for me, and many other i guess, because of the short range of bluetooth. (i’m using a nuc not a raspberry, it got bluetooth but it’s too far and anyway many don’t have bluetooth)
Maybe we could choose which one we wanna use ? I mean if it’s not too much work for you.
1 Like
For me, battery life less than days is a deal breaker. I would love to see a Bluetooth or other option. You could build a Bluetooth hub with an esp32 that bridges Bluetooth to wifi and put it in the same room. I have been thinking about doing something similar to this project for a while, but I don’t know how to develop hardware that looks this polished. I am hoping to build upon this and develop a version with very long battery life.
I have been using a the first version of this remote for a year now and battery life was not a big issue for me. When I don’t use it anymore I put it back in the dock. However I can see that a longer battery life would be a huge improvement.
All the hardware skills I gathered is from the internet. I am no engineer so probably someone with more knowledge could make it even better.
However I am curious to see how the new board and some software tweaks would improve the battery life. Otherwise the only route I can imagine is to design a full custom board around a low power chip. Maybe in the next version. That would take me years for sure 
1 Like
I have been dreaming of an open source remote like this for a while. I’m sure we will end up with something that makes control 4 and savant dealers jealous. Really excited to see how this evolves as more people are able to contribute.
It’s impossible to make something that fits everybody’s needs, but I hope it’s gonna be something that more people can use
1 Like
I agree with @bobzer that this would be a potential issue for a lot of users, whether it be because many of us have moved to more powerful hardware that doesn’t have bluetooth or because the HA server is too far away from most of the locations a remote would be useful.
1 Like
Probably the first iteration of the software is going to be wifi based as I have no idea how bluetooth communication works yet But I think having both options would be best.
4 Likes
For sure that would be best or maybe even zigbee ?
I mean for V2 if you’re going to design a new board with low power optimize chip maybe going to zigbee would be a good idea because low energy and not the same range problem as bluetooth.
Don’t know about the low energy of zwave but same range benefit and less power drowning than wifi.
Also you will discover that bluetooth is kind of difficult to work with … I mean a POC is easy but a reliable connection without unwanted disconnection it’s pretty tough.
1 Like
Rather than trying to shoehorn a new chip in there you could add a second Pi in the dock - power usage is not a concern there since it’s always plugged in. Then the remote simply speaks Bluetooth to the dock and the dock speaks WiFi to HA.
That said this definitely adds complexity and probably latency. So I’d personally probably only ever want to use wifi from the remote directly.
1 Like
How about restricting the remote design to be something like this; https://cdn-reichelt.de/bilder/web/xxl_ws/X600/NEAH-6336-REMOTE_03_ANW.png
Implement an ESP32 on the remote itself and have that communiate with a ‘brain’ just like the neeo remote which contains the Pi zero? In fact, i even opted to the guys at Neeo why they hadn’t combined the brain with the charging base to begin with.
So my ideal remote would be:
- Charging base with Pi Zero in it, containing all the radio’s (BLE/WiFi) and maybe even an IR emitter with 2 or 3 outputs
- Remote with as low power hardware as needed, connecting with the base using either websockets but indeed ‘when needed’. If you flash it correctly (in case Android studio is used) you should even be able to enable OTA FW updates
- Charging base then also has a mini-HDMI out so you could use HDMI-CEC to detect if appliances are on or off.
- Multiple charging bases can be joined using MQTT
- Putting a remote in a different charging base can change the ‘logic’ for that UI (different room, different appliances, available actions)
- For me, the possibilities are quite a lot.
The benefits for me would be that you would have a expandable eco-system (the base) which can be fully open source (SSH access) and the remote could even be just displaying a UI which is hosted FROM the brain, so make it also extendable for ‘interested’ parties.
Connection between the brain and the remote can be done by mDNS (as Neeo does) and the hardware buttons can be UDP broadcasts (which Neeo also does); no patent infringement there.
By the way, i am currently investigating if i can ‘hijack’ my current remote (The Neeo one) to display my own UI hosted on a Pi zero; if i can get that to work there is a whole lot of people who already have some sort of hardware and they might be able to ‘jump’ on this train.
4 Likes