Picochess future plans

[Jeff]

Dear picochess users, dear Jurgen,

I started to develop picochess as a personal, hobby, prof-of-concept project. I stopped investing time in it mainly because of my personal life, I'm the happy father of two kids, and this tends to be very time consumming :-)

But after I left, Shiv and Jurgen took over the project (and I must admit I didn't help with a single line since Jurgen joined), they invested huge efforts in it and it doesn't look like the little project from the beginning : it became integrated in the DGT Pi product, and it is now a full featured chess computer.

I would like to dive in picochess developpement again, using the cummulated experience of all developements until now.

One point that was frustrating for me was debugging : it's a difficult point in general software developpement, but in picochess this is even harder - software is running on a headless computer, user experience varies on hardware (power supplies, sd cards, raspberry pi models, dgt board and clocks models, engine versions,...) and it's hard to detect, understand and solve all problems arising. Also, picochess users invest some money in the hardware, and they are disapointed if it does not work.

Another point of difficulty is features : I had one new idea every week, and users also have their own ideas. Ideas were going faster than my time to implement them :)

So, with those things in mind, here is my current plan : starting pichochess v2.

Here are my goals :

1) Switch platform. Keep the raspberry pi, but switch from raspbian to arch linux 64. This will allow several things : power up the chess engines (64-bit mode), reduce the image size, simplify installation and updates (by using a custom arch repository and by making picochess and arch package). It would also make picochess more easily available to a wide range of platforms (newer raspberry pis,odroids, computers,...). I think arch 64 has matured on raspberry pi, and that it will be easy to to (I may be wrong).

2) Simplify to code and debugging. Jurgen has done an impressing job by developping functionnalities, I would like to continue the job by trying reducing the code base. To begin, I would like to stick to the functionnalities described by the DGT pi and drop the unused/rare features.

3) Rework the documentation and website : the original picochess website is obsolete, Jurgen has his own website, I also found a lot of usefull docs on DJ Dekkers's site and the google group. We also have a wiki. I think all of this should be concentrated in one place : a simple wordpress site would do the job.

4) (Optionnal, when 1-2-3 are finished) Offer 'connected' functionnalities : managing picochess from a website (configuration, display and upload games, easy remote play,...)

That's all for the plan :) I'm writing this post to gather everyone's opinion. There is no timeline, but as I plan to do a rewrite from scratch (using large parts of existing code / logic) it will obviously take some time. Also, I will need help :)

-I will try to do the plartfom switch alone and setup the packages and arch linux repository

-I will do a code refractoring from the existing base, Jurgen if you are still here, your help and advices will be welcomed :)

-For the website, I can build a simple Wordpress site, but I hope picochess users will help for the content (docs, engines updates, tutorials, videos,...) : is someone interested ? The firts work would be to list all sources of documentations, and try to centralise that in a coherent way.

-For the connected functionnalities I have in my head, I think have the skills, but will think more about it in a few months :)

Jean-Francois

[Shivkumar Shivaji]

Hi Jean-Francois,

Glad to see you after such a long time! A major problem we faced is that the adoption of Picochess was not as large as hoped. Yes, there are several users, but not as many users as the number of DGT boards worldwide.

If you are tempted to work on Picochess again, I wonder if Docker should be embraced, thus making it more modern and portable for images. I also wonder why Picochess was not adopted by far more users. Opinions?

Shiv

--
You received this message because you are subscribed to the Google Groups "PicoChess" group.
To unsubscribe from this group and stop receiving emails from it, send an email to picochess+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Jeff]

Hi Shiv :)

To me, adoption is not a problem ; 10 satisfied users is sufficient ^^ This is a niche market, chess players, insterested to play against a computer, owning a DGT board, and curious enough to get a raspberry pi.

I think that the DGT pi users just plug their machine and don't come to this forum. Also we are lucky to have many developpers for this userbase.

Docker could be a nice option, but from what I understand it would still need platform-specific binaries (engines etc)? With an arch package, everything is build from source from a single line. Also, arch is small :)

Jeff

[KAROL KIRSON]

Hi Jeff You have one user. I think that simplifying the installation so that you can easily install the progam on Pc with ubuntu system or windows and run the latest stockfish without the need to compile to armv7 architecture is the right way.

[miclang]

Hi Jeff,

here's one user too - since the good old MK802 times (it's still working)...

Still interested in Picochess' future Regards, Michael

-------- Ursprüngliche Nachricht --------

Von: KAROL KIRSON <karol....@gmail.com>

Datum: 28.04.18 11:12 (GMT+01:00)

An: PicoChess <pico...@googlegroups.com>

Betreff: Re: Picochess future plans

[Jürgen Precour]

Dear Jean-Francois,

thats great if y want to take an active role again!!


I spend alot of time on these debug messages, its more or less fine for me to track down all errors which might happen (incl. different hw). But its tailored to myself :-)


To reduce functions, im not so optimistic ;-) On my newest version i moved away the "remote play" completely, cause its setup needs alot of knowledge, and it never worked very well even y did configure your hardware.
I also began to do something myself - with the goal:  y do not need to change yr router setup - but i stopped it now! I was able to play an example game with AL over iNet - without fiddling with the router setup at all.
Despite this, im not sure, what y can switch off. As an example I once asked if i can turn off "remo te engines"....i had the feeling noone using it...but when i asked for it > "NOOO, i want to use this, but i dont know how" was my given answer > so, i made the usage easier...and this is a example feature, DGT is not interested in at all / much.
You might turn off the webserver stuff - cause thats also not soo important from DGT eyes, and if y want to stick with what DGTPI is offering, y can think about this.

If i were you , i would just turn off this "remote play" , as i did --- this is not much work to do to get rid of it.



For the website...I can offer y to open mine to interested people (writers.....) if y want. Or y do all again from beginning ... up to y.


And im always here .... if y want to ask me something ...do it here ...or send me an eMail. If too technical perhaps better by eMail (not this one!, but the one y know already - from former eMails i sended y).

Yours,
Jürgen

[Miguelon]

Hi Jean-Francois, I have been a PICO-chess user for a while but I have ZERO knowledge in programming so lately PICOCHESS has become a niche for very few people who  have a dgt board, a raspberry, clock AND know programming. Whats the main reason to have PICOCHESS? I believe the main answer is to have a tiny computer that u can carry around. I can do this already with my old DELL mini notebook or my windows tablet. Engines variety? I also have like 12 engines in my PC. Later on I bought a DGTpi but support is lousy.

Last weekend I installed the latest DGTpi version to check the ponder feature and the voice feature. The voice feature didn't work, I tried for  hours with my basic computer knowledge to make it work but couldn't make it work. Later I learned u have to play with the picochess ini file. Then I tried the "Ponder" feature while reading a chess book over the board, I became frustrated because I had to "scan the board" everytime the position changed while going thru game variations. After this I decided to sell my DGTpi.

Picohess v2 should provide features a regular computer doesn't or it does poorly. Agree with you removing a bunch of silly options. Picochess should have 2 main options: Play chess and Study chess.

I mentioned in other post that PONDER "should be able to help me while Im studying a chess book by telling me whats the next move. This can be accomplished by "Ponder" but every time you change the position Picochess gets lost. So you have to again go thru all the options to scan the position again. We should be able to do this quick. e.g : Press center bottom once to scan position with white to move, press twice to scan the board with black to move".

Picochess should also be easy to upgrade ....etc etc

Good luck with your project. People like you, Shiv and Jurguen are needed in this little world. At the end Im just an end user and my contributions can be only minimal. I wished I had learned how to program but medicine called me and I answered.

Mike

[Al]

Hi Mike,

The voice function can be selected on your DGT Pi menus.
System / Voice / Pico Voice / Voice on / English (or others) / Al or Christina
“ / “ / User Voice / “ / “ / Christina or Al (having both,1 as Pico the other as User is nice)

I use Ponder all the time, I’m not sure how you get lost? Make your suggested move, PicoChess shows you the best answer and eval, replace the piece (Pico shows takeback) then move another piece, and PicoChess shows the best answer and eval etc. Multiple takebacks are possible.

Cheers,

Al.

[Miguelon]

Al, Thank you for a quick response. I actually tried the voice opt in DGTpi  and didn't work. The speaker output is in the back of the unit right?? The problem with PONDER is that you have to use take back and sometimes you get lost in your analysis specially if the variation tree is complicated - A better option would be what I said. Press a button once to scan board and white plays or press twice to scan board and black plays. The less navigation you have to do on a DGT clock the better. DGT clock letters are not that great.

I will try tonight the voice option again

Mike

[Michael Byrne]

HI Mike,

You can also turn cap letters on in the picochess.ini - makes it much easier to decipher.  Now that should be an option menu!

uncomment this line ( remove the pound sound and space)

# enable-capital-letters = True

so it looks like this:

enable-capital-letters = True

from the terminal, do ~ $ sudo nano ../../opt/picochess/picochess.ini

Some other edits that one made enjoy if their  sound  works:

1. This edit ALWAYS leaves the sound ON while in "fischer" mode ( and why not , time is being added every move) even with less than minute to go

not only do hear every move, you will also hear the game result which is also blocked by current code.  Current code automatically turns off sound with less than 60 seconds on the clock. It also turns off any other announcements such as "game end".  Who care about the clock when YOU are playing a computer to practice? The clock is simply to adjust the amount of thinking time to provide to the engine either to adjust the strength or have the computer reply in a given time period , but it should not not to use as an auto flag to terminate the game.  All GUIs let the player to continue play even when time runs out UNLESS the user wants auto flagging

from the terminal, do ~ $ sudo nano ../../opt/picochess/picochess.py

  line 1094 ( control shift dash (next to the '9" you row,  will bring up goto line # )      

Change this line:

low_time = time_u <= 60 and not (time_control.mode == TimeMode.FIXED and time_control.move_time > 2)
               

To this, changes highlighted below:

        low_time = time_u <= 60 and not ((time_control.mode == TimeMode.FIXED and time_control.move_time > 2)
                    or (time_control.mode == TimeMode.FISCHER))

2.  If you want the engine to KEEP PLAYING even if time runs out - uncomment these 4 lines (around line #1100 a fewer lines down from above):

               elif isinstance(event, Event.OUT_OF_TIME):
                   stop_search_and_clock()
                   result = GameResult.OUT_OF_TIME
                   DisplayMsg.show(Message.GAME_ENDS(result=result, play_mode=play_mode, game=game.copy()))
Changes highlighted:

###             elif isinstance(event, Event.OUT_OF_TIME):
###                 stop_search_and_clock()
###                 result = GameResult.OUT_OF_TIME
###                DisplayMsg.show(Message.GAME_ENDS(result=result, play_mode=play_mode, game=game.copy()))

Enjoy!

Michael

--

You received this message because you are subscribed to the Google Groups "PicoChess" group.

To unsubscribe from this group and stop receiving emails from it, send an email to picochess+unsubscribe@googlegroups.com.

[Philip Catalano]

As someone who is in the process of building a chessboard rather than buying a DGT board - might I please ask you to incorporate keyboard as an entry method for the new version?  I have two different arduino programs that translate movements on the board to a "keyboard"

I think that this small change would open up the project to a LOT of people.  There are not many who buy DGT boards - and few are willing to program their expensive boards, but a DIY board costs as little as $50 total - and is a wonderful project for chess clubs and kids.  

[Jürgen Precour]

Hi Philip,

thats already implemented (for long). Take a look for "keyboard_move" inside the code.

This way (=keyboard) y gonna loose the "takeback" functionality and i would miss the piece recognition as i also miss "leds" on DGT boards. Nothing is perfect ;-)

Jürgen

[BR L]

I already did this and it works really well. I have made 2 reed switch controlled boards,This Small chessboard for casual play, and  a full size  tournament board..

They both use led's  to indicate the moves and pushbutton switches and shaft encoder switch for function selection.

My modified picochess code is on Github

Picochess-nonDGT

and I uploaded pics of my board to  this site

.Reed controlled board

[Jack Daniels]

Hi,

even though I not one of those long-term picochess powerusers, I want to throw in an at least two-fold "me too":

1. for thanking Jeff to step up for taking the lead in the future development of Picochess, thus keeping it alive!

2. Um Jürgen für seinen unglaublichen Einsatz für Picochess zu danken. Ich weiß, dass es mehr als nur ein gerüttelt Maß an Idealismus und gutem Willen braucht, um so ein Projekt so lange nicht nur am Leben zu halten, sondern sich als Freizeitentwickler mit den Wünschen und (An)Forderungen eines nicht immer einfachen Userkreises auseinanderzusetzen. Hier die Nerven zu behalten ist schwer und daher kann ich Deine Entscheidung voll verstehen. DANKE für Deine Arbeit und Deinen Einsatz und vor allem auch die Weitergabe von "Insiderinfos" bezüglich der Hardware von DGT. Ohne die würde mein PicoPi Projekt heute nicht so ausschauen und funktionieren, wie es das jetzt tut.

DANKE und alles Gute für Deine Zukunft!!!

back to english and sorry for probably violating the netiquette here by posting in german.

Finally, my 2c on Jeffs plans:

Concerning the switch to ARCH: I'm a long-time professional and private ARCH user and as such would welcome this decision. HOWEVER, before making this happen, one should consider the consequences for an "embedded" software like PicoChess, switching to a rolling release, which is requiring quite a bit more knowledge of the underlying LINUX base than Debian/Raspian. This clearly relates to the question "Who is going to use Picochess, why and how"? So I think this decision will be the cross-roads:

Do you want to make Picochess as easy to use and "low profile" as possible to attract as many "I just want to play chess" users as possible OR is the plan to make it as flexible and most suited for experienced "power users", who's number might be limited?

Anyways, as for all of these projects: complains and good/cheap advice are easy to come by, people who want/can help are rare. Therefore, I can raise another "me too" (also being father of two small kids that keep me busy): my time-resources are also limited, but I would be willing to toss my (limited) programming skills into the ring and maybe I can merge my Picochess FICS code with the main branch if this feature is asked for in the future. I might also help working on English and German documentations.

cheers

Jack

[Jürgen Precour]

Jack:

Herzlichen Dank für deine Worte!

Leider wurde gerade aktuell nicht von allen Deutschen so viel Einsatz gezeigt ;-)

And despite dev something >>> anyone can be a software tester or even record his/her voice for picochess...But i said this alot of tmes in past, aehm.

Jürgen

[Philip Catalano]

I have used OctoPrint to connect to my 3d printer.  It is very robust and works very well for connecting via the web to your 3d printer.  I mention it here because I think Jürgen did something very ingenious recently - the menu system he created is the future.  Not the exact menu system - but the idea of a menu system.  When you condense all the types there are not that many variations of how we all want our chessboards to work.  I recommend taking a moment to look here: https://github.com/foosel/OctoPrint

The code is open source and could be used in improving picochess. Web configuration is possible.  The server is already working in picochess - just a bit more functionality and it could be amazing. I am trying to learn python as quickly as I can.  

[Philip]

https://github.com/schollz/raspberry-pi-turnkey

This code should be added into future img files. I have just finished testing it on my headless Pi Zero W picochess computer. It works wonderfully.

I am against moving to arch64 because I’m using arm6 raspberry pi zero w. A pi zero fits nicely into the dgt 3000. I think a script that asks what to compile would be great. If pi64 then aaa if arm then bbb

[Dirk]

Top Wanted Features of Picochess Community

In the 60+ engines thread of Philip (see https://groups.google.com/forum/m/#!topic/picochess/KI7HTXQZszw ) Shiv is asking for the top features we want for picochess:

„What are the top features the community wants at this stage?“

Shiv

And Philips answer:

„Top feature request: easy way to change engine and settings via the web server.  

Hackaday.com etc would love an opportunity to talk about the ability to play against Alan Turing's design AND Leela on the same machine you can build yourself for less than $100“

These are my two top features for picochess:

1. I would love to see some emulated old chess computer on my DGTPI (like messtiny emulator for old Mephisto computers (ROMS published with permission on page of Ed Schroeder http://rebel13.nl/dedicated/dedicated%20as%20uci.html) 

2. Not really a feature but for me it is a must: Bug Fixing of annoying „Set Position“ error when you execute your answer move too quick after you have done the move of picochess, see issue thread „Move Detection Problem“ https://github.com/jromang/picochess/issues/267.

All other GUIs with DGT integration I know (Fritz, Shredder, Arena, Lucas) are more robust and won’t bother you with wrong position messages when doing the computer move and your move in quick sequence independent of USB or Bluetooth connection 

Dirk

[Michael Byrne]

#2 is not an issue for me at all as Pico always beeps after I move the computer pieces letting me know it’s ready for my  move. 

Sent from my iPhone

--

You received this message because you are subscribed to the Google Groups "PicoChess" group.

To unsubscribe from this group and stop receiving emails from it, send an email to picochess+...@googlegroups.com.

[Dirk]

I find all the beeps irritating and a pain - that's why I have switched off confirmation beeps and just use beep when answer move is announced.

Furthermore when you play Blitz or even Bullet then you don't want to wait until picochess is finally ready for next input.

It may depend on the slow python processing fo picochess but technically it is possible to have a robust system (even with bluetooth - usb seems to be a little faster scanning the board) like all other chess apps & programs have implemented it (android apps, Fritz, Shredder, Arena, Lucas...).

I think they just check the FEN string again when it looks illegal after expecting the computer move from board and check if it is a valid one when they assume the user move in addition.

djdekker on Github showed that even picochess was already able to do it this way before version 0.68 - so I think it should be possible and this "less robust behaviour" was introduced was introduced from version 0.68 on.

For me it is really annoying because you always have to take back the moves to get a valid position again for the computer in order to continue.

Anyway - your milage may vary and I was just giving feedback for future wishes.

Dirk

[Michael Lang]

Hi Philip, did you prepare a picochess image including that code for RPi Zero? Michael

[Philip]

I have been tweaking my own 'backup'for a little while now. It's here:https://drive.google.com/file/d/1L5VXQ8YLKJi7HnkQLPFsGjcThFa9MmcU/view?usp=drivesdk

I use a pi zero w and have tested the wifi. I couldn't be happier. It works perfectly! Upon boot it checks for whether or not the network. If it is connected then it does nothing. If not connected it reboots and broadcasts the SSID "ConnectToConnect" the password for the network is the same (with same capitalization). After you connect just open up a browser and input your WiFi settings into the form. It will reboot and connect!

The latest thing I was working on was getting the Update script to work on the Pi user account. I've modified the sudoers file but I'm still learning and it doesn't work like I want it to under the Pi account. If you type sudo -i and are then root then you can just enter the command: update

The update command updates packages, firmware, and distribution, and Auto removes no longer needed packages.

I didn't write either of these tools.

The engines are there, but as I don't have a physical board just yet I haven't been able to test them. I can confirm the PicoChess web server works though.

I'd like to get the guy's who made the Square Off Chessboard to make a PicoChess board - after all, they have suppliers for boards, pieces, magnets, and a company to build the internals.
I would love to be able to buy one for $100.

This is my not so secret goal in finding and redistributing the code/engines. I'm willing to help do a Kickstarter, but mostly I just want the board for myself.

I don't have a coding background, but I do know how to get more people to use PicoChess.

Configuring PicoChess to live report games will transition this from a flailing project into the most used chess software in the world. Currently, there are three companies that do live chess tournament management and reporting. DGT is the 800lb gorilla in this arena.

[Philip]

User: pi
Password: raspberry

[Philip]

Hi Jeff,

Brian and I briefly talked about doing a Kickstarter for a Reed switch based Chessboard that would use PicoChess. If you want to, we can collaborate on the project. He has a few Reed switch boards he's built and the design is sound.

Depending on what kind of margins we end up deciding on, I think it's very realistic to think we can offer the most advanced Chessboard ever created for $100.

Square Off has negotiated with chess.com to be able to use their product on the site, but it doesn't work yet. I think fics integration would be better suited to PicoChess since fics has more open source ideals. I think the ability to play with someone across the globe is a selling point for most people and a worthwhile endeavor.

A selling point that will change PicoChess from a niche software to widespread use is tournament functionality. For less than the rental cost of competitor products tournament organizers can buy PicoChess boards which connect wirelessly to the same wifi as the tournament host.

These two features along with all the existing features: Bluetooth, speaker, mailgun games, numerous engines, makes this the future of chess. What chess teacher would be caught without one? (On my personal one I'm building in a lipo battery - and I think this is a great 'upgrade' feature.

What do you think?

Philip

[Shivkumar Shivaji]

This def sounds exciting. I am happy to help as well. The reed switch board has to be something simple. Do you have a link to his reed switch boards?

The non robotic nature of the board makes is easy. The picochess features are easy enough to use elsewhere.

Totally agree that picochess needs to step beyond niche use.

Shiv

[Shivkumar Shivaji]

Live reports and broadcasting is not too hard. I developed the web server for picochess a while back, and the broadcast support is decent on Picochess. As long as its internet connected, it should not be a problem to broadcast more games. You might already see that a game is currently live broadcasted to all devices connected to it, it also updates in relative real-time.

I think the first goal of the kickstarter should be the hardware component which might be harder.

--
You received this message because you are subscribed to the Google Groups "PicoChess" group.

To unsubscribe from this group and stop receiving emails from it, send an email to picochess+unsubscribe@googlegroups.com.

[Philip]

Shiv,

I'm very happy to hear your enthusiasm! Here are a few sites:

Max is a distinguished English gentleman with a robust website: http://chess.fortherapy.co.uk/home/

My GitHub (I am not much of a coder, but I have tried to copy everything I can get my hands on):https://github.com/well69?tab=repositories

Sish - great start to seeing Reed switch boards:http://sishchess.blogspot.com/p/building-sish-board-tutorial.html?m=1

The very best: https://www.hackster.io/lpalcu/foldable-electronic-chess-board-fbe264

This video is why I can't settle for the DGT board: https://youtu.be/ZDDEnBZZvmo

Start with the last link. The video will inspire you.

So those are some links to get your gears going. I completely agree with you that the first point of the Kickstarter is to get the hardware out. Using square off as a sort of rough matric - they charge $329 USD for their base model which has servos and Bluetooth. I have one on order and will probably be the first owner to post a disassembly video.

So, the physical design is an 8x8 led matrix and an 8x8 Reed switch matrix (using diodes to prevent ghosting). The Reed switches should be installed vertically (or perhaps horizontally - but so far the consensus is vertically is better). The Reed switches connect to an Arduino which is coded to out put the piece movement into keyboard format (you will likely understand the code better than I do). From the Arduino it goes to the Pi.

The expenses are hardware, soldering, testing, and shipping. The code is ok as is (but could use some enhancement). I started looking yesterday for a supplier that has soldering capability and I stopped because I am uncertain whether or not it makes more sense to have PCBs custom made. I'm thinking that, at scale, it's easier to mass produce PCBs with the LEDs, Reed switches, and diodes soldered on than to trust a manufacturer to glue the Reed switch in the right place 64 times.

I started looking at a site (I forgot which, but there are plenty) that will let you design your custom PCB on their site in standard format and build it for you. I'm sure whatever company we use would do the same thing and roll the cost into the total.

I wouldn't bother with any of this if there was already a product I could buy that was even close, but there isn't. I contacted DGT hoping their new - to be released later this summer - board would contain a raspberry pi. Their response was that it would not connect to the internet (nonresponsive, but telling).

I don't know much about programming, but I'm learning... How hard would it be to have a 'settings' tab (or gear icon, or whatever) that would allow the user to change settings of the ini file from the webserver? I'd really like to be able to change engines and setting from the webserver.

By the way - for any programmers reading this - please take a look at octoprint. https://octoprint.org
Gina works full time on octoprint and it's a very robust platform for the Pi. Her business model works. I use octoprint on two 3d printers (well, one at the moment - I need to fix the other) and I've donated to her because it's just so damn useful. This is my pitch for keeping everything open source, including the physical design. The only thing I wouldn't share with the world is who we use for suppliers for various reasons.

Sorry for writing so much. I've been anxiously awaiting my vacation to end so I can get home and solder my board together. By now the parts should be at my house from China.

[Philip]

http://www.livechesscloud.com is in beta (and will probably be in beta for a while) and is free for anyone to use

[BR L]

I have posted links to 2 of my designs, although very similar there are a few differences. Pictures of my boards are on Max's website.

The first one used a 4" display to show the board position, then I saw how good the picochess webserver is, so the display was superfluous and I removed it and replaced it with a 20*4 LCD display for configuring level/engine/time/book etc

Because there are so many engines and levels, I found push button controls cumbersome, so I replaced 3 of the buttons with  rotary encoder ( up/down/press).

I wanted to be able to control my boards using picochess running on Raspberry pi or Windows, so I am currently making the LCD display connect to the arduino instead of the pi, and modifying picochess to send display messages.

Different to most designs Phillip posted, my board uses LED's in each corner of the square instead of just 1 per square, so 81 led's instead of 64.

This is the later version reed controlled board

It includes circuit diagram in fritzing format, Modified picochess code and Arduino code.

[Michael Lang]

Hi, the webserver is on but it doesn't connect to the board! Michael

[Philip]

That's probably because the base image I used was for non-dgt boards and I'm guessing you're using a DGT board?

To remedy, I'm thinking the steps in the manual install documentation that put things where they need to be if you have a DGT board probably need to be run. 

I'm on the road today, but can post a version for the DGT board in the next couple days.

[miclang]

That's right; I'm using a DGT BT board.

-------- Ursprüngliche Nachricht --------

Von: Philip <wel...@gmail.com>

Datum: 15.07.18 16:11 (GMT+01:00)

An: mic...@arcor.de

Cc: pico...@googlegroups.com

Betreff: Re: Picochess future plans