I still wonder how hard it would be to make one of Óscar Toledo's chess programs "http://www.nanochess.org/chess.html" to play from any game state you choose. |
| |
Sorry that the Linux binary/tarball are so old. I didn't even think anybody was using these. There's a lot of technical debt I'd like to get to on Escape, but only so many hours in the day! Glad you were able to get it to work at least.
SSL is definitely something I'd like to do -- I tried with letsencrypt a few years ago but it didn't work for some weird reason. Maybe time to try again. |
| |
You have less than one day to post the next entry. |
| |
for getting rid of trace and spare turing tape, make it so that the letter representing halted state (let's call it H) can combine with character to either side to become itself, and delete everything else that way, and lone H would be the destination word. If the starting word couldn't reach “blah H blah” in the first place, it wouldn't have an h to perform the "draw anything" trick with (catch 22, basically). All rules are reversible, so a word that can't reach the destination (H) can't be reached from destination either |
| |
I just realized why people cam here to comment about the chess article just now. That was when tom7 posted the video adaptation "https://www.youtube.com/watch?v=DpXy041BIlA" . |
| |
I wonder how possible it would be to create an engine that meaningfully tries to draw the game. It just wants everyone to be friends!
I can think of at least two actionable ways to do this.
First, it always chooses the stockfish move that brings the advantage as close to zero as possible. This will make tons and tons of draws. It will not be very fun, because it was will just be super-humanly boring and unexploitable (by humans, at least).
A more fun way would be an engine that detects the difference in pieces and attempts to create parity. If it is behind by a pawn, it will try to attack pawns. If it is ahead a bishop, it will attempt to donate a bishop to it's opponent. |
| |
Hi – I found what I think is a typo in one of your papers:
"rare. The dual of dangerous; the score is one minus the probability of ending the game on that square.This player has a thirst for adventure!"
Isn't rare the dual of popular? |
| |
I have some ideas for other chess engines you could implement (or could have implemented).
Longest road - the engine tries to have the longest continuous distance of contiguous pieces (measured where diagonals are root-2 and cardinals are 1)
Dominion - the engine tries to have the longest possible perimeter for the pieces of it's color.
Perfect squares - starting with a1 = 1, a2 = 2, a3 = 3, b1 = 9, etc... the engine tries to order it's pieces such that the sum of the values of the squares it controls is as close to a perfect square as possible.
The traveler - the engine selects whatever move will allow it to move a piece the most squares in any given move. Ties broken randomly among candidate moves.
Fearful - the engine prioritizes pieces that are closest to the enemy pieces (chebyshev distance) and moves those pieces as far away from the enemy as possible. Ties are broken first by moves that huddle, and then randomly.
Stockfishfried - Selects the first move of the previously moved piece that Stockfish suggests. For example, if the opponent moved a knight, select the first knight move that stockfish suggests. If the opponent moved a light-square bishop, select the first light-square bishop move. If the opponent moved a piece SFF does not have, it chooses a move at random. What's interesting about this is that it makes stockfish "self-select" from it's opponent's pieces. If their opponent only has a king left, SFF will only move it's king (unless it can't legally move it, in which case it will move randomly). |
| |
You'll excuse me but I just had to.
Secretly hoping that it doesn't break anything actually :| |
| |
Can we have TLS on escape.spacebar.org (both when using a web browser and when connecting from the game)? If Tom 7 is willing to change the server configuration, I can look into changing the client code in the game. |
| |
I'm on Fedora Linux (which currently has Escape 200912250 packaged), and after suffering data loss due to the one-bookmark bug too many times, I wanted to upgrade to the latest version. Unfortunately, the Linux build and the source tarball are both out of date. I managed to dig up the correct SVN link ( https://sourceforge.net/p/tom7misc/svn/HEAD/tree/trunk/escapex/ for the web interface or https://svn.code.sf.net/p/tom7misc/svn/trunk/escapex for checkout; the dependency "cc-lib" is alongside it) and successfully built from SVN. I hope this helps the next person. |
| |
Probably:
https://player.vimeo.com/video/301097306 |
| |
Is "bizzaro-world TED talk" supposed to be a link to anything in particular? |
| |
You have more than a day to post the next post. |
| |
I watched this at the beginning of April via jonas's comment, but it was well worth a second watch now. Fun video. Part of me wants more Tom 7 Albums-a-Day, but if this is the form of creativity that inspires you now, I'm not complaining. Good luck in the marathon! |
| |
|
|
| |
Hi Tom, at the end of the paper, you write "Please see http://tom7.org/nan" but it appears that project page is actually "http://tom7.org/nand". What do I do? Best regards. |
| |
The next month is never more than 31 days away! |
| |
Thanks! I'm using PCBWay, which also supports four+ layer, but (like basically anything that strays from the defaults) it increases the price from "suspiciously cheap" to "that's kinda what I expected." That's still somehow hard for me to justify to myself even though I'm willing to pour dozens of hours into the project, take vacation days, etc. (?). Of course the real reason to stick with two-layer is that it makes it more challenging to hand-route the board! |
| |
13620. calendar (176.63.24.170) –
30 Apr 2019 17:40:53
[ CHESSBOVIK ]
You have less than half a day to write an entry for month 04. |
| |
This one and #23 have grown on me a lot over the years. Turtledog, Hot Interrobang, Mr. Natural and Ghostlines are all choice.
Also my fake bad review above just seems mean now. Sorry! It was supposed to be funny. |
| |
|
|
| |
|
|
| |
One of the things that I've learned over the years is that an excellent way to get a good result at something is to "teach Tom7 the basics of it, and watch him eclipse you rapidly". In this vein, I have wondered over the years what would happen if somebody gave you programmable logic devices: surely performance of the NANDY 1000 would be greatly improved by using some CPLDs as the basic gates, and even then, you might have enough logic left over to implement the rest of the NANDY 1000! Either that, or we'll end up with a hardware description language embedded in SML, and either of those sound like good outcomes to me.
All the same, I enjoy seeing the fruits of someone's labor in a densely routed board, and am suitably impressed. I fully expect a 4-layer board [1] with DDR3 [2] routed for next year's SIGBOVIK :-)
[1] It's more possible than it sounds! OSHPark's 4-layer process is quite good.
[2] It's, uh, comparably possible to how it sounds! |
| |
13615. jonas (176.63.24.154) –
01 Apr 2019 17:36:10
[ CHESSBOVIK ]
You have now uploaded the video that you promise in this post: "https://www.youtube.com/watch?v=5TFDG-y-EHs" |
| |