TalkChess.com

I would like to announce the public beta of freeAquarium

what you can do with freeAquarium:

* analyze positions with different engines and save the results. These engines can run anywhere, several at once, from and to different users.
* generate a tree for every engine out of these results and minimax them, totally independent from each other.
* easily edit/view/export polyglot bin books. Every user has its own book, can compare its book to the one from other users or the pre imported ones etc...
* generate a minimaxable tree of human evaluations, also for every user.

all of this in the web, that means anytime from everywhere with every device.


If you dont want to read anything, just see the interface and play around: login with demo/demo


Have fun and let me know what you think!
mike

Interesting. I´ll try it.

Congrats on the release of the public beta Michael! I apologize that I wasn't able to produce better testing for the private beta, apparently I had terrible timing as I'm currently running out of time for several of my corr chess games and didn't even have the time to play Ed's start your engines of this week...

I support free Aquarium and hope it's a successful project.

Ovyron wrote:Congrats on the release of the public beta Michael! I apologize that I wasn't able to produce better testing for the private beta, apparently I had terrible timing as I'm currently running out of time for several of my corr chess games and didn't even have the time to play Ed's start your engines of this week...

No problem at all!
You were not the only one who had no time here...

The testing and especially the thinking you did, was very inspiring to me. I really enjoyed the discussions we had, so I hope you have some more time soon ^^

Ovyron wrote:I support free Aquarium and hope it's a successful project.

Thank you very much!

mroh wrote:I would like to announce the public beta of freeAquarium

what you can do with freeAquarium:

* analyze positions with different engines and save the results. These engines can run anywhere, several at once, from and to different users.
* generate a tree for every engine out of these results and minimax them, totally independent from each other.
* easily edit/view/export polyglot bin books. Every user has its own book, can compare its book to the one from other users or the pre imported ones etc...
* generate a minimaxable tree of human evaluations, also for every user.

all of this in the web, that means anytime from everywhere with every device.


If you dont want to read anything, just see the interface and play around: login with demo/demo


Have fun and let me know what you think!
mike

Interesting. It looks a great work, but still with some limitations. Example, if one uses a different engine it's not supported. Or, if one has to copy/paste his own previous homemade analyses he can't. Those could be hints for an upgrade.

Congrats for your work so far!

Rodolfo Leoni wrote: Interesting. It looks a great work, but still with some limitations. Example, if one uses a different engine it's not supported. Or, if one has to copy/paste his own previous homemade analyses he can't. Those could be hints for an upgrade.

Congrats for your work so far!

Thank you for the feedback!

sure it has limitations (which software hasnt), a lot of... ^^

The first point you made is true. As I wrote in the doc, I, personally, only want and need engines that run native under linux and support FRC.

But well, this is of course not set in stone. More a question of testing and using different engines. If people ask for engine xy, sure lets do it, test it etc...

The second point is not entire correct:
Already, I have two columns for that use case, a "humaneval" and a "comment" column where you can paste anything on every move for every user. If you want your already made analysis in a tree form, use the "humaneval" column, which you can minimax.

If you mean something like "I have this long analysis and want it in freeAquarium, move by move", we have to define a interchangeable format for doing this... sure lets do it ^^

So far, I made this project for me and I use it since several month for my analysis work and it works very well (for me). After some time I thought, well, this might be useful for more people than me...

But, of course, that doesnt mean that I have all the features other people need or want.
This project is intended to be an open community project, I dont want anything for it (except using it myself ^^), nothing is set in stone!

Lets talk about it, use it and let it grow!


(btw: maybe it was a mistake to call it a "beta" as people seem to think that this means work for them (writing bug reports etc), instead of just using it... for me its these kind of projects that are never finished anyway ^^ )

mroh wrote:

Rodolfo Leoni wrote: Interesting. It looks a great work, but still with some limitations. Example, if one uses a different engine it's not supported. Or, if one has to copy/paste his own previous homemade analyses he can't. Those could be hints for an upgrade.

Congrats for your work so far!

Thank you for the feedback!

sure it has limitations (which software hasnt), a lot of... ^^

The first point you made is true. As I wrote in the doc, I, personally, only want and need engines that run native under linux and support FRC.

But well, this is of course not set in stone. More a question of testing and using different engines. If people ask for engine xy, sure lets do it, test it etc...

The second point is not entire correct:
Already, I have two columns for that use case, a "humaneval" and a "comment" column where you can paste anything on every move for every user. If you want your already made analysis in a tree form, use the "humaneval" column, which you can minimax.

If you mean something like "I have this long analysis and want it in freeAquarium, move by move", we have to define a interchangeable format for doing this... sure lets do it ^^

So far, I made this project for me and I use it since several month for my analysis work and it works very well (for me). After some time I thought, well, this might be useful for more people than me...

But, of course, that doesnt mean that I have all the features other people need or want.
This project is intended to be an open community project, I dont want anything for it (except using it myself ^^), nothing is set in stone!

Lets talk about it, use it and let it grow!


(btw: maybe it was a mistake to call it a "beta" as people seem to think that this means work for them (writing bug reports etc), instead of just using it... for me its these kind of projects that are never finished anyway ^^ )

My favorites engines: Stockfish_savehash version by Daniel Jose, Sugar by Marco Zerebinati, Asmfish by Mike Byrne, McBrain by Mike Byrne.

My analyses: deep, thousands positions scored trees from opening to late middlegame with my corrsepondence chess.

BTW, I'm playing two games vs. Ulysses, maybe it's better he'll not look at my analyses......

"This project is intended to be an open community project"

Put it on github/bitbucket then, please!

Congrats on first release, nice work!

Rodolfo Leoni wrote:My favorites engines: Stockfish_savehash version by Daniel Jose, Sugar by Marco Zerebinati, Asmfish by Mike Byrne, McBrain by Mike Byrne.

I spotted a typo there that needs correction.

asmFish is not by Mike Byrne. (He has never claimed to be the author, but unfortunately some people did pick up that idea.)

And of course Sugar and McBrain are 99% Stockfish (just like asmFish is, except that asmFish embodies what would normally have been a few centuries of manual labour by a person using the nickname Mohammed Li).

syzygy wrote:

Rodolfo Leoni wrote:My favorites engines: Stockfish_savehash version by Daniel Jose, Sugar by Marco Zerebinati, Asmfish by Mike Byrne, McBrain by Mike Byrne.

I spotted a typo there that needs correction.

asmFish is not by Mike Byrne. (He has never claimed to be the author, but unfortunately some people did pick up that idea.)

And of course Sugar and McBrain are 99% Stockfish (just like asmFish is, except that asmFish embodies what would normally have been a few centuries of manual labour by a person using the nickname Mohammed Li).

Of course, you're right. But it was a too long process to write "SF derivate, bt Tord Romstad, Marco Costalba..." etc for any engine. Asmfish is an AsmfishW derivate which is a SF assembly version, 99% by Tord Romstad, Marco Costalba, ....................... Mohammed Li, Mike Byrne! LOL