yorkman wrote:First, thanks for this overdue feature. I've been waiting for SF to have this option for a very long time now.
Second, unfortunately the code isn't working properly. It only works up to a 2 GB hash size and it takes extremely long to save it...about 10 mins. For comparison, I save the asmFish hash (32 GB) in about 7 mins).
Hopefully you can fix that...or if you can at least provide instructions on how to add this feature into the source code ourselves (including a fix) and which file we're to add it in we'd really appreciate it.
Hello. Have you tried with the latest version I published few days ago? Should be solved.
Hi Jose'. Thanks again for your efforts. Can we assume now that the your code in Stockfish gave to the software, as long as persistent analysis is concerned, all the capabilities that were in Stockfish-TCEC6-32-PA_GT?
No way.
Not much is missing for a better start.
Daniel is very smart (not to be confused with Danniel who is dumb as a box of ball peen hammers). Loading the hash from a big file of EPD is all that is missing. And SF already has a function to parse EPD.
You hit the point, Dann. That's exactly what I was thinking this morning.
I remember you have a lot of positions. Could be really useful for you.
Dann Corbit wrote:Loading the hash from a big file of EPD is all that is missing. And SF already has a function to parse EPD.
So you propose to add a button to load the file instead as a hash file, as an epd file that will go to hash. It can also be done for example that these entries will be marked as read only, so they will be mandatory in engine analysis.
Just for curiosity, will you add to the epd file various related entries to route the analysis to where you want, I suppose.
Dann Corbit wrote:Loading the hash from a big file of EPD is all that is missing. And SF already has a function to parse EPD.
So you propose to add a button to load the file instead as a hash file, as an epd file that will go to hash. It can also be done for example that these entries will be marked as read only, so they will be mandatory in engine analysis.
Just for curiosity, will you add to the epd file various related entries to route the analysis to where you want, I suppose.
A step forward!
And your words inspired me:
cdani wrote:................................
It can also be done for example that these entries will be marked as read only, so they will be mandatory in engine analysis.
......................................................
If that can be done on PV nodes which reach a tuneable depth setting then you have a perfect PH system while on analysis too. Then, it'd be only matter of flag them as "read only" for saving hashes and, of course, to load them as "read only" when loading hashes.
Rodolfo Leoni wrote:
If that can be done on PV nodes which reach a tuneable depth setting then you have a perfect PH system while on analysis too. Then, it'd be only matter of flag them as "read only" for saving hashes and, of course, to load them as "read only" when loading hashes.
Maybe I'm wrong. If positions are read only, how does score propagation happen?
An idea of read only entries is for example to signal the engine that an endgame is draw, even if it thinks that is wining. So in fact you force it to avoid the endgame or to search another different path to it. Is just something I imaginated, don't know if it will be very useful.
Anyway you should be able to load various epd files, read only or not.
cdani wrote:An idea of read only entries is for example to signal the engine that an endgame is draw, even if it thinks that is wining. So in fact you force it to avoid the endgame or to search another different path to it. Is just something I imaginated, don't know if it will be very useful.
Anyway you should be able to load various epd files, read only or not.
Really interesting.
BTW, I'll post a result of a correspondence game, with a position, to show how PHs have been useful. I'm only waiting for opponent resign, so it could be today or within a month.
Dann Corbit wrote:Loading the hash from a big file of EPD is all that is missing. And SF already has a function to parse EPD.
So you propose to add a button to load the file instead as a hash file, as an epd file that will go to hash. It can also be done for example that these entries will be marked as read only, so they will be mandatory in engine analysis.
Just for curiosity, will you add to the epd file various related entries to route the analysis to where you want, I suppose.
They would be fully analyzed EPD records (so EXACT hash pv nodes). With depth, move and score.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
Dann Corbit wrote:Loading the hash from a big file of EPD is all that is missing. And SF already has a function to parse EPD.
So you propose to add a button to load the file instead as a hash file, as an epd file that will go to hash. It can also be done for example that these entries will be marked as read only, so they will be mandatory in engine analysis.
Just for curiosity, will you add to the epd file various related entries to route the analysis to where you want, I suppose.
They would be fully analyzed EPD records (so EXACT hash pv nodes). With depth, move and score.
Yes, this was clear. But on some cases the engine maybe will not find easily the path to some of the EPD positions, so maybe you will have to add intermediate positions.
Dann Corbit wrote:Loading the hash from a big file of EPD is all that is missing. And SF already has a function to parse EPD.
So you propose to add a button to load the file instead as a hash file, as an epd file that will go to hash. It can also be done for example that these entries will be marked as read only, so they will be mandatory in engine analysis.
Just for curiosity, will you add to the epd file various related entries to route the analysis to where you want, I suppose.
They would be fully analyzed EPD records (so EXACT hash pv nodes). With depth, move and score.
Yes, this was clear. But on some cases the engine maybe will not find easily the path to some of the EPD positions, so maybe you will have to add intermediate positions.
I have around a billion positions, including all the commonly played ones, so I guess the tree would not have a lot of floating islands.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
