Hum that looks like a loss of free resources to me.
If i want to review the analysis on several machines i have to re-run the analysis on each of my machine. It's then cheaper to run the server analysis.
Or if my opponent want to review the same game, he will also have to run analysis on his side as mine is not share...
It's a lot less interesting to run the local analysis then.
If i really need a deep analysis (never), easier to copy paste the pgn to a local stockfish.
How can it be more abused than fishnet ? I guess it's the same verification process and upload. Why not turn any "client" to a fishnet node if the user click Run local analysis.
Still waiting for my fishnet API key...
@gamblej said [^](/forum/redirect/post/WCiF9JP7)
> You can, however, upload them to a study.
How do I do that? I ran local analysis of a study chapter but I cannot see the results when not logged or even when logged under my account on a different machine. And I couldn't find how to make the local analysis results public. Without that, the feature (that I have hoped for for quite some time) would be rather useless for my purpose.
@gamblej i saw a server analysis with 300.000 nodes instead of one m , then i did one locally with standart qualitiy and got 1,1 m instead of one m , how is this possible ?
@gamblej i saw a server analysis with 300.000 nodes instead of one m , then i did one locally with standart qualitiy and got 1,1 m instead of one m , how is this possible ?
If you are a contributor to a study, and that study allows analysis, then you should see a Publish button at the bottom of your analysis editor.
If you are a contributor to a study, and that study allows analysis, then you should see a Publish button at the bottom of your analysis editor.
@najdorfplayer48 said [^](/forum/redirect/post/ld1EQApl)
> @gamblej i saw a server analysis with 300.000 nodes instead of one m , then i did one locally with standart qualitiy and got 1,1 m instead of one m , how is this possible ?
> Bild 22.10.26 um 17.41.png (https://image.lichess1.org/display?fmt=png&h=0&op=resize&path=HXhVu25uzCDb.png&w=864&sig=2225a06063c3c92ee547e8082ed68df6906de6ff)
Link?
You got 1.1m instead of one because you used more than one thread. As threads increase the target node count is adjusted upwards to account for a bit of parallelism inefficiency.
@mkubecek see https://lichess.org/forum/lichess-feedback/server-vs-local-analysis?page=2#OBkd9Esw
@gamblej said [^](/forum/redirect/post/ipDGGPFo)
> > @gamblej i saw a server analysis with 300.000 nodes instead of one m , then i did one locally with standart qualitiy and got 1,1 m instead of one m , how is this possible ?
> > Bild 22.10.26 um 17.41.png (https://image.lichess1.org/display?fmt=png&h=0&op=resize&path=HXhVu25uzCDb.png&w=864&sig=2225a06063c3c92ee547e8082ed68df6906de6ff)
>
> Link?
https://lichess.org/game/Az87xUzS/white#0
@gamblej said [^](/forum/redirect/post/OBkd9Esw)
> If you are a contributor to a study, and that study allows analysis, then you should see a Publish button at the bottom of your analysis editor.
Thank you, it's there now. I'm sure it wasn't there on Sunday when I created the study and ran the analysis.
Well you'll have to ask Sunday about that.