Re: [Bug-gnubg] Set analysis limit

[Christian Anthon]

Checked the code. What it does is to cut down on the move list after
the analysis has been made. It makes no sense at all. Possibly some
occult printing purpose. I'll remove it. I the limit set at 1 (equals
infinity) without having any problems. I guess it is just one of these
things that had a point a long time ago, but isn't used anymore. The
movefilters are more versatile.

Christian.

On Tue, Aug 11, 2009 at 9:12 PM, Michael Petch<address@hidden> wrote:
>
> Howdy guys,
>
> I had an interesting discussion with Chris Yep this afternoon. Among other
> things he inquired about the “set analysis limit” command (Also known as
> move limit in the Setting/Analysis box). There is some ambiguity between
> what Chris and I believe is occurring and how this command is documented.
>
> Chris believed the move limit may influence the filtering command because
> the help says “Specify the maximum number of possible moves analysed”,
> however when he ran the analysis no matter what he set it to .. Analysis
> always seemed to take the same time. His question was basically if “set
> analysis limit” had any influence on the analsysis (or the filter).
>
> At first I told him that I thought it had no influence except on the backend
> when displaying the best moves (after a move analysis is complete). The code
> in analsysis.c that uses cAnalysisMoves appears to suggest the same (After
> the move analsysis the move list is truncated (when the move limit is >=2 ).
>
> So this begs the question. If I understand thing correctly – maybe the help
> associated with this command should read something like:
> “Specify the number of top moves from an analysis to display”
> Rather than:
> “Specify the maximum number of possible moves analyzed” (This suggests the
> analysis itself may be influenced by the setting)
>
> On the same subject, I am curious about the rationale for a value of 0 and 1
> to mean “unlimited”. I would think 0 would be unlimited and a value of 1
> should be “1 move to display”. It seems the minimum number to display is 2.
> I may not know the reason, so I’m asking :) .
>
> After playing a bit with this feature I noticed that there may be a couple
> of minor glitches. When I use “set analysis limit 2” and analyze an opening
> 6-4 the move numbers in the analysis pane are listed as “1” and “-1” instead
> of “1” and “2”. As well the users move is not highlighted in red (nor does a
> star appear beside the move number). I have a feeling these are just minor
> glitches that have crept in over time, liekly suggesting setting “move
> limit” to anything other than 2 is not very common at all.
>
> Lastly, is this feature very useful at all (It might be to people using the
> command line and console output?), do people really use it? Candidate for
> code cleanup/removal?
>
> Michael
>
>
> _______________________________________________
> Bug-gnubg mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>
>