Just a heads up

https://www.gnu.org/software/gnubg/

Not sure what's new, first update in many years.

On 11/21/2022 11:39 AM, Nasti Chestikov wrote:
> Just a heads up
>
> https://www.gnu.org/software/gnubg/
>
> Not sure what's new, first update in many years.

There's some information here:

https://lists.gnu.org/archive/html/bug-gnubg/2022-11/threads.html

---
Tim Chow

On November 21, 2022 at 9:39:09 AM UTC-7, Nasti Chestikov wrote:

> Just a heads up
> https://www.gnu.org/software/gnubg/
> Not sure what's new, first update in many years.

They had announced a preview version by email
to the bug-gnubg list a few weeks ago (which my
below thread was about).

"Bug-gnubg censors their own daddy Wong's "official complaint form"

https://groups.google.com/g/rec.games.backgammon/c/HWVvHeLgCpA/m/qSyhN5kBBgAJ

Then they announced the release by email to the
bug-gnubg list again last week. I wanted to wait
and see if they would announce it here also but
they haven't.

I can understand if they feel like they are an elite
group and sneer at participating in discussions
in RGB but I can't understand why they wouldn't
announce new releases of their precious bot here?

MK

On November 21, 2022 at 12:57:05 PM UTC-7, Tim Chow wrote:

> On 11/21/2022 11:39 AM, Nasti Chestikov wrote:

>> Not sure what's new, first update in many years.

> There's some information here:
> https://lists.gnu.org/archive/html/bug-gnubg/2022-11/threads.html

Not much of an update, other than now it has
Python v3 and 8,027 files vs 7,068 previously.

For some reason, they aren't very receptive to
suggestions for fundamental improvements,
like my suggestion to add player settings for
truly random checker and cube decisions, as
well as doubling/taking points ertered by the
user, etc. even though the existing capability
of setting checker and cube levels separately
for each side is a great/unique feature of Gnubg.

Even though I maintain that equity, luck, error,
etc. calculations are inaccurate and useless,
they could at least try to improve them by, for
example, including the "position complexity" in
them as I had suggested years ago (i.e. "cost"
of errors varying by position complexities at
different stages of the games).

About a month ago Tim suggested a similar
improvement, from a different perspective on
"position complexity" (which I don't subscribe
to but neither oppose, as any progress is better
than stagnation), even including ideas on how
it could be implemented but it didn't receive a
single response from the group.

Too bad... :(

MK

On Wednesday, 23 November 2022 at 23:03:00 UTC, MK wrote:
plexities at
> different stages of the games).
>
> About a month ago Tim suggested a similar
> improvement, from a different perspective on
> "position complexity" (which I don't subscribe
> to but neither oppose, as any progress is better
> than stagnation), even including ideas on how
> it could be implemented but it didn't receive a
> single response from the group.
>
> Too bad... :(
>
> MK

The dice.c files have different file sizes. I did put them side by side in Notepad++ using the Compare plugin but my knowledge of C++ is sketchy so I don't know what (or why) the additional code is doing.

On November 25, 2022 at 10:26:00 AM UTC-7, Nasti Chestikov wrote:

> The dice.c files have different file sizes.
> I did put them side by side in Notepad++

You can actually look at revision logs of
each source file at:

https://cvs.savannah.gnu.org/viewvc/gnubg/gnubg/

You can click on the revision number of any
file to see a history of all its revisions with
brief descriptions, like this one for dice.c

https://cvs.savannah.gnu.org/viewvc/gnubg/gnubg/dice.c?view=log

One of the many options is "Diff to previous
version". If you click on that for any version,
it shows all changes in a really nice, side by
side and color coded format, even including
three lines of code above and below the lines
modified, like this last version for dice.c

https://cvs.savannah.gnu.org/viewvc/gnubg/gnubg/dice.c?r1=1.105&r2=1.106

> using the Compare plugin but my knowledge
> of C++ is sketchy so I don't know what (or why)
> the additional code is doing.

Some (most??) changes are commented but
you may need to look at more/all of the code
to understand what is doing what, which may
not be so easy even for people who know the
language well and who are already familiar
with the entire pile of code.

It's often fun enough to poke around. Enjoy... :)

MK