Re: [Scidvspc-users] Scid vs 960 is not loading default databases

[Steve A <ste...@gm...>]

Ok, there is a Mac 960 32 bit version here. Seems to work for me.... but it
is not terribly feature complete.
Please note, it upgrades (renames) any si4 databases to .si960 - which are
then unable to be opened by the normal app.

https://sourceforge.net/projects/scidvspc/files/mac/ScidvsMac960-4.21.dmg/download

Please don't ask for a 64 bit version.
64 bit Mac Tcl has never worked properly, and imho macOS Catalina and
beyond are similarly broken by Apple's dollar/iPhone obsession.

S.A

On Sat, Oct 31, 2020 at 1:27 AM Chris Lott <ch...@ch...> wrote:

> Thanks so much!
>
> I appreciate the time you put into maintaining and improving SCID.
>
> c
>
> On Fri, Oct 30, 2020 at 1:27 AM Steve A <ste...@gm...> wrote:
>
>> Ok Chris. That is the working 32 bit version. Good.
>> When i get the chance i'll make a 32 bit Chess960 version.
>>
>> S.
>>
>> On Fri, Oct 30, 2020 at 2:36 PM Chris Lott <ch...@ch...> wrote:
>>
>>> I guess the 64-bit version (4.21). I downloaded it from:
>>> https://sourceforge.net/projects/scidvspc/files/mac/ -- is there
>>> somewhere else I should be obtaining the unbroken version?
>>>
>>> c
>>>
>>> On Thu, Oct 29, 2020 at 7:56 PM Steve A <ste...@gm...> wrote:
>>> >
>>> > Are you using the 64bit (broken) or 32bit version ?
>>> > S.
>>> >
>>> > On Thu, Oct 29, 2020 at 12:31 PM Chris Lott <ch...@ch...>
>>> wrote:
>>> >>
>>> >> Any chance of an OSX build with 960 support? Every time I've compiled
>>> >> SCID vs PC/MAC myself it has not behaved well. And I fear I've forever
>>> >> hosed my TCL/TK or whatever libraries at this point when trying to get
>>> >> the XFCC functions working.
>>> >>
>>> >> Thanks in advance,
>>> >>
>>> >> c
>>> >>
>>> >> On Sat, Sep 19, 2020 at 4:35 PM Okechukwu Iwu <af...@ch...>
>>> wrote:
>>> >> >
>>> >> > Thanks for the quick response and action!
>>> >> >
>>> >> > Using the source for revision 3138 + chess 960 patch + Mr Cramer's
>>> TKText patch (applied in that order), the problem with auto-load on Scid vs
>>> 960 is now fixed!
>>> >> >
>>> >> > Another problem persisted, though. I noticed it after I sent the
>>> earlier email. The 960 patch may be breaking something in the UCI
>>> (polyglot?) code. Both Stockfish and Toga behave strangely (giving
>>> ridiculous evaluations, or just not giving an evaluation), while Phalanx
>>> behaves normally. For example, Toga claims an evaluation of ~ +5 in this
>>> position:
>>> >> >
>>> >> > rnbq1rk1/pp2nppp/4p3/2ppP3/3P2Q1/P1P5/2P2PPP/R1B1KBNR w KQ - 2 8
>>> >> >
>>> >> > From personal experience as Black, I know it can be a tough
>>> position, but not +5 :-) !
>>> >> >
>>> >> > I compiled regular Scid vs PC from r3138 source (using Gregor
>>> Cramer's TKText patch, but no chess 960 patch), and both Stockfish 12 and
>>> Toga behave normally.
>>> >> >
>>> >> > To allow them to coexist, Scid vs 960 is installed in
>>> /usr/local/{bin,share}/, while Scid vs PC lives in /usr/{bin,share}/ .
>>> >> >
>>> >> > Okey
>>> >> >
>>> >> > On 9/19/20 1:29 AM, Steve A wrote:
>>> >> >
>>> >> >
>>> >> > OK, this autoload bug is fixed in subversion patch. Cheers for the
>>> bug report.
>>> >> >
>>> >> >
>>> >> > OK.. But there are some installation "gotchas". The new gui is
>>> "scid-vs-960", but we don't have a new name for the new tkscid.. Which can
>>> cause errors, as the new binary and gui must be used together. I recommend
>>> removing the old tkscid and scid gui before doing a make clean and
>>> configure on the patched code.
>>> >> >
>>> >> > Ill try to make a windows package sometime, as this is solid code,
>>> apart from our crappy installation.
>>> >> >
>>> >> >
>>> >> > S.
>>> >> >
>>> >> >
>>> >> >
>>> >> >
>>> >> >
>>> >> > On Saturday, September 19, 2020, Okechukwu Iwu <af...@ch...>
>>> wrote:
>>> >> >>
>>> >> >> Hello, Steven A!
>>> >> >>
>>> >> >> I decided to try out the Chess 960 patch, which built "Scid vs
>>> 960". I renamed my .s?4 files to .s?960, and can open them normally from
>>> the File menu. However, when I "Save Bases as Default", close the program
>>> and reopen, the default databases are not opened. The error message at the
>>> bottom of the startup window states:
>>> >> >>
>>> >> >> "Not loading default bases as file /Chess/huge not readble"
>>> >> >>
>>> >> >> (OK, the file path is "slightly" edited. And yes, the spelling
>>> typo is real :-) ).
>>> >> >>
>>> >> >> I'm running Linux. The database files all have read permission, so
>>> I am wondering if Scid vs 960 is still expecting the default database files
>>> to end in .s?4, rather than .s?960 ?
>>> >> >>
>>> >> >> I am sorry if this has already been addressed, but I could not
>>> find it in the mailing list archives.
>>> >> >>
>>> >> >> Thanks for all your work on Scid vs PC!
>>> >> >>
>>> >> >> Okey
>>> >> >>
>>> >> >>
>>> >> >>
>>> >> >> _______________________________________________
>>> >> >> Scidvspc-users mailing list
>>> >> >> Sci...@li...
>>> >> >> https://lists.sourceforge.net/lists/listinfo/scidvspc-users
>>> >> >
>>> >> > _______________________________________________
>>> >> > Scidvspc-users mailing list
>>> >> > Sci...@li...
>>> >> > https://lists.sourceforge.net/lists/listinfo/scidvspc-users
>>>
>>