[Scummvm-devel] Announcing public SCI testing?

[Filippos Karapetis]

On Monday, July 19, 2010, Eugene Sandulenko <sev at scummvm.org> wrote:
> Great,
>
> Thank you guys for supporting this idea.
>
> Since the amount of games to test is unusually big, I am thinking about
> organizing it in manner of our pre-release testing. I briefly spoke
> with Matthew (clone2727) and he agreed to drive the process.
>

I will also have free time for this from Wednesday onwards, so I can assist too.

> Now we need to determine exact list of games for testing, particularly
> Game/Language/Platform variants, as some things may be not working.

Why? Why can't we use a list of all the PC versions of SCI0-SCI11
games? Seems to be more useful, most "big" issues have been addresed,
thus hopefully any bugs that come up should be easier to fix.

> Also we need to come up with list of known issues which should not be
> reported, so we will not be flooded with false alarms.

There already is such a list, in the SCI/TODO page, which is up to
date. I insisted in maintaining it for this purpose (and to remember
what doesn't work yet).


>
> Another thing to mention is that AFAIK, current sci-valgrind error()s
> out. I would recommend to show it with GUI dialog before exiting, where
> besides the error message will mentioned that there is console window
> which contains copy-able error message which has to be provided with
> the bugreport. Otherwise users will just report about "crashes".

True, good idea

> Speaking about our bugtracker, I am in favor of adding simple "SCI
> game" category instead of individual games, as there is not that much
> diversity and specialization among SCI developers. Opinions?
>

I'm against this, as the SCI engine follows a similar analogy to the
SCUMM engine, and we have a full list of all the SCUMM games. Unless
you want to merge all of the SCUMM games in a generalized category as
well?

Regards
Filippos

-- 
"Experience is the name every one gives to their mistakes" - Oscar Wilde