RR in Wine

[Bernard Devlin]

Hi Peter,
Don't feel bad about thinking multiple desktops worked - I think that Rev's
behaviour on 3.0 on Linux is erratic.  Sometimes it seems to be respecting
the multiple desktops, other times not.  I've also seen the message box
start up long and to the right of the screen, and then I readjust it's
position and size, and it stays put.  On subsequent restarts (with nothing
happening in between), it will be where I put it, or move back to being long
and to the right (http://quality.runrev.com/qacenter/show_bug.cgi?id=7246).
 I can understand that the broken multiple desktops is a big issue for you.
 But given the problems I've seen in the last two weeks, it is pretty minor
to me.

If you have the time could you review the bugs I've found in Rev 3.0 in the
past week or two? When I initially raised my concern that 3.0 was seriously
defective, I received virtually no support.

I cannot find any pattern to the crashing bugs, but some of those 15 or so
bugs have recipes.  I've verified those with recipes on both Fedora and
Ubuntu (since so often it's the variety of Linux distributions that gets
blamed for Rev bugs).  I also checked on 2.9 and the same repeatable bugs
are there.  (Bugs with recipes are: 7212, 7219, 7220, 7222, 7225, 7238).
 Since rev 3.0 opened in Chinese on OpenSuse (when all other applications
were in English), I haven't done any testing on that.  If you have some
version of Suse to hand, and Rev opens in English on it (or you don't mind
struggling through testing it in Chinese), it would be great if you could
verify the bugs in Suse.

BTW, in case anyone else is thinking of cross-checking these bugs, the
download links on runrev.com for the Linux version of 3.0 point to 2
different releases:

http://developer.runrev.com/components/revolution/enterprise/3.0.0-gm-1/RevEnt300Full.zip
http://developer.runrev.com/components/revolution/studio/3.0.0-gm-3/RevStd300Full.zip

I emailed support about this yesterday.  This matters, because sometimes the
update feature doesn't work (bug 5581 is reported as being fixed in 2.9 dp4,
but I saw it in later versions and Bob Warren was last week reporting bugs
in gm-2 because he didn't know there was a gm-3:  his installation told him
there were no updates available).

One guy has done us a great service in finding that there is apparently a
memory leak in rev 3.0, and he's gone back to using 2.6.1. (bug 7257).
 Whilst his test application highlights that particular memory leak, I've
run 'top' during the times I've been working with Rev 3.0, and when I'm not
running his application I don't see the same consistent ramp up of memory
usage.  That doesn't give me any confidence that his bug is the same as the
cause of my unpredictable crashes.

Between these various hangs and crashes, and the script editor being
incapable of scrolling a huge script of 36 lines, I've given up on 3.0 (and
2.9) altogether.  I'm now waiting for support to send me a download link for
2.6.1. (strangely, it's not available from the runrev download page, but
2.5.1 is still available).

Maybe 2005's technology will be more reliable than 2008's, even after the
big push on bug-fixing.  It's depressing.

Bernard

On Thu, Oct 2, 2008 at 3:21 PM, Peter Alcibiades <
palcibiades-first at yahoo.co.uk> wrote:

>
> Bernard, sorry, you are quite right.  When I first tried 3.0, I don't know
> why, I thought they were working.  I was not very well at the time, and
> don't know quite what happened.  Maybe it was an hallucination?  At any
> rate, having verified today, this is what happens.