On Wed, 13 Sep 2000, opus.species wrote:
> > Scott Raney <[EMAIL PROTECTED]> wrote:
> >
> > Both of these problems are almost certainly due to bugs and other
> > non-features in drivers for the graphics card you're using and not
> > anything to do with the hardware or OS. The standard procedure for
> > dealing with these is:
> > 1) Make sure you have the absolute latest drivers from the vendor's
> > WWW site.
> > 2) Disable acceleration in the Display and/or System control panel.
> > Trying other screen resolutions or color depths might also be a
> > workaround.
> > 3) Find a reproducible example and report it to the vendor and insist
> > that they fix it. We're always happy to help any vendor identify and
> > fix the problem, but have yet to be contacted by any of them even when
> > we ourselves have reported the bugs to them. Sometimes the bug
> > quietly gets fixed in a new driver version, but more commonly the
> > vendor just blows off anyone that reports bugs like this.
> > 4) Get a different, more reliable, graphics card.
>
> Unfortunately we can not always choose our customers nor let them buy new
> computers ... so we have to get workarounds !
It's perfectly acceptable for you to expect your customer to have a
working system, and if they don't, to get them to upgrade or
reconfigure it so that it works right. There's a really serious
problem with the idea that you have to expend more than a token effort
to try to make your application work on all systems. You may end up
losing a significant number of sales (perhaps even a majority) to the
99% of potential customers with working systems who don't buy it
because they need some feature that you can't provide because you're
wasting so much time (and/or giving up so much design flexibility)
trying to work around problems that only affect the 1% who bought some
piece of junk computer and don't know how (or care to learn how) to
fix it. At some point you just have to write the losers off...
> According to my tests, there are "incompatibility problems" between MC, some
> video card drivers and some non-really-standard file formats.
>
> - MC display always OK the RVB jpegs, but display sometimes bad the
> gray-level jpegs or the "optimized" jpegs.
There were problems displaying some types of JPEGs 2.3.1 and previous
releases, but these should all be fixed in 2.3.2. These are
independent of the graphics card problem because a given image would
look wrong on *all* systems.
> - MC display always OK the ordinary gifs, but display sometimes bad the
> transparent gifs
Sometimes these can be fixed by setting the constantMask property to
true, but sometimes it's a driver problem (masked images are usually
the most common place to see these problems). Again, it depends on
whether the image appears correct on some systems, or wrong on all of
them.
> For jpegs, the workaround is easy : open the files in Photoshop, transform
> as rvb images and save as standard jpeg files.
> The same for plain gifs.
>
> For animated or transparent gifs, i am still looking for a reliable
> solution.
I've got one: Only distribute your application on UNIX and MacOS
systems, which AFAIK never have these kinds of graphics-card
dependent problems ;-)
Regards,
Scott
> Regards,
> Claude
********************************************************
Scott Raney [EMAIL PROTECTED] http://www.metacard.com
MetaCard: You know, there's an easier way to do that...
Archives: http://www.mail-archive.com/metacard%40lists.best.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.
