CamStudio had the ability to draw a rectangle around the
capture region. At the time, other applications weren't
as easy to use.
CamStudio used AVI for the output container. And as
far as I know, doesn't have AVI2 OpenDML output capability,
which limits non-corrupted output to 4GB. Even when two
Google Summer Of Code people were available to work
on CamStudio, they didn't fix the AVI part. I could
have fixed that in five minutes, by gluing FFMPEG
to the butt end :-/ The SOC people were supposed to
have added multi-monitor support (think how quickly
that would deplete your 4GB max file size).
I think the "CamStudio CODEC" may have disappeared. They
had their own CODEC at one time, and it may have been
added to some other programs so you could actually use
the output. The program would pretend to copy at 200FPS,
when the engine was actually capturing at 7FPS and
repeating each frame close to 30 times.
But the rest of the feature set, is bonkers.
If you wrote a program in 1980, slid it into a
box, closed the lid, then opened it in 2018, what
do you think the program would look like ? Like
it was from 1980 or something. The controls make
no sense. The CODEC choices offered are... prehistoric.
Imagine using Cinepak when grabbing 1920x1080 screen ?
Think how slow that's going to be. (Some Cinepak
render runs operate at sub-1FPS rates. Such a choice
cannot end well. Cinepak was wonderful back when it
was used for 160x120 videos. Cinepak is not based
on frequency domain analysis of the picture.)
In my opinion, when there are many many competitors,
and a lot of them are re-using an existing engine,
we should use "Chrome" analysis. On browsers, anything
based on Chromium gets the "Hey, it's Chrome" description.
We're less likely to test such things, because a typical
developer doesn't really change much. It doesn't matter
if the developer dolled it up a bit (put his name on it),
it's still Chrome.
That might reduce the playing field to the level
that you could spot innovation.
Back in the day, there were people bootstrapping off
the AMCAP code (to do webcam capture). There were
people bootstrapping off the Microsoft RAMDisk
demo code (which set back the development of good
RAMDisks by about 15 years - tremendous damage...).
When code gets reused, the results aren't always
exemplary. Lazy guys like me paste an extra three
lines of code into it, put our name on it, and
suddenly you have another item on your list
to test.
As for the IceCream, I seem to have lost my copy :-)
I can't be sure what the download filename was.
I tried looking for IceCream but didn't find it.
It was a pretty large download for such a "simple"
operation. Look at the size of AMCAP for inspiration.
IceCream bolted together a whole pile of Open Source
stuff.
Paul