This came up in a recent discussion on the advanced board:
You could create another buffer object and copy the data, but that would be very innefective. We need a server-side buffer copy
http://www.opengl.org/discussion_boards/ubb/ultimatebb.php?ubb=get_topic;f=3;t=015308
The background is compacting the buffer object after transform feedback when you are unable to accurately guess the amount of data in advance.
I thought I post it over here because here it’s more likely to be noticed…
Splendid idea.
A generic CopyBuffer/CopySubBuffer would be nice indeed (use for both buffers and textures). I’ll bet something like this is already slated for LP/ME.
I know this was my suggestion and such 
but still, is it really so important? No doubt, it is nice functionality and completes the buffer object specs, but could anyone describe some situations where such functionality is really important?
P.S. I was speaking about current GL. In LP, where images are expected to be buffers, such function would be much more useful
Of course I’m referring to Longs Peak. To be honest, I don’t really care about features for the current GL anymore
As for the use cases, we have one already. I’m not sure if this feature is really important, nor do I know other use cases. But it seems worth considering, that’s why I posted it over here…
To be honest, I don’t really care about features for the current GL anymore
Good; nobody else does either
In any case, the feature seems reasonable enough, and isn’t exactly an API burden or anything.
