Robert Green has mentioned VBOs being fast for his uses a couple times
on this list, I think. Might be worth searching those posts out.
Not actual usage, but I saw them mentioned in some docs recently. It's
for a newer version of OpenGL ES, but this ImgTec document recommends
VBOs, for example:
http://www.imgtec.com/factsheets/SDK/POWERVR%20SGX.OpenGL%20ES%202.0%20Application%20Development%20Recommendations.1.1f.External.pdf
ImgTec's PowerVR SGX design is in the Droid
Qualcomm also recommends VBOs:
http://www.brewconference.com/brew_2007/pdf/Prog-401_Dorbie_v05.pdf
Although that presentation is for BREW, not Android, and skips over
their MSM7200 design which is in the G1.
The differences in the documents are interesting. ImgTec recommends
interleaving all the data for a vertex together. Qualcomm recommends
tightly packed instead. ImgTec recommends always using an indexed
triangle list. Qualcomm recommends triangle strips, stitching them
together with degenerate triangles for screen aligned quads.
On Feb 2, 8:10 pm, Peter Eastman wrote:
> I was wondering if anyone has experience with the relative performance
> of vertex arrays vs. vertex buffer objects? On a desktop computer,
> you expect VBOs to be much faster since the data can be stored in
> video memory. But hand held devices don't usually have separate video
> memory, so I'd expect little or no difference between them. How do
> they actually work in practice?
>
> Peter
--
You received this message because you are subscribed to the Google
Groups "Android Developers" group.
To post to this group, send email to android-developers@googlegroups.com
To unsubscribe from this group, send email to
android-developers+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/android-developers?hl=en
