On Wed, 16 Nov 2011 12:25:43 +0100
Pierre Ossman <oss...@cendio.se> wrote:

> > 
> > Here's what I would recommend for now:
> > 
> > (1) Change BLOCK_SIZE to 256.  This didn't seem to have any effect on
> > compression ratio, and it improved performance significantly whenever
> > the CUT was enabled.  Pierre, could you re-run your high-level tests to
> > verify that this has no ill effects at that level, either?
> 
> 
> I'm not surprised. I suspected the 16x16 blocks had less than ideal
> cache usage. I see no problems with this change, but I'll double check
> with some tests.
> 

I've looked at this now, and I'm seeing significant increases in
bandwidth when increasing the block size. And that's really expected in
most cases. With a larger block size, a tiny change will drag along a
much larger area of pixels that didn't really change.

This leads me to believe that there must be something wrong with the
numbers in your document. The sizes given for BLOCK_SIZE == 16 and
BLOCK_SIZE == 256 are identical, down to the last byte. This seems
highly unlikely as that would imply that the changed areas are
perfectly 256x256 pixels every time and that there is no unchanged
pixels that are transferred as well.

Could you re-check your tests to see what might have gone wrong? If
possibly, please also test a block size of 64. From my tests here, that
size increase does very little to increase the resulting bandwidth.

Rgds
-- 
Pierre Ossman            OpenSource-based Thin Client Technology
System Developer         Telephone: +46-13-21 46 00
Cendio AB                Web: http://www.cendio.com

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?

Attachment: signature.asc
Description: PGP signature

------------------------------------------------------------------------------
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
_______________________________________________
Tigervnc-devel mailing list
Tigervnc-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/tigervnc-devel

Reply via email to