Timothy Miller wrote:
> On 1/2/07, Lourens Veen <[EMAIL PROTECTED]> wrote:
>> On Tuesday 02 January 2007 15:55, Timothy Miller wrote:
>> > On 1/2/07, tonyb <[EMAIL PROTECTED]> wrote:
>> > > "Standard" lossless compression will reduce the amount of data by
>> > > less than a factor of four. It would still take over an hour to
>> > > move the data over the serial line.
>> >
>> > That all depends on the nature of the data. How compressible is a
>> > file filled with nothing but NUL characters?
>>
>> So, does anyone have a PCI bus trace lying around that we can have a
>> look at to see how compressible it is?
>>
>
> You know, we could possibly produce one from the tests I've been running.
I've made a simple compressor/decompressor, for 8byte wide binary input use
./srle -c -w 8 <bin8 >bin.rle-1x8
If there is more than 256 of same values (should not be) you can use
wider count field:
./srle -c -w 8 -t 2 <bin8 >bin.rle-2x8
This is not a full RLE, just an encoding of repeating values. If there
is a need to support full RLE (more complicated in hw) I can code it
next time.
If anybody has some raw binary dumps of the sampled data, please test it
and report how much does this encoding compress the data. Thanks,
The source code: http://diary.rozsnyo.com/2007/01/03/srle/
or wget http://diary.rozsnyo.com/2007/01/03/srle-0.2.tar.bz2
Daniel
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)
