> Ther is now a binding to gtkdatabox located on:
> https://github.com/persan/A-gtkdatabox
Does it use time stamps on the X axis (i.e. oscilloscope), or is it just a
XY plot?
Is it based on cairo?
How does it handle the case when several data points must be rendered in
one-pixel horizontal width?
--
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de
On 12/07/2010 09:54 PM, Dmitry A. Kazakov wrote:
> On Tue, 07 Dec 2010 21:33:52 +0100, Per Sandberg wrote:
>
>> Ther is now a binding to gtkdatabox located on:
>> https://github.com/persan/A-gtkdatabox
>
> Does it use time stamps on the X axis (i.e. oscilloscope), or is it just a
> XY plot?
Its basicly an XY plot with float values.
> Is it based on cairo?
Have not thought about it, i was looking a widget that made it possible
to build a simple UI for a pc-oscilloscope.
> How does it handle the case when several data points must be rendered in
> one-pixel horizontal width?
Have not looked deeper into the scaling effects when several X-values
should be rendered one pixel wide, but it seems like a mean value is drawn.
/Per
> On 12/07/2010 09:54 PM, Dmitry A. Kazakov wrote:
>> On Tue, 07 Dec 2010 21:33:52 +0100, Per Sandberg wrote:
>>
>>> Ther is now a binding to gtkdatabox located on:
>>> https://github.com/persan/A-gtkdatabox
>>
>> Does it use time stamps on the X axis (i.e. oscilloscope), or is it just a
>> XY plot?
> Its basicly an XY plot with float values.
I considered porting a C++ library that does the oscilloscope to Cairo Ada,
but I am not so sure now.
>> Is it based on cairo?
> Have not thought about it, i was looking a widget that made it possible
> to build a simple UI for a pc-oscilloscope.
Cairo looks a good choice. There are excellent Cairo Ada bindings by Damien
Carbonne.
Considering oscilloscope implementation there is a problem. I don't know
how to get a BitBlt under Gdk or Cairo, which would be essential for good
performance of scrolling.
>> How does it handle the case when several data points must be rendered in
>> one-pixel horizontal width?
> Have not looked deeper into the scaling effects when several X-values
> should be rendered one pixel wide, but it seems like a mean value is drawn.
I see, not a min-to-max vertical bar (the correct behavior).