Bug 87127

Summary:	jagged lines from fits to realtime data
Product:	[Applications] kst	Reporter:	Netterfield <netterfield>
Component:	general	Assignee:	kst
Status:	RESOLVED WORKSFORME
Severity:	normal
Priority:	NOR
Version:	1.x
Target Milestone:	---
Platform:	unspecified
OS:	Linux
Latest Commit:		Version Fixed In:
Sentry Crash Report:
Attachments:	What I see given Barth's scenario... Reading last 30 samples Reading last 200 samples

Description Netterfield 2004-08-13 16:21:09 UTC

Version:           0.99-devel (using KDE 3.2 BRANCH >= 20040204, Mandrake Linux Cooker i586 - Cooker)
Compiler:          gcc version 3.3.2 (Mandrake Linux 10.0 3.3.2-6mdk)
OS:                Linux (i686) release 2.6.3-7mdk

How to reproduce:
run kst/tests/dirfile_maker/dirfile_maker
In a different konsole:
[.../dirfile_maker]> kst -y cos -y ssine -n 30 dm.cur -m 1
fit a 5th order polynomial to both.
The fit to 'ssine' (which has 1 sample per frame is a nice smooth curve.)
The fit to cos, (which has 20 samples per frame) is jagged (the jaggedness is variable) but the fit to ssine is smooth.

If you change nframes for everything with <Change Data Sample Ranges> to 200, the fit to cos looks smooth, but the fit to ssine shows signs of jagedness.

W/ 300 frames, both look good.

If you, from the command line select a fixed (rather than real time) range then the fits always look smooth, apparently independent of starting frame.
(ie, [.../dirfile_maker]> kst -y cos -y ssine -n 30 dm.cur -m 1 -f 1000)

Comment 1 Andrew Walker 2004-08-13 18:29:07 UTC

Created attachment 7100 [details]
What I see given Barth's scenario...

Attached is what I'm seeing with Barth's scenario. Barth could you attach
something similar or indicate what you think the problem is.

Comment 2 Andrew Walker 2004-08-13 18:37:22 UTC

Oh dear. now I actually look at the screen capture I see what the problem is. But looking at the screen everything appears fine, all nice and smooth. I'm guessing this is a problem with the screen capture, but that Barth is seeing something similar just by eye. 

I'll ask Rick to check this out and see if he can reproduce the problem.

Comment 3 Rick Chern 2004-08-13 18:54:27 UTC

Created attachment 7101 [details]
Reading last 30 samples

Comment 4 Rick Chern 2004-08-13 18:54:48 UTC

Created attachment 7102 [details]
Reading last 200 samples

Comment 5 Netterfield 2004-08-13 19:22:42 UTC

Oh Oh - strangeness: 
on my desktop, I do not see the effect (apparently neither does anyone else).  
Only on my laptop.  I will investigate further and send a screen shot.

???

On August 13, 2004 12:54 pm, Rick Chern wrote:
> ------- You are receiving this mail because: -------
> You are the assignee for the bug, or are watching the assignee.
>
> http://bugs.kde.org/show_bug.cgi?id=87127
>
>
>
>
> ------- Additional Comments From rchern interchange ubc ca  2004-08-13
> 18:54 ------- Created an attachment (id=7102)
>  --> (http://bugs.kde.org/attachment.cgi?id=7102&action=view)
> Reading last 200 samples
> _______________________________________________
> Kst mailing list
> Kst@kde.org
> https://mail.kde.org/mailman/listinfo/kst

Comment 6 Netterfield 2004-08-14 02:15:19 UTC

I've finally gotten back to the lap-top... And I don't see it now.
I have changed nothing....  it must be some sort of quantum effect :-)

So we shouldn't hold up 0.99 for it.

Comment 7 Andrew Walker 2004-08-26 18:57:42 UTC

Could this have been caused by the threading problem recently fixed by Barth?
If so can this bug be safely withdrawn?

Comment 8 Netterfield 2004-08-26 19:15:36 UTC

Go ahead and remove with a 'works for me'.

On August 26, 2004 12:57 pm, Andrew Walker wrote:
> ------- You are receiving this mail because: -------
> You are the assignee for the bug, or are watching the assignee.
>
> http://bugs.kde.org/show_bug.cgi?id=87127
>
>
>
>
> ------- Additional Comments From arwalker sumusltd com  2004-08-26 18:57
> ------- Could this have been caused by the threading problem recently fixed
> by Barth? If so can this bug be safely withdrawn?
> _______________________________________________
> Kst mailing list
> Kst@kde.org
> https://mail.kde.org/mailman/listinfo/kst

Comment 9 Andrew Walker 2004-08-26 19:28:21 UTC

Unable to reproduce anywhere else.