Re: [CinCV] about cinelerra

On Mon, 30 Aug 2010 11:05:48 +0930
Geoff Russell <geoffrey.russell@...> wrote:

> Hi,
> 
> I'm trying to install cinelerra on an Intrepid ubuntu system using
> your instructions.
> 
> What is deb?
> 
> $ sudo deb http://akirad.cinelerra.org akirad-intrepid main

 deb http://akirad.cinelerra.org akirad-intrepid main

is the line you have to add to /etc/apt/sources.list

> sudo: deb: command not found
> 
> It isn't on my system and doesn't show up on synaptic.
> 
> Cheers and thanks,
> 
> Geoff Russell
> 
> 
> 

[CinCV] Re: about cinelerra (Christian Thaeter)

> I'm trying to install cinelerra on an Intrepid ubuntu system using
> your instructions.
> 
> What is deb?

You should also update your Ubuntu system. The current version is Lucid
Lynx, and in six weeks the next version will be released.

[CinCV] RGB-FLOAT format

Hi,

On Ubuntu 10.04 I cannot get RGB-FLOAT or RGBA-FLOAT formats to work.

The cinelerra version is 2.1CV Build date Wed Apr 7 16:05:07 UTC 2010.

The media frames get displayed in the timeline tracks, but the compositing 
window is blank (black). I loaded a project from 2 years ago (made using an 
earlier version of cinelerra) which uses RGBA-FLOAT, and that also gives a 
completely blank screen throughout.

Using RGB-8 bit or YUV-8 bit works fine (although there are some color 
differences between rendered output and what's displayed with YUV).

Thanks in advance for any pointers,

tonyhb

[CinCV] RGB-FLOAT format

Hi,

On Ubuntu 10.04 I cannot get RGB-FLOAT or RGBA-FLOAT formats to work.

The cinelerra version is 2.1CV Build date Wed Apr 7 16:05:07 UTC 2010.

The media frames get displayed in the timeline tracks, but the compositing 
window is blank (black). I loaded a project from 2 years ago (made using an 
earlier version of cinelerra) which uses RGBA-FLOAT, and that also gives a 
completely blank screen throughout.

Using RGB-8 bit or YUV-8 bit works fine (although there are some color 
differences between rendered output and what's displayed with YUV).

Thanks in advance for any pointers,

tonyhb

Re: [CinCV] RGB-FLOAT format

Re: [CinCV] RGB-FLOAT format

Re: [CinCV] RGB-FLOAT format

Hi Einar,

> Try x11-xv or x11 as video output to rule out OpenGL problems.

Yeah - I already tried this, including restarting cinelerra with these options 
changed. I also tried disabling Compiz.

> There is too few information ro diagnose your problem. Try to start
> cinelerra from terminal - you may see there messages
> helping to solve problem.

Maybe this is significant:

-- from terminal output --

mpeg3_read_prev_packet 100 result=0 title=0 [....many lines like this..]
mpeg3_read_prev_packet 100 result=0 title=0 tell=25030eb6 program_byte=25030eb6
Playback3D::print_error:
uniform sampler2D tex;
uniform float h_offset;
uniform float s_offset;
uniform float v_offset;
void main000()
{
	vec4 pixel = texture2D(tex, gl_TexCoord[0].st);
{
float r, g, b;
float h, s, v;
float min, max, delta;
float f, p, q, t;
r = pixel.r;
g = pixel.g;
b = pixel.b;
min = ((r < g) ? r : g) < b ? ((r < g) ? r : g) : b;
max = ((r > g) ? r : g) > b ? ((r > g) ? r : g) : b;
v = max;
delta = max - min;
if(max != 0.0 && delta != 0.0)
{
     s = delta / max;
	if(r == max)
     	h = (g - b) / delta;
	else
	if(g == max)
     	h = 2.0 + (b - r) / delta;
	else
     	h = 4.0 + (r - g) / delta;

	h *= 60.0;
	if(h < 0.0)
     	h += 360.0;

-- then continues with more source code, ending with --

pixel.r = r;
pixel.g = g;
pixel.b = b;
}
	gl_FragColor = pixel;
}

void main()
{
	main000();
}

Fragment info
-------------
0(51) : warning C7050: "b" might be used before being initialized
0(51) : warning C7050: "g" might be used before being initialized
0(51) : warning C7050: "r" might be used before being initialized

-- end of terminal output --

Lots of mention of rgb...

> Where did you get your cinelerra? Different sources may have different
> problems. Did you compile yourself or installed a package? What format
> is your media?

The version is 2.1.1-git20100325~ppa5~lucid. The binary was installed using 
apt-get. The OS is 32 bit Ubuntu 10.04.

Tony

Re: [CinCV] RGB-FLOAT format

Hi,

Just for my own (very pragmatic) curiosity: Is anyone here *using* any 
of the float formats for real? I would love to do so myself, since I 
usually find myself stacking several effects on the video tracks. But 
having tried quite a few precompiled Cinelerra versions over the last 
few years, this option always ended more or less with a black screen and 
nothing useful rendered.

So instead of asking what went wrong in this specific case, can anyone 
stand up and say "I'm working with it, this is what I did"?

    Eli

--

-- 
Web: http://www.billauer.co.il

Re: [CinCV] RGB-FLOAT format

Hi Tony

On Sun, Sep 5, 2010 at 9:57 PM, Tony Hardie-Bick
<tonyhb@...> wrote:
> Hi Einar,
>
>> Try x11-xv or x11 as video output to rule out OpenGL problems.
>
> Yeah - I already tried this, including restarting cinelerra with these
> options changed. I also tried disabling Compiz.
>
>> There is too few information ro diagnose your problem. Try to start
>> cinelerra from terminal - you may see there messages
>> helping to solve problem.
>
> Maybe this is significant:
>
> -- from terminal output --
>
> mpeg3_read_prev_packet 100 result=0 title=0 [....many lines like this..]
> mpeg3_read_prev_packet 100 result=0 title=0 tell=25030eb6

These are debug outputs - no harm

> program_byte=25030eb6
> Playback3D::print_error:
> uniform sampler2D tex;
> uniform float h_offset;
> uniform float s_offset;
> uniform float v_offset;
> void main000()
> {
>        vec4 pixel = texture2D(tex, gl_TexCoord[0].st);
> {
> float r, g, b;
> float h, s, v;
> float min, max, delta;
> float f, p, q, t;
> r = pixel.r;
> g = pixel.g;
> b = pixel.b;
> min = ((r < g) ? r : g) < b ? ((r < g) ? r : g) : b;
> max = ((r > g) ? r : g) > b ? ((r > g) ? r : g) : b;
> v = max;
> delta = max - min;
> if(max != 0.0 && delta != 0.0)
> {
>    s = delta / max;
>        if(r == max)
>        h = (g - b) / delta;
>        else
>        if(g == max)
>        h = 2.0 + (b - r) / delta;
>        else
>        h = 4.0 + (r - g) / delta;
>
>        h *= 60.0;
>        if(h < 0.0)
>        h += 360.0;
>
> -- then continues with more source code, ending with --
>
> pixel.r = r;
> pixel.g = g;
> pixel.b = b;
> }
>        gl_FragColor = pixel;
> }
>
> void main()
> {
>        main000();
> }
>
> Fragment info
> -------------
> 0(51) : warning C7050: "b" might be used before being initialized
> 0(51) : warning C7050: "g" might be used before being initialized
> 0(51) : warning C7050: "r" might be used before being initialized
>
> -- end of terminal output --
>

Seems that OpenGL has some problems with code loaded by Cinelerra.
Code dumped is from huesaturation
plugin. I can't help here - my current videocard has not enough
capabilities to test this.

May be it is possible to disable OpenGL warnings.

> Lots of mention of rgb...
>
>> Where did you get your cinelerra? Different sources may have different
>> problems. Did you compile yourself or installed a package? What format
>> is your media?
>
> The version is 2.1.1-git20100325~ppa5~lucid. The binary was installed using
> apt-get. The OS is 32 bit Ubuntu 10.04.
>
> Tony
>

Sorry, can't give more help.

Einar

Re: [CinCV] RGB-FLOAT format

Hi Eli,

When I had RGB-FLOAT working, I was doing a lot of colour distortion, and this 
setting gave a much finer level of detail under these conditions.

This is because when you stack a number of RGB-processing filters of limited 
precision, each one quantises its results to the colour precision, and this loss 
of precision, in the integer (especially 8-bit) case, becomes visible when a 
filter at the end of the chain tries to expand a difference between, say, a 
value of 100 and 101 to 100 and 110. In the integer case, clearly you get a 
sudden step from 100 to 110. In the floating point case, you get the full 
gradation returned, even though the final output may still be quantised to 
8-bits (per colour).

So I miss this, as I was doing a lot of colour distortion again.

Tony

On 05/09/10 21:13, Eli Billauer wrote:
> Hi,
>
> Just for my own (very pragmatic) curiosity: Is anyone here *using* any
> of the float formats for real? I would love to do so myself, since I
> usually find myself stacking several effects on the video tracks. But
> having tried quite a few precompiled Cinelerra versions over the last
> few years, this option always ended more or less with a black screen and
> nothing useful rendered.
>
> So instead of asking what went wrong in this specific case, can anyone
> stand up and say "I'm working with it, this is what I did"?
>
> Eli
>