Topic: Sinar eMotion Brumbaer Question (Read 8390 times)

Murray Fredericks · « **on:** November 04, 2007, 03:32:31 am »

Hi,

just finished a major shoot with an Alpa/ eMotion 75.. and am having trouble applying the white file corrections using Brumbaer.

The particular lens is the 24mm APO-Digitar with Centre filter.

I made White Shading Files at the start and end of the shoot for the lens and apeture. Brumbaer makes the correction file but does not seem to apply it properly to the files.

The luminosity is good to the edges but i am left with large concentric rings of green and magenta.

Files close to where the WSF file was made, seem to adjust reasonably well. But I cannot get the correction to apply to any files away (chronologically) from the WSF. I have tried to individually load the WSF then process an image, also I have tried to re-name the WSF to before all others taken on this shoot.

Any ideas?

Murray

thsinar · « **Reply #1 on:** November 04, 2007, 06:23:05 am »

Hi Murray,

Am not sure to understand what you are doing excatly: do you mean that you're using the centerfilter?

Best regards,
Thierry

Quote

Hi,

just finished a major shoot with an Alpa/ eMotion 75.. and am having trouble applying the white file corrections using Brumbaer.

The particular lens is the 24mm APO-Digitar with Centre filter.

I made White Shading Files at the start and end of the shoot for the lens and apeture. Brumbaer makes the correction file but does not seem to apply it properly to the files.

The luminosity is good to the edges but i am left with large concentric rings of green and magenta.

Files close to where the WSF file was made, seem to adjust reasonably well. But I cannot get the correction to apply to any files away (chronologically) from the WSF. I have tried to individually load the WSF then process an image, also I have tried to re-name the WSF to before all others taken on this shoot.

Any ideas?

Murray
[a href=\"index.php?act=findpost&pid=150491\"][{POST_SNAPBACK}][/a]

rainer_v · « **Reply #2 on:** November 04, 2007, 06:41:54 am »

i dont get it also what you describe.
what means:
"I made White Shading Files at the start and end of the shoot" did you say that you took two reference shots?

"Files close to where the WSF file was made, seem to adjust reasonably well. But I cannot get the correction to apply to any files away (chronologically) from the WSF."
i usually take one whitefile before every new shot, if i work withj shiftlenses.
seems so you did not do this or do i missunderstand you?

"also I have tried to re-name the WSF to before all others taken on this shoot."
so you made one wf?

Murray Fredericks · « **Reply #3 on:** November 04, 2007, 07:10:00 am »

Hi Thierry & Rainer,

Thierry:
I shot the images with the correct centre filter on the lens

Rainer:
The whole shoot was done on the 24mm at f/11. I made a White Shading File at the start of the shoot and another after the shoot at f/11.

I DID NOT make a White Shading File after each shot. I did not use movements therefore i asumed it was OK to make a 'generic' WSF at f/11 and apply to all the images.

Maybe this was a mistake?

The job was under 3 kinds of fluoros mixed...

To clarify: The Images close to when the first White Shading File was made are OK. The others are not.

Interestingly - the vignetting is not evident, but the concentric colour rings of magenta and green are.

Murray

Murray Fredericks · « **Reply #4 on:** November 04, 2007, 07:30:54 am »

Hi Again,

I think i have finally discovered the problem...after 12 hours of testing, processing, re-shooting wsf's etc

I shot a macbeth colour chart test under the fluoros and calibrated the test in the top section of the preferences pane in Brumbaer.

I have just re-processed the images with D55 setting in the top of the pane and 'ignore' in the 2nd section, using a previous chart calibration file done in sunlight.

It seems to have improved the problem greatly - it's still not perfect but I've now got some images I can give to the client.

In future i think I should make a WSF after each shot.

Murray

rainer_v · « **Reply #5 on:** November 04, 2007, 07:37:14 am »

Quote

Hi Thierry & Rainer,

Thierry:
I shot the images with the correct centre filter on the lens

Rainer:
The whole shoot was done on the 24mm at f/11. I made a White Shading File at the start of the shoot and another after the shoot at f/11.

I DID NOT make a White Shading File after each shot. I did not use movements therefore i asumed it was OK to make a 'generic' WSF at f/11 and apply to all the images.

Maybe this was a mistake?

The job was under 3 kinds of fluoros mixed...

To clarify: The Images close to when the first White Shading File was made are OK. The others are not.

Interestingly - the vignetting is not evident, but the concentric colour rings of magenta and green are.

Murray
[a href=\"index.php?act=findpost&pid=150506\"][{POST_SNAPBACK}][/a]

i am surprised. maybe it has something to do with the center filter. i never use one and let do brumbaer the job, although in the meantime i often make two conversions, one with vignetting on and one with vignetting off, layer them in photoshop and can blend than perfect and in realtime the amount of vignetting correction. ( you have to make two rxx files from one white shot, one with and one without the vignetting flag set in brumbaer ).
in general you dont have to shoot a white file at the end of the shot,- but you are right, if you apply it manually to the file it should work also.
maybe its the light sources itself ? i was not sure in your image if the concentric rings are caused by the optic/sensor or by the lightsource.
which back you use?

in any case the white shot has to be taken BEFORE each shot not after, if you want to apply it automatically. in general the e75 should behave perfect under fluorescent light.

Murray Fredericks · « **Reply #6 on:** November 04, 2007, 07:43:16 am »

Rainer,

I am surprised too...

The back is emotion 75.

The wsf was made manually, loaded manually, then all the images (ia's) were dropped into Brumbaer at once for a batch convert.

As I said, the results are much better now with D55 set in the top of the prefernces pane.

Aside: why are there two panes there anyway?? Is one meant to be the colour temp of the shooting conditions and one the temp the file srae to be mapped to?

The concentric rings are on every shot (about 50 ) in all areas of varying light...It does look like the shape of the centre filter...

Murray

thsinar · « **Reply #7 on:** November 04, 2007, 07:45:06 am »

Dear Murray,

I am pretty much sure that it is the center filter. When I was with David K. last June we did some shots on a 24mm Digitar, and DID NOT use any center filter, but corrected this with the previously shot WSF. The images were perfect.

Best regards,
Thierry

Quote

Hi Again,

I think i have finally discovered the problem...after 12 hours of testing, processing, re-shooting wsf's etc

I shot a macbeth colour chart test under the fluoros and calibrated the test in the top section of the preferences pane in Brumbaer.

I have just re-processed the images with D55 setting in the top of the pane and 'ignore' in the 2nd section, using a previous chart calibration file done in sunlight.

It seems to have improved the problem greatly - it's still not perfect but I've now got some images I can give to the client.

In future i think I should make a WSF after each shot.

Murray
[a href=\"index.php?act=findpost&pid=150511\"][{POST_SNAPBACK}][/a]

rainer_v · « **Reply #8 on:** November 04, 2007, 07:54:13 am »

i have no experience with cf filters and brumbaer, it could be that the file was influenced by this.
how looks the white file itself if you expose its own dng, applying it to itself?

the white setting should not affect color shifts, but the interpretation of the colors through ACR. which version of brumbaer you have?
it should be 1.44- which is on brumbaers website.
we are now by 1.44ß4. you can ask stefan that he send this to you,
although just details are different to 1.44 regarding the way the files are numbered if two white files are made.

Murray Fredericks · « **Reply #9 on:** November 04, 2007, 07:59:20 am »

Rainer,

version is 1.44

When the WSF is used to interpret the ia from which it came, it is perfect...

I am going to test the lens without the centre filter.

Once again though, I am amazed at how much better the rings are when D55 is used vs when my custom fluoro is used. i can't see how this would affect the rings but??

Maybe you caould ask Stephan for some more documentation in the help files about how the two colour matrix panes work and how to make the most of them...

I realise this may not be related to this discussion.

Cheers

Murray

thsinar · « **Reply #10 on:** November 04, 2007, 08:06:12 am »

Murray,

the 2 panes are there to make 2 different color matrices, one for about 5500 K and the other around 2800. Here what Stefan writes in his "Help" file (go to "Help" in"Emotion DNG Converter"):

Thierry

----------------------------------------------------------------------------------------------------------------
"The color rendition will be the better, the better the matrices are and if the color temperatures of both matrices are far from each other (i.e. 2800K and 5500K).
Color matrix:

You might want to create a custom Color Matrix for the default or a custom White Refernce file. The procedure is identical, just make sure the correct White Refernce is selected.

Take a shot of a ColorChecker 24 or ColorChecker SG chart. The shot should be evenly lit. The light should have the temperature for which the Color Matrix shall be valid.

If you can create two matrixes, make one shot about 5500K and one about 2800K. If you can only make one shot take it at 5000K if possible.

Drag the IA file onto the matrix or open it using the "Calib." button. You can not drag an image directly from eMotionReader onto the matrix, you do have to copy it and it's BR file onto a harddisk, CD or memory card first.

You will be prompted for the type of refernce chart used. Gretags ColorChecker 24 and ColorChecker SG are supported. At the moment there is no advantage to using the much more expensive SG chart.

After some seconds you will see an off color image of the shot. Move the red squares into the corner fields of the chart. Make sure that all other squares will be placed inside chart fields as well. The number of rows and columns will change depending on the relative position of the red squares.

Once finished press the "Proceed" button.

The window will close and the matrix will be set to the new values. Don't forget to set the color temperature of the matrix.

A simple test to see how good the matrix is is to convert the shot of the chart and check what the dng looks like.
----------------------------------------------------------------------------------------------------------------

Quote

Aside: why are there two panes there anyway?? Is one meant to be the colour temp of the shooting conditions and one the temp the file srae to be mapped to?

Murray
[a href=\"index.php?act=findpost&pid=150514\"][{POST_SNAPBACK}][/a]

thsinar · « **Reply #11 on:** November 04, 2007, 08:11:20 am »

Murray,

another information from Stefan's "Help" about "Color Matrices" and there effect:

"In the last step eMotionDng will add the Color Matrixes and their specific color temperature and the color temperature of the shot to the dng file.

Using the color temperature of matrices and shot, ACR will interpolate a conversion matrix from the color matrixes. The color rendition in ACR will be the better, the better the matrices are and if the color temperatures of both matrices are far from each other (i.e. 2800K and 5500K).

If you specify only one matrix (by setting the second color temperature to "ignore") the single matrix will always be treated as valid for 5000K regardless of the color temperature setting. So make sure that the matrix is valid for 5000K, if you plan to use a single matrix only.

If you only have a single matrix and this is not valid for 5000K, set the color temperature for the first matrx to the temperature the matrix is valid for and set the tempeartue of the second matrix to "auto". This will generate a second matrix automatically. This is not as good as having a "real" second matrix, but better than having a single matrix for a teperature different from 5000K.

Whenever you use only one matrix, use the first matrix and set it's color temperature accordingly. Than set the color temperatue of the second matrix to "Ignore" or "Auto"
--------------------------------------------------------------------------------------------------------------

So I guess the reason(s) of your problems are BOTH the use of center filter AND the wrong color matrix (not 5500) and "Ignore" set o the second panel.

Best regards,
Thierry

Quote

Rainer,

version is 1.44

When the WSF is used to interpret the ia from which it came, it is perfect...

I am going to test the lens without the centre filter.

Once again though, I am amazed at how much better the rings are when D55 is used vs when my custom fluoro is used. i can't see how this would affect the rings but??

Maybe you caould ask Stephan for some more documentation in the help files about how the two colour matrix panes work and how to make the most of them...

I realise this may not be related to this discussion.

Cheers

Murray
[a href=\"index.php?act=findpost&pid=150517\"][{POST_SNAPBACK}][/a]

Murray Fredericks · « **Reply #12 on:** November 04, 2007, 08:13:31 am »

Hi Thierry,

thanks - i saw that in the help menu, but assumed that to make two (2) references one at 5000k and one around tungsten, would be for tungsten and/or daylight shooting.

What about when the light is from fluoros or other kinds of daylight...?

There are all the options in the drop down menus in those panes. Is it right to assume that for accuracy you would test in the light conditions that you are shooting in?

If not, then what are all those other lighting conditions in the drop down panes there for?

Murray

rainer_v · « **Reply #13 on:** November 04, 2007, 08:14:16 am »

Quote

Rainer,

version is 1.44

When the WSF is used to interpret the ia from which it came, it is perfect...

I am going to test the lens without the centre filter.

Once again though, I am amazed at how much better the rings are when D55 is used vs when my custom fluoro is used. i can't see how this would affect the rings but??

Maybe you caould ask Stephan for some more documentation in the help files about how the two colour matrix panes work and how to make the most of them...

I realise this may not be related to this discussion.

Cheers

Murray
[a href=\"index.php?act=findpost&pid=150517\"][{POST_SNAPBACK}][/a]

it does not affect the rings, but it may affect the magenta/ green amount in the image interpretation and so they seem to be better, in fact it has nothing to do with the color rings,- but probably with the appearance they have.

Murray Fredericks · « **Reply #14 on:** November 04, 2007, 08:19:30 am »

Theirry,

thanks again.

complex now...

Can it be simplified by example?

If I shoot under strange fluoros, I do a test chart.

Do I load it in the 1st or 2nd pane? In each case, what should the other case be. Should I load my chart for 5000k (which is good) in the 'other pane' - which way around does it go?

TRhis is where I am stuck, or my powers of comprehension with regard to help menus evaporates..

Murray

Murray Fredericks · « **Reply #15 on:** November 04, 2007, 08:20:24 am »

Theirry,

thanks again.

complex now...

Can it be simplified by example?

If I shoot under strange fluoros, I do a test chart.

Do I load it in the 1st or 2nd pane? In each case, what should the other case be. Should I load my chart for 5000k (which is good) in the 'other pane' - which way around does it go?

TRhis is where I am stuck, or my powers of comprehension with regard to help menus evaporates..

Murray

thsinar · « **Reply #16 on:** November 04, 2007, 08:40:22 am »

Murray,

No, the way to go is to shoot a chart UNDER the light conditions you are using for your actual shot. That is certainly wrongly explained in Stefan's Help. In any case, one matrix under the light conditions and the second panel to "Ignore" has always given me the best results.

Thierry

Quote

Theirry,

thanks again.

complex now...

Can it be simplified by example?

If I shoot under strange fluoros, I do a test chart.

Do I load it in the 1st or 2nd pane? In each case, what should the other case be. Should I load my chart for 5000k (which is good) in the 'other pane' - which way around does it go?

TRhis is where I am stuck, or my powers of comprehension with regard to help menus evaporates..

Murray
[a href=\"index.php?act=findpost&pid=150523\"][{POST_SNAPBACK}][/a]

rainer_v · « **Reply #17 on:** November 04, 2007, 08:52:09 am »

in any case i have just made this in case very critical color reproduction is needed. so i did in museum exhibition shots with very colorfull samples in the exhibition. results have been great.
normally i use two color kalibrations,- but i use only one at a time, second color tab i set to ignore.
one shot in direct afternoon sunlight, corresponding to 5500 kelvin, one done in overcast weather corresponding to 6500K. the sunlight shot has more intensive saturated reds and greens, the overcasted shot is more "decent". depending on my daily mood i use the one or the other .....

BJNY · « **Reply #18 on:** November 04, 2007, 09:04:01 am »

I don't know the Brumbaer software, but does it work differently than Phase One's Capture One where a white plexiglas is used at 2 stops more exposure.

thsinar · « **Reply #19 on:** November 04, 2007, 09:18:01 am »

hi Billy,

we are speaking about 2 different things here: the Brumbaer " eMotion DNG Converter" is primarely a DNG converter, as its name indicates, for tethered and untethered conversions. In additon there are some features like "color balance", "destitch" (to eliminate centerfold issues), etc ... AND a feature called "White Reference". This last one allows to correct lens and sensor casts. In this the process is the same: one makes use of a plexiglas. From here on it differs: Brumbaer eMotion DNG Converter allows to apply the correct "white Reference" to the exact image in a batch process, on as many files as wanted. That's what makes the workflow so efficient and fast.

Best regards,
Thierry

Quote

I don't know the Brumbaer software, but does it work differently than Phase One's Capture One where a white plexiglas is used at 2 stops more exposure.
[a href=\"index.php?act=findpost&pid=150533\"][{POST_SNAPBACK}][/a]

Author Topic: Sinar eMotion Brumbaer Question (Read 8390 times)