Panasonic xx600\700 bondi blue (turquoise) bias issue - Page 7

Is that the kind of thing you need to see myg, or should I do it differently? Just give me the paramters and I will work it out.

All this ND filter affecting color balance in the "bondi blue issue" becomes mute when trying to explain why the WB2 trick works...
Just a quick refresher for those who might not know:
When performing a manual color white balance on a TM700, often in outdoor situations the result will have a "turquoise" or "greenish" offset, or bias, in blue colored areas like the sky or blue objects. And this regardless of iris opening or shutter speed. One trick to "remove" the bias, is to choose WB2 preset momentarily, and come back to manual WB (flashing WB icon).

So I think the white balance change with aperture variation (probably due to a tinted ND filter) is a separate problem from the "electronic" or "software" bondi blue or turquoise bug.

People also have to realize that this is a sub $1000 camcorder, and not a Pro model costing more than 10 times that, so Panasonic was bound to cut corners here and there to meet the low cost goal. Let's just hope that there might be a final firmware upgrade one day to alleviate some of the bugs. But I doubt it.

I would be curious to hear from owners and operators of pro models if there are any bugs at all in multi thousand dollars models?

BTW, what "workinghard" has observed with the ND filter is exactly what is happening on my TM700 too. It is affecting color balance, and it should not. A neutral density filter should be neutral and affect ALL the color spectrum evenly.

I know that the HMC40 is basically a TM300 on steroids but I don't think I've ever read a negative complaint about the colors.

The discussion lately in this thread underscores the utility of knowing in as much detail as possible how a given camcorder _actually_ works, in this case how the iris and the internal ND filter cooperate to produce a given "f-stop" (virtual). That's the kind of information I was trying to pull out in a largely-ignored thread I started a few days back.

[SOAPBOX]I believe that all knowledge is useful and that there's no such thing as learning "too much." It's only when you know the true limitations of a device (and there are always limitations) that you can figure out ways around them to meet your artistic & functional goals (or abandon the thing entirely if there really is no way around).[/SOAPBOX]

In this case it is the edge of the ND Filter-when fully inserted I dont have any problems.

and they probably have a higher quality filter, that must be cut better, lol...oh well.

I will play around with the Indoor2/manual later, I have some theories...

[quote=Paulo Teixeira;20059573I'm not going to really rule out the fact that maybe your TM700 would do a better job but I'm also not going to rule out the fact that it would have done the same when shooting completely in factory set auto as well.

[/QUOTE]

Paulo, a couple of points here. I don't really know that my 'late' 700 would have done any better, although I suspect it would have based on the colors I'm seeing in your clips. But I've seen plenty of posts of recent clips from the 700 that also has a greenish hue to the sky and/or water as yours did. So I don't believe this is a '600 problem', but rather a 600/700 series issue (I'm not sure about the 900 as there simply aren't enough clips out there to tell).

Again, your first two clips also didn't look right IMO and you had the camera running long enough that it should have corrected. Those drums in the center should have had orange striping and the fact that they didn't probably points to issues with other colors too.

What I am saying is that I believe some of the 600/700s out there seem to be setup differently, whether that's on the hardware or software side. I don't believe this is a '600 issue' per se. As far as evaluating clips on Vimeo, I've seen many clips from the 700 (including mine) that display colors much like I see when I put my 700 clips on my ISF'd Kuro. So I truly don't believe this is an issue of how Vimeo (or UTube) is altering color. Resolution yes, color no.

I do agree 100% with Turbe regarding the display you're viewing on and I've mentioned that in the past. But there is simply something not right about some of the units out there that are resulting in this greenish cast. Any camcorder when set to full auto should be able to display a blue sky as blue. I've owned many many camcorders in all price ranges and I've never seen any behave quite like this. Something is simply not right and I don't think we can explain it away by blaming the 'auto' feature. If the camera can't present normal looking skies when set to auto, then there is something not right.

And that's a 'bingo'. Regardless of the calibration of the monitor, there is some serious color shifting going on here at times. Yes, all AWB setups will display some variation as you zoom in on solid color hues or the like, but I've never had one that alters sky colors so radically by simply panning.

In fact, I forgot to mention that I was in a BB last week and was playing with a Panasonic 750 (which I was surprised to see). I had my own SD card and was comparing the colors with an adjacent Canon SF20. When I got home I played the SD card on my plasma via a PS3. When the 750 clip first started, it was decidedly pinkish/reddish. This only changed when I zoomed in on a BB pricing tag and the camera grabbed the white on that tag and adjusted. Once I pulled back from that zoom, the color was decidedly different and far more natural.

Now when I did exactly the same thing with the Canon, the color remained normal from the beginning to the end. I zoomed in on the same price tag and the color didn't change. I should note that both cameras were turned on from a 'cold start' and I waited a bit as I made sure both cams were adjusted properly. So you can't blame it on the fact that the cameras were turned on and shooting began immediately, it didn't.

Something is wrong!

well, it is too late for tonight. But, in regards to the ND filter issue, you can probably reproduce that indoors on a white piece of paper, if you can get it bright enough. Might be hard to get enough light though.


Not sure what the situation is with wb/indoor 2....

Honestly, if it were just the blue coloring I wouldnt be bothered, but it is the green hue clouds that is really bothering me....though I am imagining it as worse than it really is now. I think after this weekend I will have to cool it on this issue or I may start imagining boogeymen in my camcorder. I mean, its not like Panasonic cares about ME I dont think 300 dollar camcorders in a country without a real service center get top priority, and when I go to Japan next, no way they are going to care about it either Oh well!

To MYG-I dont think I am going to be able to do all that testing. I will be looking back, and if there is a really dull moment during the weekend I will jump on it, but my wife is getting irritated with me now...



Well, I am editing some footage right now...which on my computer is a pretty big deal (lots of dead time), and I recalled this statement.

On my test runs, I found that the fstop will not automatically go below 8.0.

I also found that on a normal 5pm afternoon, partially cloudy, 1/180th was too bright to be really usable at f8.0.

Were you using an additional ND filter? (Not that it really matters much at this point, but I was a little confused when I remembered you saying that)

I quoted the above from the ongoing g10 tm900 comparison thread, but it doesnt really belong there.

I did as you asked, and pulled some color samples from the video I shot. Sorry for taking so long to start this process, I have no experience and my computer is in Japanese...thus, to figure out how to even pull the color sample I had to search google, and then I had to search for images of the command prompts in MS Paint...a real pita.

While this is only one video, and only adjusting the shutter with many other variables present that could affect the outcome, it is not meant to be a final conclusion at all, rather a continuation of the discussion.

Further testing would need to be done, which I now know how to do...but I need to focus on finding a new job right now (crap that terrifies me, though to be honest I really want out of teaching! lol)

but, I pulled the still pics taken from the video in manual with all but shutter in auto mode, 1/350 shutter, 1/750th shutter, and 1/1000th shutter. I completely understand that I need more controlled conditions to say that there is a "problem," specifically manual wb as discussed earlier in the thread. I have done that, but do not have all the pics and video set up. But these are pics pulled from the exact same video and section you said the rgb values were constant in the latter portion of the clip, but I had observed some change in.

I tested in as close to the same location as I could, the same spot in the middle of the sky in each, the same region of the cloud formation, and above the exact same limb on the tree for each. I only waited about 5-8 seconds between shutter settings, and there was no major change in overall cloud cover or shade. The following are the RGB values as demonstrated in MS Paint:

Test 1 1/350th shutter
Mid-image, 57/107/134
Cloud, 81/116/138
Tree line, 53/102/132


test 2 1/750th shutter
Mid-image, 61/97/129
cloud, 82/112/138
tree, 61/100/131

Test 3 1/1000shutter
Mid-image 65/100/132
Cloud, 93/115/138
tree line, 65/100/130


What we see is that the green and blue are fairly consistent, except for the green decreasing from 1/350th and red increasing from 1/350th (which seems to relate to my prior observations of the sky looking less green at 1/750th and particularly at 1/1000th). Which is directly related to the movement of the ND filter via fstop.

I have no idea how to understand this though, I can only judge by what I see with my eyes, and try to interpret this with a total lack of background experience.

Eventually I will pull the images I have where I can observe what appears to me to be the edge of the filter passing over the screen as the iris changes to test in the same way, but until then...oh well Not really a huge deal in my opinion, just something to check out.

I want to show you my videos on youtube which demonstrate
white balance problems of Panasonic xx700 series.

As a new user, I am not alowed to post urls.

The following errors occurred with your submission:
You are only allowed to post URLs once you have at least 3 posts. If you are a spammer, please go elsewhere. If not, sorry for this issue the spammers have made for.


OK, so this is my first post

2rm

This is my second post
2rm

Third

Please, see the following videos which clearly demonstrate
white balance problems of Panasonic xx700 series.
Perhaps the most interesting is the demonstrated connection
with built-in ND filter in Part 3. Briefly said, there is no bondi blue
hue of sky if one shoots with aperture values lower than f/2.8,
i.e. without built-in ND filter. For more details see descriptions
attached to videos on youtube.

-----------------------------------------------------
Panasonic HDC-TM700 White Balance Problem - Part 1

http://www.youtube.com/watch?v=EZtP9kv_OKM

A slowed-down demonstration of TM700's inability to hold
manual WB setting in daylight. Approx. one second after manually
adjusting WB, it changes. In this demonstration, the effect is visible
as a shift of clear sky hue from blue to bondi blue. In order the effect
to be clearly visible, manual WB was set using clouds. Of course, the same
happens if one uses white sheet of paper or grey card. Thus, it is impossible
to manually set correct WB in daylight.
-----------------------------------------------------
Panasonic HDC-TM700 White Balance Problem - Part 2

http://www.youtube.com/watch?v=0zKXB5OfkIU

This video is a short and detailed demonstration of TM700's inability
to hold manual WB setting in daylight. Shutter speed and aperture values
are shown for each shot.
------------------------------------------------------
Panasonic HDC-TM700 White Balance Problem - Part 3

http://www.youtube.com/watch?v=g20oB6faYvY

This video shows that TM700's WB problems are somehow related to built-in
ND filter. Looking into camcorder lens one can see, that iris closes from
OPEN (f/1.5) to f/2.8. From aperture f/2.8 to f/5.6 iris does not move,
but built-in ND filter shows itself at 2.8, being fully inserting at f/5.6.
From aperture f/5.6 iris closes again, stopping at f/8.0 in automatic iris mode.
It means, that all aperture values from f/2.8 up use built-in ND filter.

This video demonstrates that all applicable WB settings in bright sunlight
(AWB, WB preset Sun, manual WB) work well without the built-in ND filter,
i.e. with apertures less then f/2.8, and shows blue sky as blue. However,
from f/2.8 to f/8.0, all of them show bondi blue cast on clear sky. The effect
is particularly strong from f/5.6. When shutter speed is changed so that iris
opens to value f/2.4 or less, bondi blue cast is not seen again.

I encourage owners of 700 and 900 series to check this issues with their own
units. Panasonic should provide us with firmware update to fix this problem.

Regards
2rm

I do appreciate you taking the time to do this. However, I can not help but wonder if you have bothered to read any of this thread at all?

Maybe take a look at the last few pages at least?

Now...I am confused as to how these filters work.

Anyway, my description of how the ND filter works in this thread was slightly off I think. And you are probably correct, it must moving in place at 2.8-5.6

Also, my bondi-blue hues pretty much disappear after 5.6, making me think that on my particular filter it is primarily the outer curve altering light absorption. I noticed all of this earlier and some focused on part of my video that was overexposed when I erred and only opened the iris. However, when I kept the exposure I pulled RGB values, and they clearly show the bondi-blue bias during the transition of ND filter placement.
================-

Also, some people have had other problems. There have been some shots of manual wb shifting, which this may explain, but there are other instances, as you show in your video of blue suddenly having a totally different color by changing to indoor2 and back to manual set. i have done this and never seen a change, but others have, and the difference is dramatic. I have never seen blues so off though. However, this change does not affect the shutter/iris combo, so how could it be connected to the ND Filter?

Well, I actually watched all your videos now, and they are excellent

I am not convinced that setting mwb on the clouds like that works really well, but for the purpose of this test it should not matter. You did a FANTASTIC job of capturing my own observations thus far, with a few differences in our camcorders aside.

I have read it all and other threads as well. But I was confused and had to do my own tests.



According to my experiments, there are two types of bondi blue:
1. bondi blue appears with correct WB and wrong exposure - it can be removed by adjusting iris only (it seems that ND playes no role here)
2. bondi blue caused by WB shift when shooting with apertures which use buit-in ND filter, i.e. from f/2.8 up



What I have seen is that change in blue after changing WB from manual to Indoor 2 and back occurs in daylight only. In artificial light, manual WB works perfectly well.


Before switching to Indoor2 and back WB depends on shutter/iris combination and is shifted when ND filter is inserted. After performing the Indoor 2 trick WB is correct and does not depend on ND as it should.


Thanks


Are you sure your camcorder works differently? Our only hope for correcting this issue by firmware upgrade from Panasonic is that all camcorders are faulty in the same way.

Regards
2rm

I'm just looking to send the URL to a clip of a TM700
" You are only allowed to post URLs once you have at least 3 posts "
One shot
Bertrand

I'm just looking to send the URL to a clip of a TM700
" You are only allowed to post URLs once you have at least 3 posts "
Second One
Bertrand

I'm just looking to send the URL to a clip of a TM700
" You are only allowed to post URLs once you have at least 3 posts "
Next, I hope the following will be the right

Bertrand

Do you get the same aberrations with or without Sun?

Because infrared rays and the ND filter do not always mix.

A sample taken in the morning very early in ISLAND in fully automatic mode

Download here :

>>> http://www.yousendit.com/transfer.ph...Mm1OMUN4dnc9PQ

( TM700)
Bertrand

Beautiful shot. Colors look natural.

Please do not discuss CA here. This thread is about bondi blue.

Regards
2rm

[quote=workinghard;20209118]I do appreciate you taking the time to do this. However, I can not help but wonder if you have bothered to read any of this thread at all?

Maybe take a look at the last few pages at least?

Now...I am confused as to how these filters work.

Anyway, my description of how the ND filter works in this thread was slightly off I think. And you are probably correct, it must moving in place at 2.8-5.6

Also, my bondi-blue hues pretty much disappear after 5.6, making me think that on my particular filter it is primarily the outer curve altering light absorption. I noticed all of this earlier and some focused on part of my video that was overexposed when I erred and only opened the iris. However, when I kept the exposure I pulled RGB values, and they clearly show the bondi-blue bias during the transition of ND filter placement.
=--------------------

According to this

http://www.ianperegian.com/My_FZ35_3...D_Filters.html

The so called sweet point where nd is activated is between f3.4 and f/6.8

The benefits of ND filters are that they enable the lens to operate around its sweet spot over a much wider range of light intensities, and the graded nature of the filters enable smoother changes in DOF (depth of field) and exposure under changing lighting conditions.

That information I gave on my webpage about the ND filter/s on the TM700 was obtained by Dave Jarvis, who posted it on the DVDoctor forum at:
http://forums.dvdoctor.net/showthrea...t=52758&page=3

His results for the way the ND filter/s move look to be a bit different to the results obtained by 2rm that are shown in his post above.

Cheers,
Ian