View Full Version : Dealing with Overexposure
aurangzeb
2007 August 7th, 02:24
Hello Forum Members,
I know this issue has been looked at a few times before but I thought I should bring it up again as this will be a common problem faced when shooting outdoors in bright light and we should consider it carefully.
As I see it from reading the various postings on this and other forums, there are 3 ways for "controlling" overexposure on the HV20:
(1) Using the built in zebra stripes set at 70 or 100
(2) Purchasing the Canon ND filter FS-43U, and
(3) Constructing your own variable ND filter using two linear polarizing filters (this was demonstrated by another forum member who posted some short video footage to demonstrate the effect in bright light)
What I would like to know is how well this FS-43U ND filter performs in very bright light. I have not seen much about this on this or other forums. It would be nice if someone could upload a video to show the filters comparative effect in very bright light or when parts of the image are overexposed.
Also, it would be nice if someone could post some more footage showing the effect of the variable ND filter in different lighting conditions. The best test of all, in my opinion, would be if the same scene is shot using zebra stripes, then with the ND filter and finally with the "variable" ND filter which is adjusted to different levels. That way we could easily compare what solution works best.
Thanks,
Aurangzeb
Halsu
2007 August 7th, 06:35
The best test of all, in my opinion, would be if the same scene is shot using zebra stripes, then with the ND filter and finally with the "variable" ND filter which is adjusted to different levels. That way we could easily compare what solution works best.
You don't really need an ND filter to shoot in bright conditions - the camera will use higher shutter speed and smaller aperture to compensate so that your images will not be overexposed.
The benefit of ND filter, and especially the variable ND filter is that you can use more "normal" shutter speed and apertures - the difference is visible, but it depends on the viewer's preferences which is better.
Personally, i don't think i will shoot without the variable filter in daylight again, unless i absolutely have to.
BTW, There are different strengths of ND filters - in my opinion, my 4ND filter was too weak in bright sunlight, 8ND could have been enough.
BTW2, i pretty much always use 100% zebra to see what parts of the image are overexposed, also with the variable ND filter thingy.
aurangzeb
2007 August 7th, 07:42
Thanks for your input Halsu. Will do as suggested by you, namely, forget the FS-43 ND filter. Instead, I will set the zebra stripes to 100% and, if the utside light conditions require, use a variable ND filter like the one you constructed with the two polarizing filters.
Just out of curiosity, you used two linear polarizers. There was some debate about the type of polarizers to be used. Some people on the forums I visited are of the opinion that the polarizer attached to the HV20 body must be circular (not linear) while the outer one must be linear. Is this correct?
What make/specifications of filters would you recommend?
It would be nice if you could upload some more experimental footage made in very bright light with the variable ND filter.
Thanks again.
Halsu
2007 August 7th, 08:08
Just out of curiosity, you used two linear polarizers. There was some debate about the type of polarizers to be used. Some people on the forums I visited are of the opinion that the polarizer attached to the HV20 body must be circular (not linear) while the outer one must be linear. Is this correct?
What make/specifications of filters would you recommend?
No, the polarizer attached to the body can probably ALSO be circular - this hasn't been tested but should work.
Two linear filters work, this is what i have.
Mine are Hoyas, with 52 mm threads. To use these, i also have step-up rings from 43mm to 52 mm. I haven't tested other filters.
The reason i got 52 mm filters is simple - the shop didn't have 43 mm polarizers at the time. 43 mm filters might also show in the edges when using wide angles, as the filter stack is kinda steep.
It would be nice if you could upload some more experimental footage made in very bright light with the variable ND filter.
I'll try to come up with something, as soon as i have a bit of extra time (busy times right now).
Here's the link to the thread with my old test footage:
http://hv20.com/showthread.php?t=1132
mik
2007 August 7th, 08:17
the polarizing filter as an nd filter isn't a good idea. i've heard about it but i'll never use it. for digital cameras is recommended to use circular polarizers since a linear one could affect the focus. some work depending on the focus type but if you can't test always use circular.
a true variable nd filter is expensive, i think $300 or so. it'll be best to have an nd4 and nd8 and that's all.
Goose
2007 August 7th, 08:48
Hello Forum Members,
I know this issue has been looked at a few times before but I thought I should bring it up again as this will be a common problem faced when shooting outdoors in bright light and we should consider it carefully.
As I see it from reading the various postings on this and other forums, there are 3 ways for "controlling" overexposure on the HV20:
(1) Using the built in zebra stripes set at 70 or 100
(2) Purchasing the Canon ND filter FS-43U, and
(3) Constructing your own variable ND filter using two linear polarizing filters (this was demonstrated by another forum member who posted some short video footage to demonstrate the effect in bright light)
What I would like to know is how well this FS-43U ND filter performs in very bright light. I have not seen much about this on this or other forums. It would be nice if someone could upload a video to show the filters comparative effect in very bright light or when parts of the image are overexposed.
Also, it would be nice if someone could post some more footage showing the effect of the variable ND filter in different lighting conditions. The best test of all, in my opinion, would be if the same scene is shot using zebra stripes, then with the ND filter and finally with the "variable" ND filter which is adjusted to different levels. That way we could easily compare what solution works best.
Thanks,
Aurangzeb
There is a fourth way - I know I have said this before but shooting in cine will always prevent blow highlights. It also provies the most latitude for cc in post and is unhsarpened to allow sharpening without extra artifacts in post. A polariser will actually increase contrast so you should be careful if using that to prevent blown highlights
Halsu
2007 August 7th, 10:01
the polarizing filter as an nd filter isn't a good idea. i've heard about it but i'll never use it. for digital cameras is recommended to use circular polarizers since a linear one could affect the focus. some work depending on the focus type but if you can't test always use circular.
Could you tell me WHY isn't it a good idea? It works perfectly.
The focusing problem only affects SLR still cameras with beam-splitting focus / exposure metering. There's no problems whatsoever with HV20, or other digital video cameras i know of.
a true variable nd filter is expensive, i think $300 or so. it'll be best to have an nd4 and nd8 and that's all.
I'm pretty sure those use polarizing filters too. Probably two circular polarizers, with the one in front mounted "the wrong way" so that you don't actually get any polarizing effect, only the loss of light.
Halsu
2007 August 7th, 10:05
There is a fourth way - I know I have said this before but shooting in cine will always prevent blow highlights. It also provies the most latitude for cc in post and is unhsarpened to allow sharpening without extra artifacts in post. A polariser will actually increase contrast so you should be careful if using that to prevent blown highlights
I use both the variable ND and cine mode (for the exact reasons you mentioned). I lock my exposure to 1/50 - f1.8 - 0dB gain and control the exposure with the polarizing filters only. Cine mode alone can't do this - you will get much higher shutter speeds / smaller f-stops in bright light unless you use ND of some sort.
I haven't noticed an increase in contrast with polarizers on regular subjects - they do help adding contrast to the skies, but that's a good thing.
XTXZZ
2007 August 7th, 10:21
There is a one negative side-effect, image does have more grain when zoom is used, comparing a situation without Hoya lenses.
Halsu
2007 August 7th, 10:28
There is a one negative side-effect, image does have more grain when zoom is used, comparing a situation without Hoya lenses.
This sounds like the camera is adding gain - you're darkening the image too much with the polarizers. There shouldn't be any difference on the amount of grain onless you add gain - hey, that rhymes ;-)
Here's a bit more info about the beam-splitting auto focus / metering problems on SLR cameras (they too state that the problem doesn't affect video cameras, BTW):
http://www.geocities.com/COKINFILTERSYSTEM/polarizer.htm
XTXZZ
2007 August 7th, 10:49
Might be so, havenŽt yet studiet thoroughly. IŽll test.
mik
2007 August 7th, 11:15
it does seem the 300$ is just 2 polarizing filters :))
but i see only linear don't alter the final result (colors and contrast), not 2 circular or combined. plus it's a selective filter.
http://www.dgrin.com/showthread.php?t=38397
http://www.largeformatphotography.info/forum/archive/index.php/t-16872.html
Halsu
2007 August 7th, 13:59
it does seem the 300$ is just 2 polarizing filters :))
but i see only linear don't alter the final result (colors and contrast), not 2 circular or combined. plus it's a selective filter.
With two linear filters, like in my system, there's also polarizing effect going on. This is kind of selective, depending on how the filters are rotated, one can reduce reflections and / or deepen the blue of the sky.
This is often a desired effect, but i guess not always.
With two circular polarizers, the first one inserted "the wrong way", there shouldn't be any selective polarizing effect, the filter should only darken the image, without affecting it otherwise.
My educated guess is that this is the way the $300 filter is built.
mik
2007 August 8th, 03:19
it could be.
i do understand how 2 linear filters let the light pass but i don't have a clear picture about how the light travels thru 2 circular, one put the other way. maybe we'll find a physics description, drawing anything.
Powered by vBulletin® Version 4.2.0 Copyright © 2013 vBulletin Solutions, Inc. All rights reserved.