Photo Forum / General Photo Topics / Australian Photography / February 2007
Something different - using a *DARK* ND filter
|
|
Thread rating:  |
mark.thomas.7@gmail.com - 23 Feb 2007 11:29 GMT Just wondering if any other crazy people have tried using an X400 (yup!) ND filter. It gives around 9 stops exposure decrease, for about AU$88... (dang)
Basically I want try some odd stuff like shooting a city street in daytime for 5 seconds or so, to remove a lot of the people/cars - you know, that old stuff! I have some other interesting ideas for it...
So rather than stack a few 8x (I don't like more than one extra layer of glass in front of my lenses..), I made a few enquiries to see if you can get a really dark one, and the X400 came to my notice.
Anyone else used one, any insights before I have a go?
(crossposted to aus.photo and rec.photo.digital, hope you don't mind)
Jeff R. - 23 Feb 2007 11:53 GMT > Just wondering if any other crazy people have tried using an X400 > (yup!) ND filter. It gives around 9 stops exposure decrease, for [quoted text clipped - 9 lines] > > Anyone else used one, any insights before I have a go? Well, the two most obvious ideas would be: a welding mask filter or a solar filter. Both are available in different "grades" (densities) and both are unfortunately heavily coloured - usually green for the welding glass and lots of different colours for solar filters. Mine show either a "gold" or a "silver" coloured sun.
A welding glass (still fine for monochrome) you can get for a few bucks from Bunnings. Solar filters cost more - even cheap ones. I bought a 12"x12" sheet of mylar filter material from Gary Hand Optics (USA) for not-too-much, but I confess it never occurred to me to try a "conventional" daylight shot with it. I got it for the last transit of Mercury (bloody clouds).
ISTR the local astronomy shops ask about $30-60 for a solar filter (glass), but these are quite clean optically.
Thanks for the idea. I'll give it a go too.
-- Jeff R.
Paul - 24 Feb 2007 11:34 GMT Baader Solar film shows a white sun.
>> Just wondering if any other crazy people have tried using an X400 >> (yup!) ND filter. It gives around 9 stops exposure decrease, for [quoted text clipped - 29 lines] > -- > Jeff R. Zen Diver - 23 Feb 2007 11:59 GMT > Just wondering if any other crazy people have tried using an X400 > (yup!) ND filter. It gives around 9 stops exposure decrease, for [quoted text clipped - 11 lines] > > (crossposted to aus.photo and rec.photo.digital, hope you don't mind) Have you considered using crossed polarizers. In order to do this you will need a linear polarizer stacked in front of a circular polarizer in order for TTL metering to work. This will give you a variable ND filter which will offer from -2 stops to completly obscured. I know it involves 2 layers of glass but it is very flexible.
jon
POTD.com.au - 23 Feb 2007 21:37 GMT >> Just wondering if any other crazy people have tried using an X400 >> (yup!) ND filter. It gives around 9 stops exposure decrease, for [quoted text clipped - 19 lines] > > jon I made this setup a couple of years ago and have never got around to using it. Thanks for the reminder, I must give it a go
BTW... I know you can buy these variable ND filters ready to go.... it's either Lee or Singh-Ray that has them. (I think it's Singh-Ray)
Graham Fountain - 24 Feb 2007 02:59 GMT >> Just wondering if any other crazy people have tried using an X400 >> (yup!) ND filter. It gives around 9 stops exposure decrease, for [quoted text clipped - 17 lines] > which will offer from -2 stops to completly obscured. I know it involves > 2 layers of glass but it is very flexible. I've used 2 stacked ND8s (effectively ND64) but not an ND400 - although looking at it in the hoya catalog they look pretty cool. I don't know how you'd get on with framing though - even with 2 ND8's the light coming through made it very hard to line up the shot, the camera couldn't AF, and it was hard to visibly assess focus. I guess that could be an advantage of crossed polarisers - align the polarisers, and align for the best polarising effect. Frame and focus, then turn the rear polariser to get the desired darkness (I'm guessing that if the front polariser was turned, it could have an adverse effect on the polarisation - similar to using a polariser 90 degrees from optimum, where reflections get boosted above the desired image). I guess this is an example where drop-in filters like Cokin would be preferable to screw-on filters like Hoya - with drop-in filters you could frame and focus using full light, then drop the filters in without disturbing the camera.
> jon Mr.T - 24 Feb 2007 04:44 GMT > I've used 2 stacked ND8s (effectively ND64) but not an ND400 - although > looking at it in the hoya catalog they look pretty cool. I don't know how > you'd get on with framing though - even with 2 ND8's the light coming > through made it very hard to line up the shot, the camera couldn't AF, and > it was hard to visibly assess focus. Surely since you would be shooting on a tripod anyway, that's all pretty irrelevant. Simply remove the filter to frame and focus.
> I guess this is an example where drop-in filters like Cokin would be > preferable to screw-on filters like Hoya - with drop-in filters you could > frame and focus using full light, then drop the filters in without > disturbing the camera. You can't screw on a filter without disturbing the camera? (however I do admit it would be slightly easier with a filter frame.)
MrT.
Graham Fountain - 24 Feb 2007 06:55 GMT >> I've used 2 stacked ND8s (effectively ND64) but not an ND400 - although >> looking at it in the hoya catalog they look pretty cool. I don't know how [quoted text clipped - 14 lines] > You can't screw on a filter without disturbing the camera? > (however I do admit it would be slightly easier with a filter frame.) I challenge you to screw a filter onto most lenses without disturbing it's focus and/or zoom position. Some lenses you might get away with it, but on a lot you won't. Anything with a moving front element, even if the front element doesn't turn, will be almost impossible. Even my rangefinder which has a fixed front element and uses the back element to focus, and is quite stiff in it's action (the old days when focus wasn't sloppy) will be disturbed by this action because of the slight give in the assembly. It's harder to do on lenses with a hood too, and a hood is almost a necessity when shooting long exposure.
> MrT. Mr.T - 24 Feb 2007 13:31 GMT > I challenge you to screw a filter onto most lenses without disturbing it's > focus and/or zoom position. Some lenses you might get away with it, but on a [quoted text clipped - 3 lines] > stiff in it's action (the old days when focus wasn't sloppy) will be > disturbed by this action because of the slight give in the assembly. Wow, you mean you are unable to see what zoom setting and/or focus distance has been set, in case you accidently move them? I can't say I've ever had a problem setting lenses that way. In fact I miss the depth of field markings that were on manual lenses, because I often set focus that way.
>It's harder to do on lenses with a hood too. I can't see why you don't just attach it to the filter first?
But if that's all too difficult for you, then I agree, don't use one.
MrT.
Rob - 23 Feb 2007 12:02 GMT > Just wondering if any other crazy people have tried using an X400 > (yup!) ND filter. It gives around 9 stops exposure decrease, for [quoted text clipped - 11 lines] > > (crossposted to aus.photo and rec.photo.digital, hope you don't mind) Why not try two polaroid filters stacked and get to the darkest point? may save the cost of trying to obtain a ND filter.
AustinMN - 23 Feb 2007 15:06 GMT > mark.thoma...@gmail.com wrote: > > Just wondering if any other crazy people have tried using an X400 [quoted text clipped - 15 lines] > Why not try two polaroid filters stacked and get to the darkest point? > may save the cost of trying to obtain a ND filter.- Hide quoted text - While this would result in a very dark filter, it would also introduce a significant color cast. What the color cast would be depends on the specific polarizers used, but it tends towards the blue (shorter) wavelengths.
Austin
Pete D - 23 Feb 2007 19:56 GMT >> mark.thoma...@gmail.com wrote: >> > Just wondering if any other crazy people have tried using an X400 [quoted text clipped - 22 lines] > > Austin We are talking digital here, this is easily corrected.
AustinMN - 23 Feb 2007 20:38 GMT > >> mark.thoma...@gmail.com wrote: > >> > Just wondering if any other crazy people have tried using an X400 [quoted text clipped - 24 lines] > > We are talking digital here, this is easily corrected. When the difference exceeds a couple of stops? Thanks, but I'm not interested in posterized reds.
Austin
Mr.T - 24 Feb 2007 04:48 GMT > > > While this would result in a very dark filter, it would also introduce > > > a significant color cast. What the color cast would be depends on the > > > specific polarizers used, but it tends towards the blue (shorter) > > > wavelengths.
> > We are talking digital here, this is easily corrected. > > When the difference exceeds a couple of stops? Thanks, but I'm not > interested in posterized reds. Why not use porarizers that are not quite so crapppy then? I haven't seen that problem even with the cheap "Arrow" brand filters!
MrT.
Lionel - 23 Feb 2007 13:26 GMT >Just wondering if any other crazy people have tried using an X400 >(yup!) ND filter. It gives around 9 stops exposure decrease, for [quoted text clipped - 9 lines] > >Anyone else used one, any insights before I have a go? I haven't, unfortunately, because I'd like to try some of the same experiments you've mentioned. If you do end up getting one, please let us know about your results - I for one, will be very interested.
 Signature W "Some people are alive only because it is illegal to kill them." . | ,. w , \|/ \|/ Perna condita delenda est ---^----^---------------------------------------------------------------
AustinMN - 23 Feb 2007 15:34 GMT On Feb 23, 5:29 am, mark.thoma...@gmail.com wrote:
> Basically I want try some odd stuff like shooting a city street in > daytime for 5 seconds or so, to remove a lot of the people/cars - you > know, that old stuff! I have some other interesting ideas for it... It takes more than 5 seconds to accomplish that. See below.
> So rather than stack a few 8x (I don't like more than one extra layer > of glass in front of my lenses..), I made a few enquiries to see if > you can get a really dark one, and the X400 came to my notice. > > Anyone else used one, any insights before I have a go? I have never used an X400, but I did use a pinhole apature in front of a 50mm lens back in my film days. I think I calculated an approximate apature of about f200. Start with a roudn piece of foil the size of a 1A/UV filter (it works better if you paint one side flat black), put a tiny hole in the center, and attach it to the filter with the black side towards the lens.
Bracket the exposure like crazy, and expect some diffraction artifacts (but these are not too bad with daytime shots).
If you are shooting digital, expect significant noise problems.
On the subject of "empty" daytime streets, depending on traffic patterns, it can take a long time to get an adequate exposure. It is not unusual for a pedestrian to stand still for 10 seconds while waiting to cross, for example. Even with a 30 second exposure, there will be an obvious ghost-like presence on the image.
One technique I have experimented with is to use a _very_ sturdy tripod heavily weighted, and take an image every 15 seconds until I had about 40 images. I then discarded any that had very bright or dark transient items (such as a white car or a pedestrian with a black shirt), and all but one where the same object was in several images. I averaged together the remaining 32 images. Although this was usable, I decided to go a couple steps farther. I did a difference between each of the original images and the combined image. I then subtracted this difference from it's original image. Once done to all 32 of the originals, I had a set of images where the transient objects were mostly gone (much closer to the intended image), but parts of the image which were identical to the average were not changed at all. Then I averaged together the 32 reprocessed "originals." The results were better, but still not perfect.
The original exposures were relatively short - around 1/250. If I were doing it over again, I would try using as long an exposure as I could, in an attempt to eliminate hard lines. In this particular situation, noise from long exposures isn't important because of the averaging of significant numbers of images, but it can bring out hot pixels in certain situations.
Austin
Rich - 23 Feb 2007 20:47 GMT On Feb 23, 6:29 am, mark.thoma...@gmail.com wrote:
> Just wondering if any other crazy people have tried using an X400 > (yup!) ND filter. It gives around 9 stops exposure decrease, for [quoted text clipped - 11 lines] > > (crossposted to aus.photo and rec.photo.digital, hope you don't mind) Why not just get a solar filter material and put it in a filter holder? Optically, it will be 100x better than stacked NDs.
mark.thomas.7@gmail.com - 23 Feb 2007 21:39 GMT Thanks for all the very helpful replies. To answer a few issues that were raised:
- the reason I don't want to use welders glass or solar filter is simply that I want to shoot in colour... I know welders glass gives a huge colour cast (as well as being optically quite poor), and I suspect solar filter has the same colour issues.
- similarly with crossed polarisers - my experiments have shown that when you get to the sort of darkness I need, the colour cast is so strong as to be uncorrectable. Maybe with very high quality and perfectly matched polas this is less of a problem?, but they would work out more expensive anyway.
- yes, I know 5 sec exposure won't 'remove' the people, but it will look weird enough, and yes, I'll try stacking exposures also.
Thanks again. I think I will order the X400 today (not stocked locally) and hopefully do some shooting next week - I'll post the results, but don't hold your breath as my life is a little busy right now... (O;
Rob - 23 Feb 2007 22:01 GMT > Thanks for all the very helpful replies. To answer a few issues that > were raised: [quoted text clipped - 17 lines] > results, but don't hold your breath as my life is a little busy right > now... (O; Well have you considered a pin hole attached to your camera. No cast :)
These can be made by drilling into a body cap to start with.
or start here at B&H Photo - Specialty Cameras < Cameras < Pinhole
you can buy pin holes separately.
rm
Warren Block - 23 Feb 2007 22:52 GMT > Just wondering if any other crazy people have tried using an X400 > (yup!) ND filter. It gives around 9 stops exposure decrease, for [quoted text clipped - 3 lines] > daytime for 5 seconds or so, to remove a lot of the people/cars - you > know, that old stuff! I have some other interesting ideas for it... Seems like this would be possible in PP. Take two shots a few seconds apart, and subtract the difference digitally. It would remove anything that moved, anyway.
 Signature Warren Block * Rapid City, South Dakota * USA
Warren Block - 23 Feb 2007 23:09 GMT >> Just wondering if any other crazy people have tried using an X400 >> (yup!) ND filter. It gives around 9 stops exposure decrease, for [quoted text clipped - 7 lines] > apart, and subtract the difference digitally. It would remove anything > that moved, anyway. Here's a tutorial on something like that:
http://www.dsphotographic.com/index.php/articles/how-to-remove-tourists-from-you r-photos/
Not as automated as I was thinking, but probably more practical.
 Signature Warren Block * Rapid City, South Dakota * USA
Wilba - 24 Feb 2007 00:59 GMT > Here's a tutorial on something like that: > > http://www.dsphotographic.com/index.php/articles/how-to-remove-tourists-from-you r-photos/ Can anyone help me do this in PS Elements? I can't find a way to add a layer mask, and using the selection brush tool in mask mode just paints over the image rather than revealing the layer underneath.
Or am I wasting my time using Elements ...? :-)
dj_nme - 24 Feb 2007 02:07 GMT >>Here's a tutorial on something like that: >> [quoted text clipped - 5 lines] > > Or am I wasting my time using Elements ...? :-) This works in Corel Photopaint, but should also work in PS Elements: Open your image and paste as new layer all of the other versions where the people have moved around. Use the eraser tool to remove part of the layers that have people on them to show the layer below it where there isn't. Basicaly using the eraser to make "holes" in the layer to let the background show through.
Wilba - 24 Feb 2007 03:43 GMT >>>Here's a tutorial on something like that: >>> [quoted text clipped - 13 lines] > Basicaly using the eraser to make "holes" in the layer to let the > background show through. That's exactly what I want and it works perfectly. Thanks!
I'm sure there would be more sophisticated ways to do it in Elements, but that works for my level of expertise.
Cynicor - 24 Feb 2007 03:15 GMT > Here's a tutorial on something like that: > > http://www.dsphotographic.com/index.php/articles/how-to-remove-tourists-from-you r-photos/ Here in New York, we have two additional methods for removing tourists from photos. The first is firing a pistol wildly in the air for a few seconds before taking photos. The second is yelling "Hey, I think the Today Show is filming something over there!"
Jeff R. - 24 Feb 2007 03:28 GMT Mark...
Subsequent to your initial post -
I did some friggin' about with one of my solar filters - a glass one which renders the sun "silvery". I taped it in front of my 12-24 and took a 30 second shot of the backyard at f/4. (Needed more exposure, but wanted to use the camera's shutter timer.)
Here is a thumbnail of the result, replete with lots of spots on the filter (courtesy of the short f.l. lens)
http://www.mendosus.com/jpg/solar-thumb.jpg
The colour is much better than I'd feared - as you can see. Very magenta, but almost controllable.
Here's a crop of part of the left-hand side, with some colour/exposure tweaking: http://www.mendosus.com/jpg/solar-1.jpg Again, not brill, but better than expected.
Certainly promising enough to continue.
Next time: Camera in full shade; Longer focal length; Longer exposure.
Thanks for the idea.
-- Jeff R.
Charles - 24 Feb 2007 03:45 GMT >Just wondering if any other crazy people have tried using an X400 >(yup!) ND filter. It gives around 9 stops exposure decrease, for [quoted text clipped - 11 lines] > >(crossposted to aus.photo and rec.photo.digital, hope you don't mind) Many years ago I tried using a Kodak ND4, 13.3 stops. Exposure was 20 minutes or a stream, I really wanted to get the flowing effect. Overall it was disappointing, never used it again.
those things certainly have gone up in price the last few decades.
k - 24 Feb 2007 06:42 GMT | Just wondering if any other crazy people have tried using an X400 | (yup!) ND filter. It gives around 9 stops exposure decrease, for [quoted text clipped - 9 lines] | | Anyone else used one, any insights before I have a go? cross a pair of polarizing filters
k
textilis - 25 Feb 2007 04:58 GMT > cross a pair of polarizing filters > > k Never tried this, but theoretically wouldn't it produce a centre pinprick of light?
Rob - 25 Feb 2007 10:05 GMT >>cross a pair of polarizing filters >> >>k > > Never tried this, but theoretically wouldn't it produce a centre pinprick of > light? Pin hole lens?
k - 25 Feb 2007 23:35 GMT | > cross a pair of polarizing filters | > | > k | | Never tried this, but theoretically wouldn't it produce a centre pinprick of | light? no :)
k
Rob - 25 Feb 2007 23:52 GMT > | > cross a pair of polarizing filters > | > [quoted text clipped - 7 lines] > > k You must have to use two circular polarisers then :)
Graham Fountain - 26 Feb 2007 07:54 GMT >> | > cross a pair of polarizing filters >> | > [quoted text clipped - 10 lines] > > You must have to use two circular polarisers then :) Nope not even that. The only difference between a circ-pol and a regular pol is that the circ-pol has an element after the polarising element that de-polarises the light, this is so AF sensors will still work. You can't do the crossed polariser trick with circ-pols, because by the time the light hits the 2nd filter, it is no longer polarised. The 2nd polariser can be a circ-pol, but the 1st can't be. Irrespective of the type of filter, you won't get a central pinprick of light. Remember that this polarising action occurs across the entire face of the filter, not just in the centre. When light passes through a polariser, the intensity varies depending on the angle of polarisation of the light. A polariser doesn't work by letting 0% of light in unless it's polarisation matches the polariser and if so lets 100% in. The proportion of a ray of light that is passed by the polarising filter (assuming a perfect filter) is equal to the cosine of angle of the light's polarisation, relative to the polarising filter's angle. This is basic vector theory - a ray with intensity of 1 at an angle of 30 degrees, will have an intensity of .86 (cos 30) in the X direction and .5 (cos 60) in the Y direction. After passing through the polarisor, only the X direction remains, so the light ray will have intensity .86 and now be polarised in the X direction. In theory, if the image was composed of randomly polarised light and the polariser was perfect, the filter would pass on average 71% of light through (average of cos(0-90)), or a 1/2 stop light reduction. When you have crossed polarisers, because the light coming from the first polariser is polarised in that direction, all of the light is presenting a vector to the 2nd polariser. If the 2nd polariser is at a 30 deg angle to the first, the light it passes will only be cos(30 deg) of what it would otherwise pass, or 86% (0.86). Because the first filter took 71% of the light away, the combined effect is a about 61% of light making it through, or almost a 1 stop reduction. Turn the filter to 60 degrees, and we now have 50% of 71%, or 35% passing through (1-1/2 stop reduction). Go to 89 degrees, and the 2nd filter only passes 1.7% of the 71% of the original light, so we have 1.2% of the original light coming through, or 6-1/2 stops lost. At 90 degrees, the 2nd filter passes 0% and so we have a total blackout. Now these sums are for a perfect filter, which we don't have. In real life, the polarising filters don't do a perfect job of blocking polarised light, so they do let through some light at 90 degrees, even though theory says they shouldn't. Not only that, but due to the opacity of the glass and polarising element, they block more than just the polarised light, so they have a greater reduction to light that is in the polarised direction than theory says they should. That is why in practice a polariser drops between 1-2 stops of light, when in theory it should only drop 1/2 a stop, and why you won't get total 100% blockout by using crossed polarisers.
mark.thomas.7@gmail.com - 26 Feb 2007 11:27 GMT Some interesting discussion here - I'm a little internet-time- challenged for the next few days, but I'll wade through them gradually - forgive me if I haven't responded to particular issues just yet...
I did a little bit of research on the crossed polarisers business, and found very little useful information. But it seems that polarisers work best on the red/green end of the spectrum and become less efficient as they get to the blue/violet - that makes sense to me, given it is just a 'venetian blind', and the blue wave-heights are smaller (how's that for an over-simplification..!)
I've tried rotating several different brands of polarisers (no *really* expensive ones though!), and they all gave strong blue/violet color casts (to my eye) when they were totally/almost crossed. I did not do any photographic tests to verify it - has anyone here or are we all shooting from the hip? (O;
I did find this: http://www.jr-worldwi.de/photo/index.html?polarizer.html
There's some very interesting images and tests there (although some images are missing), and it shows that crossed Pola's can work as an IR-pass filter, and they do strange things to colour at the extremes..
I've got two pola's, but they are a 58 and 49, so testing this out will require some fiddling and it isn't high on my agenda of things to do before I die...
Anyway, I've ordered the X400, so I'll let you know what happens and hopefully post some weird masterpieces... yeah right.
mark.thomas.7@gmail.com - 26 Feb 2007 12:36 GMT And here is an example of the color problem:
http://forums.dpreview.com/forums/read.asp?forum=1024&message=8084020
If you can colour-balance that one back into something approaching reality...
I *salute* you!!
|
|
|