Quick Question

mattchewone

mattchewone

Suspended / Banned
Messages
884
Name
Matt
Edit My Images
Yes
Could someone please explain a few things for me so i can understand them fully.
Was reading about ND grad filters and it was saying that the top half of the filter could be 2stops darker. Does this mean if your shooting at F/2.8 that that the top half would be f/5.6?

Also i friend of mine told me this calculation just wondering if it works and if there are any similar for anything else.
To calculate the minimum shutter speed for the sharpest image, times the focal length by 1.6 (crop sensor).
50mm*1.6 = 1/80


Many Thanks
Matt
 
mattchewone said:
Was reading about ND grad filters and it was saying that the top half of the filter could be 2stops darker. Does this mean if your shooting at F/2.8 that that the top half would be f/5.6?
Click to expand...

Nope. Let's say you're taking a landscape shot and you take two meter readings. The foreground meters at 1/125 and F2.8 and the sky meters at 1/125 and F5.6. Obviously you can't get both exposed correctly regardless of which setting you use so you've got a few choices.

Either compromise, say 1/125 at F4, the sky will be slightly overexposed the foreground slightly under.

Take two shots at the correct exposure for each and blend them in an editing programme to get the result you want or

Use a grad filter of appropriate strength on the sky. Take the shot exposing for the foreground, the graduation reduces the exposure required for the sky and you should end up with the correct exposure for the image overall.
 
1. Basically yes, in the sense that the very darkest part of the filter will let in 2 stops less light than the very brightest part - so the effect will be that the darkest part will only be letting in 1/4 of the light that the brightest part lets in. But of course, the lens is still at f/2.8 so depth of field etc are unaffected.

2. This is just a rule of thumb based on focal length x the ability to hold a camera still. It works like this: On a 35mm camera, most people need a shutter speed that matches the focal length of the lens for camera shake not to be apparent, e.g. 50mm lens = 1/50th sec, 1000mm lens - 1/1000th sec. If you're using a cropped frame digital with, say, a 1.6 crop, then just multiply the shutter speed by 1/6.

As I said though, it's just a rule of thumb. Some people can hold cameras still, some can't.
 
Got you? Whats the best metering mode? Evaluative? spot? Partial?

When you say take a reading how do you actually do this properly? As i dont tend to use M mode that much at the moment as i dont want to screw pics up so i use AV or TV?
I have used M a few times, i just adjust the shutter and aperture until the exposure is perfect. Is this right? It takes time to do this, how do people do it so fast lol.
 
Garry Edwards said:
1. Basically yes, in the sense that the very darkest part of the filter will let in 2 stops less light than the very brightest part - so the effect will be that the darkest part will only be letting in 1/4 of the light that the brightest part lets in. But of course, the lens is still at f/2.8 so depth of field etc are unaffected.

2. This is just a rule of thumb based on focal length x the ability to hold a camera still. It works like this: On a 35mm camera, most people need a shutter speed that matches the focal length of the lens for camera shake not to be apparent, e.g. 50mm lens = 1/50th sec, 1000mm lens - 1/1000th sec. If you're using a cropped frame digital with, say, a 1.6 crop, then just multiply the shutter speed by 1/6.

As I said though, it's just a rule of thumb. Some people can hold cameras still, some can't.
Click to expand...

So would it look like this - 50mm = 1/50 * 1.6 = 1/80mm or have i worked it out wrong lol.
 
mattchewone said:
So would it look like this - 50mm = 1/50 * 1.6 = 1/80mm or have i worked it out wrong lol.
Click to expand...

Yup, that's about right, but remember that it is a very general rule of thumb. Image Stabilised lenses or camera bodies will allow for a slower shutter speed when handheld also which should be taken into account too.
 
Thank you very much!

Anyone got any tips on metering whats the best mode etc. How to actually meter as im not too sure i know how.
 
mattchewone said:
Thank you very much!

Anyone got any tips on metering whats the best mode etc. How to actually meter as im not too sure i know how.
Click to expand...

If you are going to be using grad filters, you pretty much have to go with spot metering.
If you are going to be using a polariser, put that on and adjust it before taking any metering. Then before you put the camera on the tripod and set up the composition, turn to either TV or AV mode and take some spot meter readings around various bits of the foreground and sky. This means filling the spot-meter circle on your camera fully with bits of foreground or bits of sky and noting what exposure settings your camera sets itself to. Take a number of readings of different bits of foreground and sky (clouds vs clear blue sky or trees vs grass) You should, in general, notice that the sky is mostly one setting, with the foreground mostly another setting, usually darker.
Example, go to AV, set it to what f-number you want to use in the picture (say f8). Various spot meters in the foreground might read 1/20 f8, while the sky might read 1/90 f8.

If you know your f-stop tables you can see that the sky is roughly 2 stops brighter than the ground. (if, like me, you don't know them, a handy tip is to go to manual mode and set it to 1/20 then count the number of steps until it reaches 1/90 remembering if your camera is set for 1/2 or 1/3 stop jumps)

In manual mode then, use the setting from the foreground spot metering, and use the appropriate filter to reduce enough light from the sky to make it only as bright as the foreground (in this case, it's 2 stops brighter, so use a 2 stop filter.)

You are now using manual mode and spot metering. Apparently that makes you a better photographer too, which is a nice little bonus.
 
How do i know if my camera is set to jump 1/2 or 1/3 stop jumps?
Also where is the spot meter circle?

Thanks alot that is really helpful! Is there an actually Fstop table available to print out so i can read before bed time and learn lol!
 
There's some terrible maths going on in here :lol:

mattchewone - 50mm*1.6 = 1/80... errr no it's not 50 x 1.6 = 80 not 1/80

Garry - agree with this bit e.g. 50mm lens = 1/50th sec, 1000mm lens - 1/1000th sec.

But this ??? If you're using a cropped frame digital with, say, a 1.6 crop, then just multiply the shutter speed by 1/6.

So if I'm shooting a 300mm lens with a 1.6 crop the minimum shutter speed you're suggesting here is 1/300mm (in keeping with the first part of your reply) but divided by 6 = 1/1,800th sec

:lol::lol::lol:

Soz, bored and this looked like a bit of fun

DD
 
Garry Edwards said:
2. This is just a rule of thumb based on focal length x the ability to hold a camera still. It works like this: On a 35mm camera, most people need a shutter speed that matches the focal length of the lens for camera shake not to be apparent, e.g. 50mm lens = 1/50th sec, 1000mm lens - 1/1000th sec. If you're using a cropped frame digital with, say, a 1.6 crop, then just multiply the shutter speed by 1/6.

As I said though, it's just a rule of thumb. Some people can hold cameras still, some can't.
Click to expand...

I'd agree with that except for the 1.6 crop making any difference. Movement of the camera causes the image to move on the sensor. How far it moves (and hence how much blur) will not change just because you are not recording the outside of the frame with a bigger sensor.
 
RobertP said:
I'd agree with that except for the 1.6 crop making any difference. Movement of the camera causes the image to move on the sensor. How far it moves (and hence how much blur) will not change just because you are not recording the outside of the frame with a bigger sensor.
Click to expand...



Hmmm, soz Rob but I can't agree with that

I know what you mean, but you're cropping (effectively magnifying) the scene and it's the magnification that shows up camera movement; hence more magnification at 500mm than 50mm, and the same with the crop too

Imagine you've shot a FF at 500mm and it looks sharp enough at 100%, but you zoom into the image by 1.5x or 1.6x (i.e. you're cropping it) and now it's not so sharp as you can now detect camera shake - it's the same problem with cropped sensors and the min ideal focal length as you're not just cropping into an image, you're magnifying the scene that results in an image

Make sense anyone ???

:thinking:

DD
 
Guess you could look at it both ways. if the pixel density of the sensor is similar for both then I still think the crop will make no difference to how much movement blur there is.

If you choose to magnify either image you will still see the same degree of blur.
 
i didnt think the cropped sensor made any difference to the calculation since it doesnt increase the length of the lens?:shrug:
 
dave, you dont zoom in though. you just reduce the final image size?

(probably completely wrong, but be good to get my head round it)
 
I wasn't talking about zooming into the image (which would make no difference), but zooming/magnifying the scene which clearly does

A 50 mm lens is wider in the capture angle than a 500mm one, you're magnifying the scene even though you're still using the same pixels - the degree of movement of the lens and it's magnified shake effect is clearly different, we all accept that, which is why we say 50mm = 1/50th, 500mm = 1/500th

What I'm saying is that it's the scene magnified that matters, not the crop, so as I see it a more powerful lens is simply more prone to camera shake, hence the effect of the crop is to make the lens more powerful in respect of the magnification of scene captured, hence it needs a faster shutter speed

Look at it another way, what would the slowest recommended shutter speed be on a MF camera for a 500mm lens? It would be less than 1/500th wouldn't it, due to it capturing more angle of the scene and hence less magnifying than the same 500mm focal length on a 35mm camera or cropped digital; and it would be less still on a 10x8" plate camera

It's the same problem you get focussing more closely too, my 90mm macro is easy to use handheld photographing a full length portrait at 1/90th sec and there's no apparent camera shake; but it'd look a mess handheld at 6" trying to photograph a spider at 1/90th as the magnification of the scene AND the camera movement has increased

Does that make more sense now? Or am I really talking out of my windy-place

:D:D:D

DD
 
DiddyDave said:
Does that make more sense now? Or am I really talking out of my windy-place
Click to expand...

Only the field of view changes, not the magnification? Upscaling after cropping is totally different to actually magnifying optically.
 
robmiller said:
Only the field of view changes, not the magnification? Upscaling after cropping is totally different to actually magnifying optically.
Click to expand...

Nope - that's wrong in the context of what I was writing, about the scene in front of you and which bit you decide to shoot

However we like to say it, the more magnified or the narrower the angle of view we shoot, the more powerful the lens is the more it's prone to camera shake- yes?

If so, a smaller sensor is still effectively narrowing the field hence it's more prone to camera shake than a larger sensor with the same focal length

So... I've given in now, more life to live outside of "Quick questions" :lol:

:wave:

DD
 
Regarding the use of grad filters, why not rely on your SLR to give you a meter reading - in average or evaluative mode? It seems to work for me. To be absolutely certain you can bracket exposures 1/3 or 1/2 stop either side and choose the best exposure later.

You can use the spot meter to estimate the best grad to use. If the sky is, for example, 2 stops lighter than the foreground , use a two stop grad. If its one stop lighter, use a one-stop grad.

Sky = 1/90th second, foreground = 1/20th second >>>> 2 stops (approx)

Sky = 1/200th second, foreground = 1/100th second >>>>> 1 stop

But there seems to be 2 separate threads going on here at the same time.

:)
 
I have caused mayhem lol!

Thanks to everyone for there input although im more confused now then when i started lol. Mainly on the rule of thumb for minimum shutter speed when handheld on a 1.6crop sensor.
Would it best to go for the 50mm = 1/50? or 50 x 1.6 = 1/80?
Or shall i just try both and work out which work and get the sharpest image?


Also where is the best place to buy decent but not too expensive ND grad filters? Do they come in sets?
 
RobertP said:
Guess you could look at it both ways. if the pixel density of the sensor is similar for both then I still think the crop will make no difference to how much movement blur there is.

If you choose to magnify either image you will still see the same degree of blur.
Click to expand...

Pixel density doesn't come into it, it's the Circle of Confusion that defines the amount of blur and sensor size is a factor there. Because a 1.6 crop sensor has a smaller CoC it requires less movement of the camera before it becomes noticable on a print when compared to a full frame camera.

But that's not the end of the story, to get the same shot from a full frame camera you either use a lens 1.6x longer and therefore need the same shutter speed as the crop camera or crop from the full frame in which case you're still enlarging the CoC 1.6x and the shutter speed still needs to be higher.

The problem with comparing crop and full frame is that you need to remember that a full frame will always give the same results as the crop if it was cropped too - including DoF, camera shake, etc.
 
pxl8 said:
Pixel density doesn't come into it, it's the Circle of Confusion that defines the amount of blur and sensor size is a factor there.
Click to expand...

Pixel size and circle of confusion are directly related, as are pixel size and pixel density, so in fact pixel density precisely defines circle of confusion. Sensor size is only a factor if we specify a constant number of megapixels, since then it defines the pixel density.


None of this is actually important because the 1/focal length rule was created for making acceptable prints from film, not for approaching the circle of confusion of a digital camera. If you want to do that then you can forget handholding unless using very high shutter speeds and relatively short focal lengths.
 
Jamougha said:
Pixel size and circle of confusion are directly related, as are pixel size and pixel density, so in fact pixel density precisely defines circle of confusion.
Click to expand...

The CoC is the same for a given sensor size no matter what the pixel count is, see here:

http://www.dofmaster.com/digital_coc.html

Presumably because the pixels are all much smaller than the CoC to start with.
 
pxl8 said:
Pixel density doesn't come into it, it's the Circle of Confusion that defines the amount of blur and sensor size is a factor there. Because a 1.6 crop sensor has a smaller CoC it requires less movement of the camera before it becomes noticable on a print when compared to a full frame camera.

But that's not the end of the story, to get the same shot from a full frame camera you either use a lens 1.6x longer and therefore need the same shutter speed as the crop camera or crop from the full frame in which case you're still enlarging the CoC 1.6x and the shutter speed still needs to be higher.

The problem with comparing crop and full frame is that you need to remember that a full frame will always give the same results as the crop if it was cropped too - including DoF, camera shake, etc.
Click to expand...


Phew!

Glad another is on my side, albeit with yet another take on the explanation of the problem/solution

Thought my windy parts had gone overdrive for a mo !!!

:lol:

DD
 
pxl8 said:
The CoC is the same for a given sensor size no matter what the pixel count is, see here:

http://www.dofmaster.com/digital_coc.html

Presumably because the pixels are all much smaller than the CoC to start with.
Click to expand...

I think you're getting confused between two common usages of the term CoC.

The strict technical definition of CoC is related to the size of the Airy disk caused by diffraction at the aperture. If you like it's the minimum spot size that can be resolved by the lens, due to physical limitations. Providing that the pixel size is much larger than the CoC it has no bearing on the camera's ultimate resolution. With 12mp on a DX sensor this means up to about f/11.

Photographers tend to use the term CoC to refer to the minimum permissible circle of confusion. If you're not diffraction limited then this is what defines the resolution of the camera, assuming a sufficiently high-quality lens. For handholding this is generally the number that matters - whether you have enough resolution to see the blur or not. Only if you were handholding at f/16+ would CoC matter. The minimum permissible CoC is generally accepted as twice the pixel size.

edit - actually I think my explanation of CoC is wrong but the general point is correct. The minimum permissible circle of confusion is what matters here unless you're diffraction limited.
 
In simple terms I see it thus:

The minimum size of a "point" on the film/sensor that would be considered sharp when viewing a print at a "normal" viewing distance.

And that is the basis under which I put forward my earlier comments regarding sensor size and shutter speed. With a smaller sensor the enlargement to print size is greater therefore the size of the point needs to be smaller to compensate. The same "size" limit also applies to movement due to camera shake.
 
Aha, wait, looking on wiki there is in fact a third definition of circle of confusion which matches yours. Now I understand what you're saying; we were getting caught up in 3 different meanings for the same term!

So yes your original post was perfectly correct. My apologies for having doubted you.:D
 
mattchewone said:
I have caused mayhem lol!

Thanks to everyone for there input although im more confused now then when i started lol. so where is the best place to buy decent but not too expensive ND grad filters? Do they come in sets?
Click to expand...

seeing how everyone else has wandered off topic that much they seem to be ermmm wherever .:lol:
to answer the question ND grads can be bought seperatley ( sp ) or in sets 1 good place IMO is here.

http://www.formatt.co.uk/stills-filters/products.aspx

or the same filters via a different supplier

http://www.teamworkphoto.com/index....78_767&zenid=297ddde76bb37cbb8596f4b979dfb86b



another make of filter is cokin

worth having a look as it includes example photos

http://www.cokin.co.uk/pages/main.htm
 
OK - back OT

I hate ND grads with a vengeance, they are NEVER right IMO for a shot and always look false when used. They slow down any semblance of creativity and are never the right shape for the scene unless it's solely a seascape, where blending two exposures is better

Grrrrr

DD
 
Thanks very much! Anyone recommend a decent filter just a clear one to prevent the lens glass getting damaged.
Cheers.
 
DiddyDave said:
OK - back OT

I hate ND grads with a vengeance, they are NEVER right IMO for a shot and always look false when used. They slow down any semblance of creativity and are never the right shape for the scene unless it's solely a seascape, where blending two exposures is better

Grrrrr

DD
Click to expand...

Cause blending two images together never looks false and unnatural :)
Both are techniques that can be done to various degrees of success.

It also (for me) means loading the images in another piece of software and playing with them there to do the blend, something I am loathe to do. I'd far rather slow down the creative process (I don't see that as a bad thing) than have to play with it later.

I don't refuse to blend images, I've done it in the past, I'll do it again as there are times when it's the only way, I just really dislike doing it.
I'm still trying to perfect my technique using grads, and still having some very bad results from time to time. Maybe I'll reach a point where I agree with you.

At the end of the day, you do what you feel happiest with. For you thats blending, for me it's (currently) using grads. Neither is wrong.
 
Thanks for the comments. Get a nice feel for different opinions. Its good to see people do things there own way and not copy each other etc etc.
 
Eau Rouge said:
Cause blending two images together never looks false and unnatural :)
Click to expand...


Don't usually do that either, although I do 'HDR' sometimes for fun and in extreme conditions - my HDR program is easier to use than Photomatix so blending is no trouble at all

What I could never get to grips with was the greying of the mountain tops caused by the ND Grads as there are no Valley shaped ones, and on many of my mates' images you can see that the filter was used

Of course there are plenty of occasions when they've been used well, but they are slow and cumbersome I find - so for me, they're a :nono::nono::nono:

This (below) is a typical 'normal' DD shot - not HDR, not blended, no filters used (though some around me had their filters on aplenty); looks okay to me

Takes all sorts as they say

:thumbs:

DD

Black_Nab_for_TP.jpg
 
Gawd, I'm agreeing with DD again but...

ND's are great for film/video but I do find the effect they have on the tops of buildings, trees, etc. ruins the image - something that a blend can neatly side-step :thumbs:
 
You must log in or register to reply here.
Back
Top