Hard vs. Soft ND grads on Cropped sensor

Messages
865
Name
Anand
Edit My Images
Yes
Just as the title suggests, is there any recommendation for one over the above? just read somewhere, where Lee apparently recommend the hard grads for using with DSLR cropped sensors like a D5000 I have?

Thanks.
 
I was under the impression that hard grads were for when you have a straight line like a straight horizon and a soft grad is for things like rolling hills where its not a straight line
 
Not really to do with sensor size.

Soft grads for more undulating horizons and hard grads for very level horizons such as seascapes.
 
I was under the impression that hard grads were for when you have a straight line like a straight horizon and a soft grad is for things like rolling hills where its not a straight line

Thanks Andy,That's definitely what I thought..but kind of re-thinking after reading what I read...let me find the link
 
Not sure about that, if you are a working landscape photographer using a crop camera (unlikely though) there is probably a case for having both hard and soft grads.

You would use a hard grad when the area to be graduated to has a distinct line with few or any intrusions (such as a seascape), and a soft grad when the terrain or area has a lot of intrusions (such as a cityscape or mountainous area).

If you can not afford both sets I would probably limit myself to a hard grad 1.2 (4 stops) and a soft grad 0.9 (3 stops).
 
here's the quote from Speedgraphic

..On full frame format, soft is more generally useful, but photographers specialising in landscapes where there is often a sharp dividing line between sky and foreground (e.g. seascapes) will usually prefer a hard graduation. For smaller sensor formats, Lee recommend using a hard grad set for most scenarios.

and the link

Linky
 
Perhaps they think the transition area of soft grad covers too much of the sensor.
 
Perhaps they think the transition area of soft grad covers too much of the sensor.

That's very droll.

What difference would it make? If you were using a 50mm on FF or a 35mm on crop the AOV is about the same therefore the amount of transition would be about the same.
 
Sensor size makes no difference at all.
 
That's very droll.

What difference would it make? If you were using a 50mm on FF or a 35mm on crop the AOV is about the same therefore the amount of transition would be about the same.

Definitely that sounds like a good point..wonder why lee have mentioned it though then.

It's only because I am hoping to buy a Lee DSLR starting kit which contains a hard ND grad, but I was hoping to buy a soft one..So wondered whether this was worth a buy :)..
 
I just said perhaps that's their theory ?

As per my first post - the choice of hard / soft is to do with subject.
 
Thanks Phil. Guess I am going to bite the bullet and go for this DSLR starter kit then.... Will probably leave theory aside and listen to more practically experienced toggers here :)
 
Sensor size makes no difference at all.

My impression was that the effect was linked to the focal lenght multipler.

The wider a lens is the smaller the graduated area of a filter will be in the image so the harder the transition will appear.

So if you were using exactly the same focal lenght on a crop a hard grad might give a similar effect to a soft grad on FF but if you were using a focal lenght that gives the same field of view(say 10mm on a crop rather and 16mm on FF) the effect for grads would be the same aswell.

If thats correct then really the advice should be, move to harder grads as your field of view gets smaller for the same effect.
 
Last edited:
Well, I guess the actual manufacturers of the filters dont know what they are talking about as the TP collective have said otherwise.

According to Lee, Hards are better on crop cameras. for the reason Moreorless has stated above. The sensor size does make a difference, hence Lee saying so.

I'd much rather follow thier advice than anyone elses, however you should still have both hard and soft to cover all eventualities.
 
Well, I guess the actual manufacturers of the filters dont know what they are talking about as the TP collective have said otherwise.

According to Lee, Hards are better on crop cameras. for the reason Moreorless has stated above. The sensor size does make a difference, hence Lee saying so.

I'd much rather follow thier advice than anyone elses, however you should still have both hard and soft to cover all eventualities.

In terms of the way people are actually going to use them I think Jim is correct though, crop users will tend to use the same field of view as FF users not the same focal lenght so there will be no difference.

Perhaps Lee's advice dates from the days when there wasnt any ultrawide coverage for crop sensors(at least none that would take filters). If 16mm was the widest you could go on a crop then hard grads might be a better choice.
 
That's very droll.

What difference would it make? If you were using a 50mm on FF or a 35mm on crop the AOV is about the same therefore the amount of transition would be about the same.

I suspect you need to rethink this Eddie. The attentuation from 0EV to -xEV ( -x being the highest density part of the filter) will be to the limit of the entire image circle. A crop sensor is not going to utilise the same extremities as a FF sensor and hence a soft grad would be unlikely to provide max attenuation on a crop.

I think :thinking:

Bob
 
I'm trying to get my head round this sensor size making a difference.

Surely, regardless of any camera format, you grad accordingly to the composition?

I can't see how a hard edge could work in a mountain area on a crop in preference to a soft, yet a full frame you would preferentially use a soft for the same composition?

Andy
 
I'm trying to get my head round this sensor size making a difference.

Surely, regardless of any camera format, you grad accordingly to the composition?

I can't see how a hard edge could work in a mountain area on a crop in preference to a soft, yet a full frame you would preferentially use a soft for the same composition?

Andy

Its not the sensor seize, its the field of view.

If you use the same grad for a shot taken at 10mm and 20mm then the graduation will appear to be twice as hard for the 10mm shot because the area of graduation takes up half the space on the image.
 
but wont that be a different composition ?

Does that reasoning apply at 16mm vs 32mm on FF for example?
 
Last edited:
I suspect you need to rethink this Eddie. The attentuation from 0EV to -xEV ( -x being the highest density part of the filter) will be to the limit of the entire image circle. A crop sensor is not going to utilise the same extremities as a FF sensor and hence a soft grad would be unlikely to provide max attenuation on a crop.

I think :thinking:

Bob

Bob, I don't quite follow.

If the AOV is the same then surely the same grad filter will have the same effect on both formats, if not then what am I supposed to use on my H1D which has a 1.6x increase on a FF DSLR or on my bridge camera which has a much smaller sensor than a crop camera?
 
I suspect you need to rethink this Eddie. The attentuation from 0EV to -xEV ( -x being the highest density part of the filter) will be to the limit of the entire image circle. A crop sensor is not going to utilise the same extremities as a FF sensor and hence a soft grad would be unlikely to provide max attenuation on a crop.

I think :thinking:

Bob

I think Eddy is right, the way I read it. Focal length is irrelevant and it's the field of view that makes the difference. It's simply how much of the filter is 'seen' by the lens, hence wide angles show quite strong grad effects, and on longer lenses it's pretty much invisible.

Other factors are the f/number, which has a depth of field effect that pulls the grad line slightly more into focus at higher f/numbers, ie harder, but that's pretty marginal TBH. And the other one, similarly a DoF effect, is how far the fillter is held in front of the lens. Again, that's marginal but if you hold a grad right next to the lens without a holder, and then with the holder in the furthest slot, there's a very slight hardening.

I don't know where Lee are getting their into, and since Anand's link didn't work for me it's difficult to say. Possibly a reference to the DoF changes with full-frame vs crop? Dunno. It's not helpful the way it's been quoted here though.

Edit: not all manufacturers definition of hard and soft is the same, and not always consistent either. Lee have what I would call the best interpretation, and most reliable. HiTech are quite noticeably harder, SRB too.
 
Last edited:
My impression was that the effect was linked to the focal lenght multipler.

The wider a lens is the smaller the graduated area of a filter will be in the image so the harder the transition will appear.

So if you were using exactly the same focal lenght on a crop a hard grad might give a similar effect to a soft grad on FF but if you were using a focal lenght that gives the same field of view(say 10mm on a crop rather and 16mm on FF) the effect for grads would be the same aswell.

If thats correct then really the advice should be, move to harder grads as your field of view gets smaller for the same effect.

Just being silent and trying to absorb the content..I suppose this is more complicated than I initially thought...may I thank everybody for the inputs so far.
 
Bob, I don't quite follow.

If the AOV is the same then surely the same grad filter will have the same effect on both formats,

You may well be right Eddie but I'm getting a different answer when I look from a different viewpoint.
The gradient will affect the intensity projected on the image circle (parts of) and clipping the image circle (as crops do) will remove the darkest region (soft grad) whilst it will likely (?) be maintained with a hard grad.
I can't see how the AOV comes into it....stick a grad on a 24/1.4 or a 135/2 and the graduated region will still go from the horizon line to the periphery.

Bob

Edit Warning to self....you could end up looking silly here
 
Last edited:
True but it could be the field of view itself rather than the subject that could effect which grad you choose.

Yeah I think the speedgraphic quote on Lee was specific to the sensor size whereas your point was more angle of view.

I was more disputing the speedgraphic point :thumbs:
 
You may well be right Eddie but I'm getting a different answer when I look from a different viewpoint.
The gradient will affect the intensity projected on the image circle (parts of) and clipping the image circle (as crops do) will remove the darkest region (soft grad) whilst it will likely (?) be maintained with a hard grad.
I can't see how the AOV comes into it....stick a grad on a 24/1.4 or a 135/2 and the graduated region will still go from the horizon line to the periphery.

Bob

I agree there will be a difference between different focal length lenses on the same format, but a 10mm lens on a 1.6 crop will show almost the same image as a 16mm lens on a FF camera from the same distance. If you then put the grad on the horizon of each camera the result should be similar to all intents.

I suppose we need someone with a plethora of cameras, lenses and grads to try it out.
 
I stand by my earlier comment about sensor size making no difference.

I agree with those who say focal range does though.

A grad is more pronounced at wide angles, so on a crop at 17mm will be very hard and obvious, whereas on FF at 100mm the effect will be less. Reverse this and you get the same results, whatever the sensor maybe, 17mm on FF will be more obvious than say 50mm on a crop.

Sensor size makes no difference, it's focal length that makes a difference.

To see how the focal setting affects the way the grad works put one on a wide-angle zoom and take a sequence of shots starting with a shot from the wide end, a shot at 30mm then one at 50+mm and see how it changes filters effect.
 
Last edited:
TCR4x4 said:
Well, I guess the actual manufacturers of the filters dont know what they are talking about as the TP collective have said otherwise.

According to Lee, Hards are better on crop cameras. for the reason Moreorless has stated above. The sensor size does make a difference, hence Lee saying so.

I'd much rather follow thier advice than anyone elses, however you should still have both hard and soft to cover all eventualities.

Ok, so what's different about using a grad on a crop at 10mm and FF at 16mm?

There is no difference in the way the grad works, so that's just simple, but well meaning mis-information by the manufacturer, probably written years ago before the introduction of crop UWA lenses.
 
Last edited:
I agree there will be a difference between different focal length lenses on the same format, but a 10mm lens on a 1.6 crop will show almost the same image as a 16mm lens on a FF camera from the same distance. If you then put the grad on the horizon of each camera the result should be similar to all intents.
I agree that the image will be the same but the gradient won't as the filter is not on the object plane.
Envisage a filter, clear in the centre and then with concentric rings each of which is -1EV darker as you progress outwards. The image circle of a crop will see fewer of the circles and hence will not record the darkest region at the periphery.

I suppose we need someone with a plethora of cameras, lenses and grads to try it out.
I'll see if I can find such a person :thumbs:

Bob

Still preparing to look silly
 

That quote is nonsense, and I'd be amazed if Lee makes any such recommendation either.

If focal length is adjusted by the crop factor to maintain field of view, the grad effect is substantially the same.

If you want to be picky about it, the f/number should also be adjusted by the crop factor to maintain identical depth of field, and the filter held exactly the same distance from the front of the lens. If you do that, the grad effect will be 100% identical regardless of format.
 
There is no difference in the way the grad works, so that's just simple, but well meaning mis-information by the manufacturer, probably written years ago before the introduction of crop UWA lenses.
I think you're wrong Jim and it's not mis-information.

Bob
 
Canon Bob said:
I think you're wrong Jim and it's not mis-information.

Bob

Why do you think I'm wrong? It's simple physics!

I bring you back to my other point. What's the difference with using a grad at 10mm on a crop and 16mm on FF?
 
Last edited:
Be amazed... Second FAQ

http://www.leefilters.com/camera/advice/faq/

However, it seems that some other places have read this as small format as in crop, whereas lee consider small format crop to 35mm. so its open to interpretation.

This is what it actually says "Hard Grads tend to be used on smaller format cameras (DSLR or 35mm) and are used for general landscape shots with a definite horizon. Soft Grads tend be used on larger format cameras or on very wide angle lenses and are used for balancing exposure evenly, when there is no definite horizon i.e. in cities or other situations when you may have an uneven skyline."

And I don't have any problems with that. It does not say that hard grads are better on crop format DSLRs, which is what folks appear to be reading into it here. It's confusingly expressed though.
 
Why do you think I'm wrong? It's simple physics!

I bring you back to my other point. What's the difference with using a grad at 10mm on a crop and 16mm on FF?
I'm just not seeing focal length anywhere in the equation Jim. I stick a grad on 24mm lens or a 135mm lens and I still have a range from 0EV to the maximum density of the filter. The reasoning I'm seeing from other people would seem to infer that a wider lens uses more of the grad and it just doesn't.
The light passing through the filter will affect the image circle (about 45mm diameter for Canon EF lenses) and that circle is the same size whatever the focal length of the lens producing it.

Bob
 
HoppyUK said:
This is what it actually says "Hard Grads tend to be used on smaller format cameras (DSLR or 35mm) and are used for general landscape shots with a definite horizon. Soft Grads tend be used on larger format cameras or on very wide angle lenses and are used for balancing exposure evenly, when there is no definite horizon i.e. in cities or other situations when you may have an uneven skyline."

And I don't have any problems with that. It does not say that hard grads are better on crop format DSLRs, which is what folks appear to be reading into it here. It's confusingly expressed though.

...and that enforces my point about focal ranges (the wider, the more effect).
 
Canon Bob said:
I'm just not seeing focal length anywhere in the equation Jim. I stick a grad on 24mm lens or a 135mm lens and I still have a range from 0EV to the maximum density of the filter. The reasoning I'm seeing from other people would seem to infer that a wider lens uses more of the grad and it just doesn't.
The light passing through the filter will affect the image circle (about 45mm diameter for Canon EF lenses) and that circle is the same size whatever the focal length of the lens producing it.

Bob

But angle of view does affect it. Try my example earlier, put a grad on a standard zoom at take 3 shots, one at it's widest, one in the middle of the range and another at the top of the range. You'll see the effect of the grad diminish as you zoom.
 
Back
Top