Struggling still with Landscape clarity and sharpness!

This might have been said already, if so apologies ... I have not ready the full thread.

Your scottish landscape shots on Flickr are all lacking in drama in the sky, a key landcape technique to master. As a result they also suffer from poor exposure balance between fore ground and sky, this can result in the lack of clarity and punch you feel is missing. You are doing the right thing by not only doing sunny day landscapes, just work on the exposure balance.

Others have posted the technical cures for these issues. Next time you are out thin drama and exposure balance.
 
jerry12953 said:
Strictly speaking I don't think a polariser does cut thro' haze. Haze is actually dirt/dust particles in the air - you get it generally during anti-cyclonic conditions, especially with east winds in the UK - and a polariser can't help. Haze is an absolute killer for long distance subject matter. What I think a polariser does is to cut out the scattered light which is present in an otherwise clear atmosphere.

I'm not sure if you're suggesting that it' isn't worth using hyperfocal distance. It's very useful technique when you need a very large depth of field. It's probably good enough for most situations to know the general principles.

A sun compass is handy but not absolutely necessary. If you remember that the sun is due south at 12 noon (GMT), due west at 6 pm (GMT) and due east at 6 am (GMT), you can work out when it will be at any time of the day. It moves around in a clockwise direction by 15 degrees every hour in the northern hemisphere.

I hope you don't think I'm nit-picking but this thread is progressing towards a kind of checklist for the landscape photographer, and it's worth getting everything correct.

jerry
Click to expand...

You're certainly not nit-picking and I absolutely agree that things must be right :)

And I'm certainly not saying that hyperfocal distance isn't worth using - I'm sorry if that wasn't clear. What I'm saying is that standards of sharpness have risen with digital in my opinion, and therefore the established parameters for acceptable sharp focus need to be tightened. I simply calculate hyperfocal distance, then add a stop to sharpen up the extremes of the hyperfocal range.

I find that using the hyperfocal rule, infinity focus isn't sharp enough. Of course, be definition it lies right at the limit of what is deemed acceptable sharpness. I have also read of some photographers taking this a step further by actually focusing on infinity, as this is usually where the most important detail is in landscapes, and then stopping down to get sufficient foreground sharpness. This makes no sense to me. It disregards the potential problem of diffraction (and sensor dust :) ) it promotes excessively long shutter speeds and the attendant risk of camera shake, and it is plain unnecessary.

I'm very happy with hyperfocal distance focusing, plus a bit for luck :)

PS Another tip for calculating the position of the sun - satellite dishes point almost exactly due south.
 
HoppyUK said:
PS Another tip for calculating the position of the sun - satellite dishes point almost exactly due south.
Click to expand...

A SKY satellite dish in the UK points to about 28 degrees east of due south.
 
I think I'll stick to hyperfocal focusing... Thanks for all the posts guys... like someone said this post has moved on somewhat!!!

What did I ask originally!!!! ??? ER....

Thanks again chaps....

I think I've reached a better understanding!

M
 
jerry12953 said:
I think I'll leave that for you to sort out, then, tdodd!

The way I see it if you focus on infinity there's all that lovely depth of field beyond it just going unused.....
Click to expand...

I have shot a series of photos with my 1D3 and 16-35 lens @ 16mm at full stop apertures from f/2.8 to f/22. I focused manually on the most distant object in the scene - an electricity pylon - using Live View and 10X magnification. The object was too small to isolate with AF but not a problem with Live View.

I have posted the results to an online album and have included all seven shots as....

- full frame images resized to 800x533;
- 100% crops of the focused subject (plus a large part of the rest of the scene);
- 100% crops of the grass "at my feet", which is actually about 3m away from me. The bamboo cane marks 3m from the sensor along the ground.

The images were shot raw and processed to JPEG in Lightroom with no edits - just standard Lightroom default settings. Here are the sample shots at f/4....

20090621_165750_8453_LR.jpg


20090621_165750_8453_LR-2.jpg


20090621_165750_8453_LR-3.jpg


The full album is here - http://picasaweb.google.co.uk/EezyTiger/LandscapeFStop. If you view as a slideshow you should be able to observe as parts of the images become sharper (or softer) as the aperture shrinks.

To my eyes the results at f/8 look best overall, but even at f/2.8 the extremes of near and far distance are far from complete mush. Viewing the entire scene shrunk to 800x533 I am hard pressed to find a winner or loser amongst them.
 
Realising that a near distance of 3m (~10') may not suit everyone's needs I collapsed the tripod legs to their shortest length and repeated the same shots with the same approach. This time I marked the close distance with the bamboo cane at 1.5m. The results are in a separate album, here - http://picasaweb.google.co.uk/EezyTiger/LandscapeFstopII

To my eyes f/11 just had the edge at the 1.5m distance, viewed at 100%, but f/8 is pretty close. Apart from f/2.8 there seems to be little to choose between the full images resized.
 
jerry12953 said:
I'm amazed that you have the time and inclination to do this, mate!
Click to expand...
Well I said I'd do it, and since the idea of the forum is to help each other I thought the results might be useful to others as well as me.

The figures at http://www.dofmaster.com/dofjs.html say that with a 1D3 and 16mm lens at f/8 the hyperfocal distance is 1.41m, giving acceptable sharpness from 0.7m to infinity. By focusing as I did, at ~300m, the field of acceptable sharpness is calculated to be from 1.38m to infinity, which was adequate for both my scenes. In fact, had I focused at 10,000m the near point of acceptable sharpness would have been 1.39m. Thus, by the calculations, it seems to have mattered little whereabouts I focused within a range from 1.41m to 300m or even infinity. That's a pretty flexible range. If there was something midway into the scene that I wanted to be as sharp as possible - a rock/tree/bridge/stream/building etc. - I may just as well have focused on that, since near and far elements would have been acceptably sharp regardless of where I focused.

FWIW I fired off a quick shot this morning, just at 16mm and f/8, this time focused at approx 3m. In terms of near and far field sharpness I could see no difference whatsoever between today's image and the equivalent from yesterday. Based on the calculations that comes as no surprise, but a visual confirmation does no harm.

Obviously at different focal lengths and different bodies the calculations do vary, so one size does not fit all, but so long as one picks a sensible choice of aperture and focus distance I doubt very much that those things will be the cause of poor clarity. Far more likely is the atmosphere itself, at a distance. Of course, the lens IQ, any filter attached, or camera instability could be factors, but my vote goes to the angle of the light and the quality of the air and how the light interacts with airborne particles. Heat haze is a curse. Dust is a course. Mist/moisture is a curse. Flat lighting is a curse.... If you are looking for clarity.
 
I suggest that with a focal length of 16mm, wherever you focus, almost everything else will be in focus as well, even (almost) wide open, as you have demonstrated.

It will be much more challenging when using a longer focal length. You'd need to be much more careful with your focussing then.

I agree that the atmospheric conditions you mention in your last paragraph are often the most critical when out in the landscape.
 
Don't forget that 16mm on a 1D3 gives the field of view of a 21mm lens on full frame, or 13mm on a 1.6X crop body, so while it is indeed quite wide, it is not insanely so. It is easily possible to shoot quite a bit wider than that. Had I shot at the 35mm end that would have been equivalent to 46mm on a full frame body, which is probably not your everyday choice for landscapes. I thought my choice seemed reasonably representative of a sort of middling focal length for landscape stuff.
 
Thanks Tim. Interesting, as always :)

I wonder how it would look with longer focal lengths. It is pretty hard to get anything out of focus with a really wide lens.
 
Oops! Double post.
 
Reading through this thread I have some observations....
1) you use a tripod and release, but do you lock up the mirror, and do you wait for the wind to drop.
2) You have been given some good advice on DOF But you must realise your lenses are diffraction limited on your sensor... at F11 or below.. That means in practical terms there is no advantage for you to stop down more than this. see this thread.
http://www.tawbaware.com/forum2/viewtopic.php?t=5527
3) apparent sharpness is greatly influenced by the light, cross lighting give the greatest advantage.
 
In reality.... does mirror lock up really make any difference? I'd love to see an example of two shots where someone can prove to me it does... because in any shots i've done I see no difference and with my camera the 5d it seems to take 30secs to go through the shot taking process on a 1/2 second exposure... this does seem excessive for the mirror lock up puropse...

can someone shed any light or am I being nieve?

Mark
 
I can't comment on the improvements or otherwise from using MLU but when seeking absolute sharpness, perhaps when checking AF accuracy, I use it combined with the 2 second self timer on my 50D and 1D3 and the mirror locks up as soon as I press the shutter button, then two seconds later the shutter trips. I can't think why it would add 30 seconds to the process on the 5D. Even long exposure NR should do no more than double the time of the exposure but that has nothing to do with using MLU, as far as I know.

If the 5D doesn't have a 2 second timer then even the 10 second timer should only add 10 seconds. If you have a remote release then obviously you can pause as long as you like between MLU and shutter release. Again, on the 50D and 1D3 MLU is an option I've set up on my favourites menu, so enabling and disabling it takes just a moment and no faffing around hunting for it.
 
tdodd said:
I can't comment on the improvements or otherwise from using MLU but when seeking absolute sharpness, perhaps when checking AF accuracy, I use it combined with the 2 second self timer on my 50D and 1D3 and the mirror locks up as soon as I press the shutter button, then two seconds later the shutter trips. I can't think why it would add 30 seconds to the process on the 5D. Even long exposure NR should do no more than double the time of the exposure but that has nothing to do with using MLU, as far as I know.

If the 5D doesn't have a 2 second timer then even the 10 second timer should only add 10 seconds. If you have a remote release then obviously you can pause as long as you like between MLU and shutter release. Again, on the 50D and 1D3 MLU is an option I've set up on my favourites menu, so enabling and disabling it takes just a moment and no faffing around hunting for it.
Click to expand...

Ah... maybe I should read my instruction book... are you saying when you use MLU and a cable release you need to press the button once to lock up the mirror and a second time to take the pic??? This may be my problem? as i've only been pressing it once and waiting... and waiting ... and waiting! In fact it was a while ago so i'll need to give it another go... Im sure its prob me... and I never did read the instruction booklet... just went for it!
 
Yup. Press once to raise the mirror. Press again to release the shutter. Self timer will take care of the double press for you, but if you need to capture a "decisive moment" relying on a timer release won't be much use. For a landscape, though, it's a nice easy option - unless you need to catch a break in between gusts of wind.
 
I think sometimes we all must be guilty of not reading our manual... feel such a twit now... what a dope! Will def be using mirror lock up now... even if it does or does not make a dif... I cant believe I sat there for 30 secs plus at times when all I needed to do was press again! Idiot!

M :bang:
 
Mirror lock up will help with expsures in the 1/15 second to a couple of seconds range (roughly).

At high speeds like 1/125 upwards and anything longer than a few seconds, it will not make a lot of difference.


Steve.
 
markelliot said:
3590228623_cf46920f45.jpg


This is an example...
Click to expand...

Sorry, if I've missed something in amongst this thread, have only done some skip reading, but there's some awful sharpening and jpeg artifacts in the 'original' version of these images. And as far as I am aware, the 'original' on flickr is as saved by you.
 
According to the Flickr page the photo was saved with a Photoshop Quality of 3, which if I'm reading that right does sound tremendously low.

If there is a raw original available it would be interesting to see what image the camera actually recorded. Maybe there is a workflow problem here rather than a picture taking problem.

There does appear to be a flare spot in the top left of the image and there is a lack of contrast the deeper into the image (further from the camera) you look. A spotless, filterless (or at least spotless and filtered) lens and a lens hood would seem to be wise - I don't know whether or not a hood was in use or how clean and shiny the lens/filter was.

Also, looking at the histogram, there are no blacks in this scene. The "clarity" (contrast) might be improved by darkening the shadow region, boosting contrast and increasing clarity. Other than that the hazy mountains look to be quite normal enough but editing could improve things.

Here's one of mine from 3 years ago, shot as a JPEG, firstly with no edits and then with the Blacks deepened, Clarity turned up a lot and Vibrance boosted.

20060706_124530_LR.jpg
20060706_124530_LR.jpg


Haze is a killer, but a bit of tweaking can help out. Nowadays I only shoot raw so I have freedom to edit without suffering the initial loss of the JPEG conversion in camera.
 
Gawd this is a good thread. I'm learning loads.

Andy
 
markelliot said:
I use the canon software as a raw editor... so thats a good start.

I tend to just apply the sharpness slider right away and just do it once... It's on a scale of 0 to 10.... so what your saying is I should initially slide to 2 and then PP... then sharpen it a little at the end...
Click to expand...

I use a value of 5

and then also sharpen in CS3 depending on the photo..with the smart sharpen option...
 
Steve Smith said:
Mirror lock up will help with expsures in the 1/15 second to a couple of seconds range (roughly).

At high speeds like 1/125 upwards and anything longer than a few seconds, it will not make a lot of difference.

Steve.
Click to expand...

^^^ Right (y)

I did some tests and found that between 1/15sec to 1/4sec were the worst speeds for mirror slap. Can be very bad in fact :eek: If you have a wobby centre column, it affects much longer speeds, too. As Steve says, I would avoid anything between 1/30sec to 1sec without mirror lock up.

If you want to see what mirror slap is all about, put a long lens on the tripod and pull up the centre column a few inches. Shoot at 1/8sec. Absolutely hopeless.

Using a tripod is no guarantee of shake free images.
 
Maybe I do have a workflow issue... Im not sure it's a strong point of mine as Im from a film background.... I am sort of self taught and to be honest have not even read much... just trial and error... and I'd rather be out taking pics than stuck in front of a comp... I get bored really quickly and always have a backlog of processing.

I save my photos in two formats. The first is a HQ jpeg at 10 quality no. Then I save a second using the save for internet function in photoshop. I use this file for flickr and forums etc as it loads up much faster. I use the full quality file for printing. I've dabled doing the bulk processing in photoshop too but off the top of my head cant even remeber how I did this! And I tried in Canon DPP but could not get it to compress the files as small as I could in photoshop for the web.

I think I had a poleriser on the lens for the Lake shot... so prob no hood. And it was a cheap jessops one... I've now upgraded to a hoya pro 1.

So any workflow ideas would be good!

Thanks again for your comments... been really helpful!

M




tdodd said:
According to the Flickr page the photo was saved with a Photoshop Quality of 3, which if I'm reading that right does sound tremendously low.

If there is a raw original available it would be interesting to see what image the camera actually recorded. Maybe there is a workflow problem here rather than a picture taking problem.

There does appear to be a flare spot in the top left of the image and there is a lack of contrast the deeper into the image (further from the camera) you look. A spotless, filterless (or at least spotless and filtered) lens and a lens hood would seem to be wise - I don't know whether or not a hood was in use or how clean and shiny the lens/filter was.

Also, looking at the histogram, there are no blacks in this scene. The "clarity" (contrast) might be improved by darkening the shadow region, boosting contrast and increasing clarity. Other than that the hazy mountains look to be quite normal enough but editing could improve things.

Here's one of mine from 3 years ago, shot as a JPEG, firstly with no edits and then with the Blacks deepened, Clarity turned up a lot and Vibrance boosted.

20060706_124530_LR.jpg
20060706_124530_LR.jpg


Haze is a killer, but a bit of tweaking can help out. Nowadays I only shoot raw so I have freedom to edit without suffering the initial loss of the JPEG conversion in camera.
Click to expand...
 
68lbs said:
Sorry, if I've missed something in amongst this thread, have only done some skip reading, but there's some awful sharpening and jpeg artifacts in the 'original' version of these images. And as far as I am aware, the 'original' on flickr is as saved by you.
Click to expand...

You have lost me a bit with your coments... can you explain a little better... how do you mean by awful sharpening efects? Original on flickr was saved by me and not sure what your on about here to be honest!
 
HoppyUK said:
^^^ Right (y)

I did some tests and found that between 1/15sec to 1/4sec were the worst speeds for mirror slap. Can be very bad in fact :eek: If you have a wobby centre column, it affects much longer speeds, too. As Steve says, I would avoid anything between 1/30sec to 1sec without mirror lock up.

If you want to see what mirror slap is all about, put a long lens on the tripod and pull up the centre column a few inches. Shoot at 1/8sec. Absolutely hopeless.

Using a tripod is no guarantee of shake free images.
Click to expand...

Thanks for this guys... I've made myself a note in my we camera bag notebook to remeber this... will deff adopt the 1/30sec to 1 sec in future. Will do a few test shots when I have a chance... no doubt prob wont noltice but will see anyway... and always do it as a matter of course! Also are we def saying no point if its an exposure longer than 1 sec...
 
Mark, referring to comments made by 68lbs, if you look at your Original sized image online, first look at the sky in the top left hand corner. You will note it is comprised of horrible square splodges, kind of like a mosaic. That is a problem with excessive compression of the file. Making the file too small comes at a price, and the price is IQ.

If you now scroll down to the horizon you will note that the very top edge of the land has a dark line all the way along it - like lip liner or eye liner. In the sky immediately above the land there is an odd sort of halo in the few pixels above the horizon. The excessive darkening of dark edges (the land) and the halo in the sky is a clear indication of excessive sharpening. Sharpening works by increasing contrast where edges meet, by darkening dark things and brightening light things. The high levels of JPEG compression have quite possibly exacerbated the sharpening effect, making it very obvious and not very appealing.

Regarding workflow, if you are shooting landscapes I doubt you are churning out hundreds of images an hour, unlike a sports or action photographer. For starters I recommend you shoot raw instead of JPEG. That will give you far more control over fine tuning your images, sorting white balance, adjusting tones and sharpening precisely as you would wish. You will also have a lot more dynamic rnage to play with, with the ability to pull back blown highlights that would be lost forever in a JPEG from the camera. Once the camera spurts out a JPEG file a lot of those things are set for you. Sure, can can make some changes, but some things will have been "baked" into place never to be undone. At the very least, give raw a go. If you are nervous then by all means shoot raw + JPEG and then at least you can see whether raw offers you an advantage. Use a calibrated monitor for editing and pay attention to the histogram, both when shooting and when editing, as well as your eyes.

FWIW I started shooting raw within a week of getting my first DSLR, three years ago. I have not shot to JPEG since. I used to use DPP for editing and conversion to JPEG but have migrated to Lightroom ever since version 1 was released. I use Lightroom for all my image file management, editing, conversion etc.. I have Photoshop but hate it so much that I do not even have it installed.
 
Can I ask what tripods you all use. I have a velbon Sherpa with a silk pan and tilt head for landscapes. I also use a silk AM340 for times when I need to travel light. I like this but I know it is a comprimise but the best I could find for travelling. I feel I need to replace the velbon though... for something a bit heavier... any suggestions? I was thinking a manfrotto 190
 
I use a Manfrotto 055MF3 (£200) and Manfrotto 468MGRC2 head (£175).
 
tdodd said:
Mark, referring to comments made by 68lbs, if you look at your Original sized image online, first look at the sky in the top left hand corner. You will note it is comprised of horrible square splodges, kind of like a mosaic. That is a problem with excessive compression of the file. Making the file too small comes at a price, and the price is IQ.

If you now scroll down to the horizon you will note that the very top edge of the land has a dark line all the way along it - like lip liner or eye liner. In the sky immediately above the land there is an odd sort of halo in the few pixels above the horizon. The excessive darkening of dark edges (the land) and the halo in the sky is a clear indication of excessive sharpening. Sharpening works by increasing contrast where edges meet, by darkening dark things and brightening light things. The high levels of JPEG compression have quite possibly exacerbated the sharpening effect, making it very obvious and not very appealing.

Regarding workflow, if you are shooting landscapes I doubt you are churning out hundreds of images an hour, unlike a sports of action photographer. For starters I recommend you shoot raw instead of JPEG. That will give you far more control over fine tuning your images, sorting white balance, adjusting tones and sharpening precisely as you would wish. Once the camera spurts out a JPEG file a lot of those things are set for you. Sure, can can make some changes, but some things will have been "baked" into place never to be undone. At the very least, give raw a go. If you are nervous then by all means shoot raw + JPEG and then at least you can see whether raw offers you an advantage. Use a calibrated monitor for editing and pay attention to the histogram, both when shooting and when editing, as well as your eyes.

FWIW I started shooting raw within a week of getting my first DSLR, three years ago. I have not shot to JPEG since. I used to use DPP for editing and conversion to JPEG but have migrated to Lightroom ever since version 1 was released. I use Lightroom for all my image file management, editing, conversion etc.. I have Photoshop but hate it so much that I do not even have it installed.
Click to expand...

Since taking that shot... and for about 6 months I have been shooting RAW and do notice a huge improvement...

I'd love a simple checklist on how to PP RAW images for landscape in Canons DPP... then how best to save the image for the internet so it's compressed but at the same time not mess its IQ too much so I get bad shots.
 
HoppyUK said:
I did some tests and found that between 1/15sec to 1/4sec were the worst speeds for mirror slap. Can be very bad in fact :eek: If you have a wobby centre column, it affects much longer speeds, too. As Steve says, I would avoid anything between 1/30sec to 1sec without mirror lock up.
Click to expand...

Just to expand on this:

At high shutter speeds, mirror slap is not really an issue as the speed is fast enough to capture the image before it moves too much - just like hand holding.

At slow shutter speeds over about two seconds it is not much of an issue either. this is because the time the camera is vibrating is a very small percentage of the time the shutter is open. The longer the time, the less significant it is.


Steve.
 
tdodd said:
I use a Manfrotto 055MF3 (£200) and Manfrotto 468MGRC2 head (£175).
Click to expand...

Thats allot of money fro a tripod combo... but the more I think about it... and for the type of images I want to take it is probably money better spent than comprimising and then spending money on more lenses... I do want a 70-200 and a 17-40mm but prob best upgrading the tripod first!
 
Steve Smith said:
Aaagghhh!!!!

Internet spelling!!

a lot (two words!)




Steve.
Click to expand...

I cant spell to save myself... Im an engineer give me some maths to do... but dont ask me to spell! And not really after spelling advice anyway... wont do much for my photos!

M
 
You must log in or register to reply here.
Back
Top