Hi Abhimanyu
The exposure looks OK to me on this shot but the main problem is that it is not sharp. I'm not sure whether you used a tripod but if not there could be camera shake with the 1/4 sec exposure used (according to the EXIF info). If you did use a tripod, then the problem could be that the autofocus hasn't worked properly in the lowish light conditions. I would use manual focus under these circumstances and probably would have made the bridge the focus point. You've got plenty of depth of field anyway.
It's a bit hard to tell what the noise is like with the web size image but according to the EXIF info you've used an ISO setting of 800 (possibly to get as fast a shutter speed as possible ?). With a tripod the shutter speed is not important and you could use an ISO of say 100. An ISO setting of 800 on a small sensor camera such as the SX50 would be a bit noisy I think.
Finally I would consider applying a bit of perspective correction if you have suitable software. This will help straighten up the buildings on the left.
Hope this is of some help
Dave
I think that is an very helpful critique that Dave has provided and I would agree with what he has written.
I would add that this picture does, in my view, demonstrate, you ability to 'see' a picture; i.e. to identify a scene that will make a good picture. It is a good composition and makes very good use of light and colour. All that is needed now is the acquisition and application of those additional knowledge and skills such as referred to by Dave and you will be on the way to making very good images.
It's also worth mentioning that if you are using a tripod, make sure it's weighted down enough during long exposures to account for movement by breezes.
Hey Abhimanyu,
If it were not for the camera shake this would have been a stella shot! Well done and shows you have an eye for composition and also know how to expose an image which is a GREAT start! The technical stuff you can fix as you go, but the developing an eye for what makes an image a good image is the hard part so you are more than half way there!!!
Look at the verticals (edges of the buildings) and you will see them leaning into the frame. If you were to move the exact centre focus point level with the horizon you may have ended up with more or less vertical lines (along with some inherent distorsion of the lens of course). I think Colin may have adjusted the distortion with his redo of your image. You can see that the vertical edges are now more upright/vertical in his version, which indicates to me that a pro thinks about things like that so maybe we should too
Here are some excellent links that I found helpful:
Using and Chosing a Tripod
Using Wide Angle Lenses
Avoiding Camera Shake
Well, done and happy shooting
Basically ...
If one doesn't have a tilt/shift lens then for this kind of shot one needs to make sure the camera sensor is perpendicular to the ground plane so that the building verticals don't converge or diverge - and sometimes this means having to use a much wider angle lens and then crop the image afterwards.
Hey Colin,
So by perpendicular I take that to mean around 90 deg to the ground surface (or at least parrallel to the verticals in the scene if the ground is particularly steep). When it comes to taking wide and cropping are we simply removing the most distorted portion of the lens in this way or have I missed the point? (i.e. I say that because a sensor plane can be orientated to perpendicular regardless of lens FL)
Last edited by Hans; 6th April 2013 at 12:20 AM.
Sorry, I should have worded that a little better. Parallel to the verticals is what's needed - that'll also be perpendicular to the ground if the ground is level (but only if the ground is level) (assuming the buldings aren't sloping!).
In many cases - if folks keep the sensor parallel to the verticals - they'll be cutting off the tops of the tall buildings. So usually what people do is "point the lens upwards" to get everything in the shot - and that in turn gives the issue with converging parallels (ie "sloping buildings"). (on a side note, it's not toooooooo bad with tall buildings, but it's vitally important with interior wide-angle photography)When it comes to taking wide and cropping are we simply removing the most distorted portion of the lens in this way or have I missed the point? (i.e. I say that because a sensor plane can be orientated to perpendicular regardless of lens FL)
With a tilt & shift lens you can get the sensor parallel and then shift upwards to get what you want in the field of view - but if you DON'T have a tilt & shift lens then you need to select a lens that has a field of view that's wide enough to get the building tops in - which in turn will also be wide angle the other ways as well and probably be close to having you feet in the shot - so the extra area needs to be cropped out.
A third alternative is to point the lens up - get the distortion - and then fix it in Photoshop, but that means choosing 1 of 2 approaches, both of which lose data;
(i) Use the perspective correction tools (that have the downside of cropping in from the sides losing FoV at the bottom), or
(ii) Using a transform that keeps the bottom in tact, but progressively lowers the resolution as you get closer to the top (you effectively stretch the top horizontally and then crop off the excess) (which is what I did to fix the OP's image in my quick makover).
That makes perfect sense, thanks again mate
With regard to camera shake, and Colin showed what the use of an editor can do for one with marginal shots, in a large city and on a relatively flexible bridge you need to consider traffic vibration and perhaps wait for the traffic lights to halt the vehicles. A situation where a tripod could produce a worse result than a careful hand held with image stabilisation that the camera has.
There is another approach in that your camera has manual control over shutter and aperture and if traffic is likely to be a serious problem it could be worth while choosing to shoot at 1/15 shutter or even 1/30 where OIS should be more able to give you a sharper shot while 'in-camera' you would get a dark result this could be adjusted in editing. even the free Paint.Net could do it, to achieve the result you have but sharper[ Here one uses Curves or levels rather than the brightness/contrast tools]. PN also has a perspective tool to correct the verticals, on both sides of the picture. It is always worth considering what the companion tools of camera and editor can do for us in difficult situation such as this. Being aware of the shutter and aperture settings the camera is picking for us when we take half trigger prior to shooting the picture. If you are not aware of this technique it is quite possible that you pressed the trigger and then relaxed before the camera took the shot resulting in the softness. Modern cameras are much faster with this but there is still a measurable delay as the camera works out what settings to use.
If it were me, I'd have shot it F11 @ 30 seconds @ ISO 100 on "old faithful" (my Gitzo 1548 tripod with RRS 58 Ball head).
The OP's shot was 1/4 - normally IS / VR won't give you much improvement at those low speeds, although another technique if one REALLY want to use shutterspeeds like that is to just shoot a burst of shots and then get the software to align them later.
This was a 5 MINUTE exposure from a SWING BRIDGE ... how's that for stabilisation?