A few of my shots have that problem. I have not been able to fathom the reason for this. Any ideas?
Canon 5D MkII,EF24-70mm.f2.8LII Shot at f22 1/250 ISO 400
That's a lot of noise.
Definitely ISO400? Yep.
In the exif data it says no Colour Representation, I would have thought it would show sRGB. Do all your pics show that?
Ken
freddayan wrote:
A few of my shots have that problem. I have not been able to fathom the reason for this. Any ideas?
Yes that is a lot of noise.
And ugly Canon crosshatch noise to boot.
Looks equivalent to about what I would expect from a reading of ISO1600 - 3200 for a daylight shot to me.
I don't know why it has done it but F/11 would have given you ISO100.
Actually all my pictures have Adobe RGB (1998) setting. Also ISO 400 does not generate noise in almost any pictures.
I rechecked a large number of pics taken under the same circumstances- no noise. I am truly baffled!
Could it be a problem with your card?
Original is 14.7Mp, ISO 400, 1/250-sec at f/22. Quite curious.
If it were, it would be universal, I surmise...
freddayan wrote:
If it were, it would be universal, I surmise...
Fred, do you ever use live view for any of these? Just making sure as that would introduce noise.
I quite frequently get at least some noise if it is overcast or in the shadows as you are. And yes, it would help if you put the lens in its sweet spot and lowered your ISO. Thats not an answer but could help in the future.
If it happens a lot, try putting everything to defaults, and see if it doesn't change. Good luck.
I'm guessing you metered off the sky and brought the shadows up in PP. Anytime you under expose you will get noise or pixilation to some extent depending on how under exposed the photo is, my guess at least.
fstop22 wrote:
I'm guessing you metered off the sky and brought the shadows up in PP. Anytime you under expose you will get noise or pixilation to some extent depending on how under exposed the photo is, my guess at least.
You could be right, I was answering as if it was an unprocessed pic.
So you have raised a good question.
Have the shadows been brought up in post?
Yes I did! Thank you I shall look out for this in the future.
I used Aperture to PP. BTW where is a lens' sweet spot?
freddayan wrote:
where is a lens' sweet spot?
Usually, 2-stops to 3-stops above wide open, or so. For many lenses, that is f/5.6 or f/8.
Here is the Modulation Transfer Function (MTF) chart for my Nikkor 105G macro lens.
http://www.normankoren.com/Tutorials/MTF.html
freddayan wrote:
Yes I did! Thank you I shall look out for this in the future.
I used Aperture to PP. BTW where is a lens' sweet spot?
Fred, when you reply, hit the"quote reply" button on that posters post, so we know who, or what you are responding too.
But if it's in regard to the, "live view(LV)", every time you turn it on, the sensor is exposed and develops heat, which introduces noise. Leave it on only long enough to focus, then turn it of.
Your meta data should show if you were using the LV on those particular, problem shots. Hope that was the problem. Good luck.
That image looks over-sharpened to me. Sharpening will increase noise.
freddayan wrote:
A few of my shots have that problem. I have not been able to fathom the reason for this. Any ideas?
My guess is that the shadows were underexposed. The subjects are in open shade and a (very rough) guess would be 1/400 at f 5.6-8 would have been more reasonable.
That way you could bring the highlights down rather than the shadows up. I'm fairly certain you'd still have detail in the sky with that approach using raw.
Lowering the iso won't help much with this particular issue but I'd likely have shot at 100-200 anyway (which would change the exposure to 1/iso in the recommendation above - which is just based on sunny 16 with a couple of stops for open shade). Center weighted or focus point linked metering might have worked better for you to get the exposure specifically from your subjects.
If you want to reply, then
register here. Registration is free and your account is created instantly, so you can post right away.