Just a Blue Tit but manual exposure used this time

Messages
5,491
Name
Nigel
Edit My Images
Yes
Tim's (TDodd) advice spurred me on so I used manual exposure on this one. Had to pull back the exposure a tad. Practice makes perfect I guess. :bonk:

C&C very welcome

4601646516_b96a032108_o.jpg
 
Looks good to me!

I can honestly say that I never liked to shoot manual, but nowadays I feel more in charge with the camera in manual, and it's giving me some nice results here and there ;)

Definately worth trying the more you learn (y)
 
Hi, Nigel. I've only just stumbled upon this thread. It looks like you have clear blue skies and bright sunshine lighting your subject. As a guideline an exposure based on the Sunny 16 rule/guideline would be a good place to start. However, your EXIF shows an exposure 2 full stops brighter than a Sunny 16 equivalent. I'd be interested to know how you arrived at the exposure you did. I'm not surprised you had to pull back a bit (quite a bit?).

Apart from the little hiccough on exposure it is a really nice composition. (y)
 
Hi Tim. This seems like ages ago now, but I remember the conditions well and also my thought process. Although the bird was in full sunlight, I was sat in the shade, semi hidden. There were no nice white clouds to take a meter reading from as I was hoping to set these at +2 or +3 and hence these would be the brightest thing in my image.

I did try a couple of test shots with the bird in some further away branches using the sunny 16 rule but these looked a little dark. I then tried to ETTR but I cant remember the histogram in camera showing any hot spots so i kept the settings as per the exif. In DPP, there are no highlight warnings showing with the CR2 file

The pic in Camera Raw needed the exposure pulling back by one stop to eliminate the warnings however. It was then on with my usual PP.

I dont know if any of this helps, but your thoughts would be very welcome as always
 
When you say the images looked a little dark, using Sunny 16, were you making your judgement based on the appearance of the image on the back of your camera, or the histogram, or something else? Relying on the image itself is a very unreliable way to check exposure accuracy. The ambient light levels can cause your pupils to open up wide or stop down, but the brightness of the image displayed by the camera does not alter to compensate. Sat in the shade your eyes might open up and make the image look bright, but if you were looking out towards a brightly sunlit scene then they would stop down and make the image on the camera look dark.

I would also add that while Sunny 16 is a good starting point for establishing the correct exposure, the sun does look to be well off to one side, so you might need to sneak the exposure just a little higher, maybe by +1/3 above Sunny 16. Also, at 06:10 in the morning maybe the sun was still a little below full strength, having risen less than one hour earlier, which might account for another 1/3 stop or so. So, Sunny 16 + 1 might be pretty good for the conditions. Going for +2 seems like a lot.

As for ETTR and histograms, most of the tones in the scene are not that bright. The white feathers on the cheek are the danger zone, and the area likely to clip first, but that area is so small in the frame that you would not spot a problem from the histogram. If you had enabled highlight clipping warnings then I would expect to see those flash if you pushed the exposure to the limit, and to flash aggressively if you pushed too far. Do you have highlight warnings enabled in the camera? I recommend you do.

If you want me to take a closer look then I'll need a copy of the raw file, but it sounds to me like you were nudging the exposure to where you thought it should be, possibly based on duff information, or information you could not see, rather than based on any accurate metering technique. There's no harm in taking that approach, per se, but if it led you to make a wrong exposure then I guess the technique needs some work. :)
 
Thanks for your thoughts Tim. I have highlight warning enabled, but dont seem to remember the cheek feathers showing as clipped, but I may be wrong, and probably am. The "dark" comment would have been based on the histogram, as I rely on histogram info all the time

It would be good if you could take a look at the raw file to see what you think. Its a 25Mb file, so i'm wondering how I can get it to you :shrug:
 
If you don't have your own hosting webspace to which you could FTP the file I think the easiest thing would be to upload it to rapidshare.de or yousendit.com and then PM me the download link.
 
Nige, I've had a look at the file. In DPP you do have some red and green channel clipping on the bird, both on the white feathers and yellow. Unfortunately the highlight warnings in DPP do not light up unless you have all three channels clipped, which is more than a little stupid, but if you drag the cursor over various areas of the bird you will see the individual RGB values for each pixel. If you look at the RGB tab you will see from the RGB histogram that you are a bit lively in the red channel....

20100712_220000_.JPG


In Lightroom I'd say you've got a pretty much perfect ETTR exposure there, with tiny amounts of clipping once again, but the bird does look overexposed and a -1 decrease in exposure is probably closer to a realistic rendition, and of course, shooting ETTR is not the same as shooting for the finished product, so I'd say you were pretty much bang on with that exposure. Certainly I'd be happy with it.....

20100712_220524_.JPG
 
Thanks for your help Tim. It's good to know I'm on the right track, but practice will I hope get me a bit closer to perfect. The only thing i cant seem to be able to do is to get the pixel readings when I drag the curor over the pic. Can you give a bit more detail as to what I need to do to see this? Thanks
 
When you are editing an image you will see pixel RGB values displayed in the bottom left of the screen when your cursor is somewhere over the image. There are also X,Y coordinates displayed for the cursor position. You will see in ths example that with the cursor/mouse positioned over the brightest area of yellow that both the red and green channels are at 255, but that is not enough to trigger a highlight warning.

20100712_225235_.JPG
 
Last edited:
Back
Top