Bought 7D for Low Light Performance

Nick D

Well-known member
Messages
182
Reaction score
0
Location
Eastern, PA, US
.......... And have not been disappointed. I also wanted a crop body not a FF. that is why not a 5D II ( in case anyone wondering), also like the AF system, and 8fps. This camera offers a lot of versatility that i like. Yes I would love a 5D but need to build up glass collection a bit first.

17-55 f2.8 @ 17mm, jpeg original, manual mode,f8, 1/25, iso 4000, +1EV in LR, NO Flash, No NR other than in camera "Standard" setting



100-400L @100mm,jpeg original with in camera standard nr, f4.5 1/125sec. iso 12800 Av, -1/3EV, no flash



10-22 @ 15mm jepeg orig. with in camera standard nr only, f5.6, 1/13, iso 6400, no flash



--
Nick D

nickd224.smugmug.com
 
I've been pleased with mine, too; ISO 3200 has quite low noise and 6400 is usable, I think at least as good as 1600 on my 450D. I haven't tried 12,800 yet but I think, judging from 6400, that it wouild produce noisy JPEGs.

FF
 
The middle pic, of my dog, is at 12,800 with a bit of shaddow area. to my eye it looks much better than my 50D at 6400.maybe even comparable to 3200.

I am happy that i do not see banding noise which was horrendous with my E3 and evident with my 50D. the 50D was nice if exposed properly but these images, particularly the last 2 were shot near the end of daylight on a cloudy, rainy day, opposite were sunlight was entering the house. It had all the earmarks of other times when i failed miserably with other cameras. So i do not care if it was me or camera. this 7D has me very satisfied.
--
Nick D

nickd224.smugmug.com
 
noise control is great, but how do you find the autofocus in different modes in low light? my experience is a bit of a hit and miss.. just like my 450D was. any similar observations?
 
The AF systems never never designed to operate without assist in low light. They having done any magic to make the 7D work better in low light than previous models.

You'll also find some lens work better than others. Namely my 50 f1.8 works better than the f2.8 lens I have. Of course it lets in more light so that's why :)
--
http://www.flickr.com/photos/sebastianip/
 
Having played with the 5DMkII and a 24-70 L F2.8 combo at the Singapore Night Safari while my missus used the trusty 50D with the 17-55mm F2.8 I would say you just might be dissapointed if you moved to the 5DMkII.

It's a great camera, don't get me wrong on that at all, but for AF in low light, the 17-55's IS helps the AF work well. You cannot use AF assist there, it's all just natural shooting. I was having no end of difficulty with AF, and due to the very low contrast/low light, wasn't able to manually focus worth a damn either. Yet the missus was getting between 8-10 shots for every one I got.

We investigated further, and when the IS was turned off, we both struggled. So give me that lens over the 50 primes any day (Including the 50mm F1.2 I took caving). IS really helps stabise the image for AF to work.

On the bright side, in the next few Months, rumour has it that the 24-70mm L F2.8 is being released in IS version. This should be a killer lens for the 5DMkII, and would put things back in it's favour. But until it surfaces, stick with what you have got. Shot for shot, if we both got the shot, the 5DMkII had less noise, but on the ones I manually focussed, most lacked crisp focus so after NR the 50D shot was the keeper still.
--
The Aussie Viking
 
even though they are fine examples of high ISO (and I like what I see a lot), the room in the third picture is fantastic. So cozy and warm,makes you want to be there and relax.
Rgds
 
... not really a good indicator of high iso performance. The main problem at high iso is, of course, high noise. Resizing an 18MP image down to 1 MP (or so) for the web eliminates most of the visible noise. Any camera looks good if you shrink the image down enough. It would be more helpful to have posted 100% crops.

But then we would miss seeing your terrific photos!

--
JerryG

My galleries at:
http://www.pbase.com/jerryg1
 
... not really a good indicator of high iso performance. The main problem at high iso is, of course, high noise. Resizing an 18MP image down to 1 MP (or so) for the web eliminates most of the visible noise. Any camera looks good if you shrink the image down enough. It would be more helpful to have posted 100% crops.

But then we would miss seeing your terrific photos!

--
JerryG
True, but how often in real life do you print or view pics at 100%?

"Any camera looks good if you shrink the image down enough."

Not true for high ISO, for one, many cameras , P&S, do not go high at all.

Rgds
 
... not really a good indicator of high iso performance. The main problem at high iso is, of course, high noise. Resizing an 18MP image down to 1 MP (or so) for the web eliminates most of the visible noise. Any camera looks good if you shrink the image down enough. It would be more helpful to have posted 100% crops.

But then we would miss seeing your terrific photos!

--
JerryG
True, but how often in real life do you print or view pics at 100%?
That's not the point. It makes no sense to mask the noise by a 20:1 size reduction, and then try to judge how good or bad the noise is. At least a 100% crop, or possibly a crop after reducing to 15MP (for comparison with the 50D), or 10MP (for comparison with the 40D) would still allow judging the qulity and quantity of noise compared to other cameras.

Besides, how often do you use an 18 MP camera to display a 1MP image? Any thing more than 3-5 MP is overkill for web displays' On the other hand, 18 MP makes lots of sense for 12" x18" and larger prints, and a noisy file will show at those sizes.
"Any camera looks good if you shrink the image down enough."
Not true for high ISO, for one, many cameras , P&S, do not go high at all.
OK, I may have overstated that a bit.

Here's an example of what I'm talking about. I used my wife's SD1200 at ISO 1600 to take indoor pictures of a speaker at a photo club function. Prints are generally awful at ISO 1600 from this camera, but the 2" x 3" web pictures that they were used for looked beautiful. You could not tell they were from a tiny-sensor elph. That's because the 10 MP images were scaled down to about 1/20 MP.

Good night.
--
JerryG

My galleries at:
http://www.pbase.com/jerryg1
 
... not really a good indicator of high iso performance. The main problem at high iso is, of course, high noise. Resizing an 18MP image down to 1 MP (or so) for the web eliminates most of the visible noise. Any camera looks good if you shrink the image down enough. It would be more helpful to have posted 100% crops.

But then we would miss seeing your terrific photos!

--
JerryG
True, but how often in real life do you print or view pics at 100%?
That's not the point. It makes no sense to mask the noise by a 20:1 size reduction, and then try to judge how good or bad the noise is. At least a 100% crop, or possibly a crop after reducing to 15MP (for comparison with the 50D), or 10MP (for comparison with the 40D) would still allow judging the qulity and quantity of noise compared to other cameras.

Besides, how often do you use an 18 MP camera to display a 1MP image? Any thing more than 3-5 MP is overkill for web displays' On the other hand, 18 MP makes lots of sense for 12" x18" and larger prints, and a noisy file will show at those sizes.
"Any camera looks good if you shrink the image down enough."
Not true for high ISO, for one, many cameras , P&S, do not go high at all.
OK, I may have overstated that a bit.

Here's an example of what I'm talking about. I used my wife's SD1200 at ISO 1600 to take indoor pictures of a speaker at a photo club function. Prints are generally awful at ISO 1600 from this camera, but the 2" x 3" web pictures that they were used for looked beautiful. You could not tell they were from a tiny-sensor elph. That's because the 10 MP images were scaled down to about 1/20 MP.

Good night.
--
JerryG
Hello Gerry,
in essence , I agree with you.

If you need to judge true noise performance, large files are the only way to go. I was making a rushed mental shortcut, some images are so good for different reasons that technical aspects mean not that much. Like here, I like the room so much, the light, colour, I could not care less abt noise. In images posted, at the size posted, there is not much noise to speak of.
Peace.
Rgds
 
Look at the mantel in the third photo.

FF
 
... not really a good indicator of high iso performance. The main problem at high iso is, of course, high noise. Resizing an 18MP image down to 1 MP (or so) for the web eliminates most of the visible noise. Any camera looks good if you shrink the image down enough. It would be more helpful to have posted 100% crops.
I'm not sure his point was to be "helpful."

He was displaying the images how he displays them.

A lot of photographers display a lot of images at relatively small sizes over the web. Of we print 4x6s. or even 8x10s.

In other words, we do a lot of resizing.

A lot of us don't care about a technical evaluation of what a camera is capable of. We care what the images look like, as we use them. And that's what I took these photos to be representative of: good examples of what this camera can do in practical use.

And yes, you can see the difference between these images and similar images taken with a 50D.
 
I just shot a Canine Obedience Trial yesterday and used my 7D. They don't allow flash, so I didn't want to use a 5D2 for this, as I needed lots of reach. My rationale was that 70-200 2.8 IS at 7D's 1.6 crop at 3200 will get me a faster shutter than a the 100-400 4.5-5.6 at 5D2's 1.0 crop at 6400. (The 300mm prime would not have worked because I needed shorter focal lengths as well and with a ton of action in three separate rings separate cameras just didn't work.)

I started at 3200 and then pushed it to 4000. Got home, pixel-peeped, and was stunned. There's noise at full size, but it's soooo incredibly fine and film-like that it wasn't an issue. And that's a full size. Even at 8x12 you don't see it. And for orders that have already come in, I have found that putting a layer of noise reduction over it, and then changing that opacity to 50% to hold some additional texture from the original layer, it's even better.

That's a roundabout way of saying that this was my first chance to put the high ISO noise factor to the test on the 7D and I was pleased beyond belief with the results.
 
...

A lot of photographers display a lot of images at relatively small sizes over the web. Of we print 4x6s. or even 8x10s.

In other words, we do a lot of resizing.
...
I don't disagree with those points, but the OP's title and first sentence implies that his images demonstrate the low light performance of the camera. In fact such seriously resized posts don't do that at all. Those small images probably don't look different noise wise then similar sized images would look from the 50D, 40D, and possibly 30D. I'm not saying the the 7D images aren't superior to the earlier cameras; I am saying that resizing them makes it hard to see the performance he is so pleased with.
A lot of us don't care about a technical evaluation of what a camera is capable of. We care what the images look like, as we use them. And that's what I took these photos to be representative of: good examples of what this camera can do in practical use.
A resized web post is about the least challenging "practical" use I can think of for a camera as sophisticated as the 7D. Impress me instead with a 16 x24 inch print ;)

--
JerryG

My galleries at:
http://www.pbase.com/jerryg1
 
look at his faboulus photographs when he kriss crossed America with a small camera...does anybody care about the noise...I was doing film at 3200 ISO some 25 years back...well, these pictures indeed were noisy, but I didn't care either ;-)
good pics, go ahead...thx for sharing
--
http://www.flickr.com/photos/guatitamasluz/
 
Hi all again and thanks for comments. i have been away from here and well i feel absolutely stunned. i took those at that size. i did not even realize that. but i will go through and post truer pics pic



here was one at 4000 iso with 17-55. this is the largest res. smugmug lets me copy down. it was taken in large jpeg mode, standard noise red.

Sorry to have unintentionally mislead anyone. I did same thing to me. but i still have other large jpegs where i am happy with image. will redo testing and post.

--
Nick D

nickd224.smugmug.com
 

Keyboard shortcuts

Back
Top