Any low light performance gained?

For those with any experience, would I be gaining any low light ability by going from a 60D with a Sigma 17-50 f/2.8 to a 6D with the 24-105 f/4? Obviously the 24-105 is one stop slower, but does the low light performance of the 6D offset that one stop at all? Thanks for any thoughts.

A lot depends on your tolerance for noise. The 6D will be at least 1 stop better than the 60D, if not two. But much of that gain is the processing of .jpg so if you use raw it may not be 2 stops better. So you arent going to be gaining much giving away a stop in aperture.

dan.k78 wrote in post #15635758
For those with any experience, would I be gaining any low light ability by going from a 60D with a Sigma 17-50 f/2.8 to a 6D with the 24-105 f/4? Obviously the 24-105 is one stop slower, but does the low light performance of the 6D offset that one stop at all? Thanks for any thoughts.

What issues are you having currently? AF in low light? Trying to deal with noise in low light?

There might be things we can point you toward if the latter so that you don't have to change up equipment, but instead change how to you post process.

Not really having issues, I was just curious if the noise performance of the 6D offset the 1 stop in aperture between those two lenses.

dan.k78 wrote in post #15635758
For those with any experience, would I be gaining any low light ability by going from a 60D with a Sigma 17-50 f/2.8 to a 6D with the 24-105 f/4? Obviously the 24-105 is one stop slower, but does the low light performance of the 6D offset that one stop at all? Thanks for any thoughts.

I went from a 7D with a Sigma 17-50 to a 6D kit, to answer your question... yes there is a big performance gain due to the 6D's excellent high ISO performance. ISO 12800 is probably cleaner than the 7D at 3200.

From a back-of-the-envelope physics point of view, the larger area of the 6D sensor ought to give you 2*log2(1.6)=1.36 stops of improved iso performance. Factoring in the improved QE of the 6D sensor (50% versus 40%) increases this difference to 1.68 stops. (This estimate makes certain assumptions about the nature of the noise limiting the iso performance.)

Based on DxO's actual sensor tests, in which they measure the ISO at which the camera hits some arbitrarily defined signal-to-noise ratio, the difference is 1.53 stops.

So, even after giving up a stop of aperture, the 6D ought to give around a half-stop of improved iso performance over the 60d. However, that assumes that the two lenses in question have identical light transmission characteristics (i.e., the difference in t-stops of the two lenses is the same as the difference in f-stops). This might not be the case, and I don't know what the t-stops are for the lenses in question.

All of the above ignores differences in the jpeg engines of the cameras if you're shooting jpeg rather than Raw. I can tell you from experience that the in-camera jpeg engine of the 6D applies some nasty noise reduction at very high ISOs, losing lots of detail. Raw is definitely a better choice for this camera at high ISO.

From a real-world point of view, I went from a Canon 1000D with Tamron 17-50 f/2.8 to a 6D with 24-105 f/4 and I see a nice improvement in low-light performance. If I want great low-light performance, I switch to a fast prime and see a truly dramatic improvement compared to the 1000D. (For reference, the 1000D's DxO score for ISO is 719, compared to the 60D at 813 and 6D at 2340.)

I have just recently acquired a 6D with the 24-105L lens, coming from a 60D and the Sigma 17-50, and the difference is a world apart. The very first night I got the 6D, I shot a pic of a friend in the balcony of a rather dark theater, at 25,600 ISO, and the pic was somewhat grainy, but you can even make out the wrinkles of her face, so much so, that she did not like the pic, but for the sheer ability to obtain an image under those circumstances, it was pretty amazing, at least to me. To be honest, I really never liked shooting anything over 1600 ISO on my 60D, but with the 6D, I have no problems shooting at 10,000 ISO, and pretty much none at 6400, and no problems at all at 3200 ISO.

you will gain a little, but probably not too dramatic (guessing 1 stop or less). You will gain a MUCH larger shooting range along with sharper photos (noticably sharper). Combine the DOF factor, and it's a pretty large change even though 1 aspect isnt that large of a change.

I have a 6d right now, renting it for the weekend, and the ISO performance compared to my 60d is no joke. I don't see the 6d not being able to make up the stop difference easily.

I looked up DxO's T-stop measurements for the two lenses in question. Sigma 17-50 f/2.8 is measured at 3.1 T-stops and the Canon 24-105 f/4 is measured at 5.1 T-stops. (Actually, that's a pretty poor number for the Canon! I'm surprised it's that bad.)

This means that, in terms of light transmission, the Canon lens is actually about 1.4 stops worse than the Sigma, rather than 1 stop as one would assume based on aperture alone. If one believes the DxO ISO ratings for the two cameras, which are 1.5 stops apart, then in terms of low-light capability the two camera+lens combinations in question (60D + Sigma 17-50 vs. 6D + Canon 24-105) ought to perform almost exactly the same.

However, I think that the experience of most people is that the 6D combination actually performs a good deal better in real-life use. What might account for the discrepancy between the DxO measurements and user experience? One possible answer (pure speculation, as I have not used a 60D) is that the noise is more random on the 6D and more patterned on the 60D. On my 1000D the noise at high ISO shows heavy banding, while the 6D noise is pleasantly random-looking. Maybe the same difference exists between the 60D and 6D.

FF is nice but getting better ISO performance camera and then sticking f4 zoom is not my cup of tea. I had 5dmk3 and didn't like at all 24-105mm f4 for indoor shots without flash.

bobbyz wrote in post #15638218
FF is nice but getting better ISO performance camera and then sticking f4 zoom is not my cup of tea. I had 5dmk3 and didn't like at all 24-105mm f4 for indoor shots without flash.

I have a T2i right now and a Sigma 18-50 which stops down to F/4.5 at the telephoto end and I've had zero issues using it indoors in low light and at the highest ISO I'll tolerate with my T2i (1600). Shots turn out fine as long as you have IS. If you're shooting faster moving objects then I guess that'd be another story.

This shot was at 1/10 sec @ F/4.5 and ISO1600

HTTP response: NOT FOUND | MIME changed to 'image/gif' | Redirected to error image by FLICKR

Like Manauta said above, I rarely went over ISO 1600 on my 60D. I frequently resorted to 1/30 and slower exposures under low light just to keep the ISO at 1600.

So I upgraded to a 5D3 specifically for the higher ISO capabilities and it's dead-on focusing. It was like night and day! ISO 6000 is no problem with noise! So far, I haven't ventured higher...no need to.

Since I was using my 24-105 (and other Ls) with the 60D and the 5D3, I can't make any comparisons with the OPs Sigma lens.

EverydayGetaway wrote in post #15638696
I have a T2i right now and a Sigma 18-50 which stops down to F/4.5 at the telephoto end and I've had zero issues using it indoors in low light and at the highest ISO I'll tolerate with my T2i (1600). Shots turn out fine as long as you have IS. If you're shooting faster moving objects then I guess that'd be another story.

This shot was at 1/10 sec @ F/4.5 and ISO1600


Shootout by EverydayGetaway, on Flickr

His shot with a comparable Nikon at the same ISO and at the same FL (50mm) but at F/1.8 was pretty much useless because he didn't have IS on his lens. I would say that you'd definitely get a noticeable improvement in low light ability being able to shoot at much higher ISO's cleanly and using an L lens with IS. I'm upgrading to the 6D kit for the same reason... I almost wasn't going for the 24-105L kit for the same logic of it being a "slow" lens, but the DOF at F/4 on a FF camera is about the same as a crop camera at F/2.8 from what I've seen in the sample galleries here, and the low light ability of the super high ISO's and stabilization of the lens make it a huge upgrade over your current kit (and mine).

Try some kid instead of that guy in same room, same settings.

Typical low light indoor will be f2, ISO6400, and 1/30 or so. That is approx 3 stops lower light than your example. Here is an example, late afternoon time and door is open to her right. 1/90, f2, ISO1600.

bobbyz wrote in post #15638218
FF is nice but getting better ISO performance camera and then sticking f4 zoom is not my cup of tea. I had 5dmk3 and didn't like at all 24-105mm f4 for indoor shots without flash.

For me, 2.8 on the 7D was still not enough for indoor low light movement. The original question is comparing performance between APS-C with f2.8 and FF with f4.

I will still put on my 50 1.4 or 85 1.8 for indoor shots of my 2 year old running around.