Which extension 1.4x or 2x or none?

Charlie wrote in post #18289937
a more magnified view will show more flaws.

A valid point about the so-called "magnification" (a misnomer IMO) of crop sensors. Actually, there's another factor at play: Unless you're using one of the 5DS bodies, the typical 18-24 MP crop sensor has greater resolving power (finer pixel pitch) than the current Canon FF sensors, so even at the same image scale, you might see it more (if you're really looking for it).

Charlie wrote in post #18289937
typically, telephotos are good throughout the frame anyhow, the center IQ argument may work with a 50 f1.8, doesnt work well when you've got a 400+ that's sharp all over.

This isn't about a $100 plastic lens. There are $2000 L zooms that are soft at the edges, and get worse with the TC's even though they're compatible with them. Even some of Canon's shorter L primes soften at the edges. I'm not trying to start a crop-versus-FF war, but in general the crop sensor simply cuts off the worst part of the image circle. Most of Canon's prime teles and superteles are incredibly sharp across the frame, and those are the ones people are most likely to use the TC's with. But there's still a degradation.

And I stand by my original statement above, that everybody has his or her own idea of what "acceptable" IQ is. Some people want to look at each pixel at 100% and stress over every little bit of softness and CA. Other people just want to see a satisfying image at a reasonable viewing size. I'm not making judgements either way.

Mike

Elton Balch wrote in post #18290181
Those are obviously great photos but I'm guessing the OP isn't going to rush out and buy a 5d iv to replace the 60d and a $7,000 lens in addition to the 2x converter although for photos like that it might be worth it!

I didn't suggest that he should, I was simply refuting the often repeated line that the 2x significantly degrades IQ and should only be used as a last resort. In my experience when people complain about the 2x it's because they either put it on lenses that aren't really suited to an extender, or they need to work on their technique. Shooting at 800mm and beyond isn't the easiest thing in the world. The truth is that there are people out there who are capable of getting better shots while using the 2x than what most people can get with a bare lens.

The 2x works great on the 400... f/2.8

On the 400/5.6 you definitely want the 1.4x, and I'd stop down 1 or 2 thirds of a stop.

mcoren wrote in post #18290204
A valid point about the so-called "magnification" (a misnomer IMO) of crop sensors. Actually, there's another factor at play: Unless you're using one of the 5DS bodies, the typical 18-24 MP crop sensor has greater resolving power (finer pixel pitch) than the current Canon FF sensors, so even at the same image scale, you might see it more (if you're really looking for it).

We often hear this perspective of 'more resolving power' due to finer pixel pitch of APS-C sensors vs. FF sensors. Rightly so, considering the most recent ASP-C as 269 pixels/mm, while the 5DIV has only 187 pxiels/mm.

If we start with the understanding that the same lens will deliver an IDENTICAL subject size at the focal plane, when we compare pixels-on-subject, yes there are 44% more pixels per axis on the subject when imaged with the newest APS-C cameras. Effectively, there is 'more reach' and finer detail with the APS-C camera.

If we start with 'same AOV lens for both formats', e.g. 625mm on APS-C vs. 1000mm on FF, we have a different animal to consider. The FF sensor captures an area of 2.3' x 3.5' at a subject distance of 100'.
The APS-C sensor also captures an area of 2.3' x 3.5'. Now when we make a 16x24" print from both,
we enlarge the FF image by 16.9X enlargement, and we enlarge the APS-C image by 27.3X.

• So we end up with 9.85 pixels/mm on print from APS-C image, and 11.0 pixels/mm from FF image...MORE PIXELS per millimeter from the FF image!
  
• And if we start with a superb lens that might provide 150 line-pairs/millimeter delivered to the focal plane, the APS-C print has 5.5 line-pairs/mm while the FF print has 8.9 line-pairs/mm...more lens resolution delivered to print from FF image, as well!
  

However it seems that people don't always try to equalize the AOF between formats, they go with a FF to get a wider shot with a lens (ie. "to use the lens as it was meant to be used"), or they get a crop body to increase that AOV (ie. "because they are reach limited").

Wilt wrote in post #18290406
So we end up with 9.85 pixels/mm on print from APS-C image, and 11.0 pixels/mm from FF image...MORE PIXELS per millimeter from the FF image!

Way to pick and choose convenient numbers to illustrate your bias!

Why stop there? Why not use the 50 MP FF 5DS versus the original 6 MP Digital Rebel? Then the results will look even better for you.

For a meaningful comparison, assume 2 sensors with the same number of pixels.

Wilt wrote in post #18290406
We often hear this perspective of 'more resolving power' due to finer pixel pitch of APS-C sensors vs. FF sensors. Rightly so, considering the most recent ASP-C as 269 pixels/mm, while the 5DIV has only 187 pxiels/mm.

If we start with the understanding that the same lens will deliver an IDENTICAL subject size at the focal plane, when we compare pixels-on-subject, yes there are 44% more pixels per axis on the subject when imaged with the newest APS-C cameras. Effectively, there is 'more reach' and finer detail with the APS-C camera.

If we start with 'same AOV lens for both formats', e.g. 625mm on APS-C vs. 1000mm on FF, we have a different animal to consider. The FF sensor captures an area of 2.3' x 3.5' at a subject distance of 100'.
The APS-C sensor also captures an area of 2.3' x 3.5'. Now when we make a 16x24" print from both,
we enlarge the FF image by 16.9X enlargement, and we enlarge the APS-C image by 27.3X.

• So we end up with 9.85 pixels/mm on print from APS-C image, and 11.0 pixels/mm from FF image...MORE PIXELS per millimeter from the FF image!
  
• And if we start with a superb lens that might provide 150 line-pairs/millimeter delivered to the focal plane, the APS-C print has 5.5 line-pairs/mm while the FF print has 8.9 line-pairs/mm...more lens resolution delivered to print from FF image, as well!
  

So the 'APS-C advantage' materializes when we use SAME FL on both bodies, but disappears when we use a 'format appropriate' FL on each body.

Hey Wilt, you are really overthinking this! I'm my case, I'm buying a 7d ii (arriving today!) as a substitute for using a 1.4x or 2x extender with my 100-400 form extra reach on my 5d iii. Most who post here understand the differences between full frame and APS-C and the compromises that occur in various shooting situations. In my case I'm not going to rush out and buy an uber expensive 1000 mm lens for my full frame to get better prints! The 7d ii with my 100-400 and my 1.4x extender will be just fine.

Ok, two things:

1) to answer OP's question: for the 100-400, I'd stick with the 1.4x so that you can keep AF. And, the IQ degradation is less than the 2x. Once I got dialed in with the 100-400 + 1.4xIII combo, I was getting some excellent quality photos at f/8 @ 560mm.

2) the crop vs. FF pixels on target stuff is confusing! I have an 80D and a 5D4, and I basically now just reach for the body that best suits the light and distance situation, regardless of pixels on target. Gets even more confusing if we start talking about cropping in post, which file has better IQ and malleability

So, my operating procedure when shooting birds with the 400mm DO II and have both 1.4x/2x extenders to use: 80D in good light, 5D4 when lower light, or if I don't need as much reach. Try to shoot natively as much as possible for best IQ and AF performance. 1.4x is great, no complaints aside from stop of light loss. 2x extender is borderline for me, I try not to use it. Would rather crop in post, but evidence suggests using extender is better. Oh well..

Seems to be working so far. When I'm out shooting, I don't want to short circuit my brain by puzzling over this stuff!

.

MatthewK wrote in post #18290728
.........to answer OP's question: for the 100-400, I'd stick with the 1.4x so that you can keep AF. And, the IQ degradation is less than the 2x. Once I got dialed in with the 100-400 + 1.4xIII combo........

But I don't see how that answers the OP, as the OP never mentioned anything about a 100-400mm zoom lens.
.

MatthewK wrote in post #18290728
.........the crop vs. FF pixels on target stuff is confusing! I have an 80D and a 5D4, and I basically now just reach for the body that best suits the light and distance situation, regardless of pixels on target. Gets even more confusing if we start talking about cropping in post, which file has better IQ and malleability

When I'm out shooting, I don't want to short circuit my brain by puzzling over this stuff!

.
I completely agree. Photography is about how things look. It is about aesthetic qualities and how they make you feel when you see them and how much you like the look of the images you create.

The very essence of photography and other fine arts does not lie in anything that can be quantified. Math really isn't a good tool to be using to figure out what kinds of photos will make you happy when you look at them.

Besides, mathematical discussions bore me to tears and I hate having to pick thru paragraphs of math-related stuff, hoping to see some non-math sentences hidden somewhere in the midst of all the numbers.

.

Tom Reichner wrote in post #18290733
.

But I don't see how that answers the OP, as the OP never mentioned anything about a 100-400mm zoom lens.
.

.
I completely agree. Photography is about how things look. It is about aesthetic qualities and how they make you feel when you see them and how much you like the look of the images you create.

The very essence of photography and other fine arts does not lie in anything that can be quantified. Math really isn't a good tool to be using to figure out what kinds of photos will make you happy when you look at them.

Besides, mathematical discussions bore me to tears and I hate having to pick thru paragraphs of math-related stuff, hoping to see some non-math sentences hidden somewhere in the midst of all the numbers.

.

Crap, you are right. I saw someone else mention the 100-400, and just went with it. But luckily, the same applies since 400mm f/5.6 is the same for the zoom and the prime

And I whole heartedly agree with your statement. These days I'm learning to better put more emphasis on the process of making a photo, and it's made me much more happy than I was when I was stressing about sharpness. I'm ok if my photos aren't the sharpest in the universe. I don't care anymore. I care more about whether the sun is shining, if it's windy or calm, if wildlife is active, can I incorporate a new technique, can I use a lens I haven't used in while,

No bias, simply presenting hard FACTS!
I own both APS-C and FF, and usually find APS-C to be fully sufficient! But just this morning, I did need FF.

Folks are so prone to accuse of Bias...Why?

I suggest 1.4x rather than 2x. I feel that 2x pushes the tele-extender technology a bit too far.

To test autofocus capability I mounted my Kenko 1.4X PRO 300 Teleconverter DGX with my 70-300L on my 60D. Although the aperture reported on the camera's LCD was f/8, the camera at least tried to autofocus. I found it effective to focus manually before pressing the shutter button to autofocus. The camera does better when the subject is almost in focus. (This is true even without the teleconverter with an f/5.6 lens.) I think that the Kenko teleconverter "lies" to the camera to enable autofocus at f/8.

The Kenko DGX teleconverter has been superseded by a newer model, but Amazon.com still has a few.

My 70D does a little better with this lens+teleconverter combination, but still struggles a bit. The 70D is supposed to autofocus at f/11 in live view mode, but I have not tried this. And, of course, the 80D is supposed to autofocus at f/8, but I don't happen to own one (yet.)

As for image quality, I found that using the lens without the teleconverter and cropping gave me similar results, but reduced the effective pixel count from 20 megapixels to about 10 megapixels. This probably matters only if I plan to do further editing or cropping.

Wilt wrote in post #18290831
No bias, simply presenting hard FACTS!
I own both APS-C and FF, and usually find APS-C to be fully sufficient! But just this morning, I did need FF.

Folks are so prone to accuse of Bias...Why?

Wilt, you are biased toward printing
In any case, as mcoren has noticed, pixel density (and pixel quality - I say) needs to be factored in.

As far as I can work out, the 60d will only autofocus with f/5.6 or wider lenses (I might be wrong). Therefore, autofocus would not work with either TC.

I often use a 2x III on a Sigma 120-300 f/2.8 on a 7d2, but that results in only f5.6. I am generally happy with the shots I get with this combination. Using a 2x TC on a f/5.6 lens means you are starting at f/11 and more likely shooting at f/16. On top of any IQ reduction you might see due to the TC, you may well also see degradation due to diffraction. There will likely also be increased noise, due to the need for higher ISO settings. Depending where you live, it might often be too dark to shoot at those settings. It would often be the case in northern Scotland, where I live.

I am not sure that you will be happy with the results with a 2x. A 1.4x would have less problems, losing only 1 stop, but you probably still need good light to get decent results.

Honey Monster wrote in post #18291374
As far as I can work out, the 60d will only autofocus with f/5.6 or wider lenses (I might be wrong). Therefore, autofocus would not work with either TC.

I often use a 2x III on a Sigma 120-300 f/2.8 on a 7d2, but that results in only f5.6. I am generally happy with the shots I get with this combination. Using a 2x TC on a f/5.6 lens means you are starting at f/11 and more likely shooting at f/16. On top of any IQ reduction you might see due to the TC, you may well also see degradation due to diffraction. There will likely also be increased noise, due to the need for higher ISO settings. Depending where you live, it might often be too dark to shoot at those settings. It would often be the case in northern Scotland, where I live.

I am not sure that you will be happy with the results with a 2x. A 1.4x would have less problems, losing only 1 stop, but you probably still need good light to get decent results.

I'm often surprised how quick ISO can jump when shooting at f/8, even when shooting on a bright sunny day.