70D vs 7D - preliminary review for bird photography

AlanF

Stay at home
CR Pro
Aug 16, 2012
8,890
12,206
My 5DIII is out of action while being repaired and so I have reverted to my old 7D. I am one of those hiker/bird photographers who uses the 300mm f/2.8 II + 2xTC III combination at 600mm for its relative lightness. The 7D autofocusses quite quickly with most lenses in general under all conditions. It is a bit slow with the 300mm f/2.8 II + 2xTC III for well-defined targets and fails for diffuse ones and hunts badly. Out of frustration, I tested a 70D in a shop with the 300mm f/2.8 II + 2xTC III and it was fine. The 70D is reputed to be less noisy than the 7D and slightly sharper (DxO tests give the figures). So I bought one.

Here is a brief review.

1. Most importantly, despite being described as having the same AF system as the 7D, the 70D works very well with the 300mm f/2.8 II + 2xTC III combo and appears to have a more consistent and faster AF. So, the over-riding consideration for me is fulfilled.

2. I prefer the screen and toggle switch of the 7D as I use the camera as a digital telescope in the field for identifying birds. But, the swivel touch screen of the 70D does have advantages.

3. The 70D lacks the spot focus of the 7D - it just has the "open square" in the viewfinder.

4. The weather here has been atrocious so field trials have been difficult. My impressions are that the 70D is indeed sharper than 7D at the extremes of resolution.

As there are so many variables under field conditions, I conducted a test under controlled conditions. I took a very sharp close-up of a kingfisher with my 5DIII in the summer and had a large size print made. So, I used the print as a target. I used the 300mm f/2.8 II without the TC for testing as it is so sharp and will not be the limiting factor in resolution at the distances used indoors. The lens was AFMAd using FoCal on both bodies prior to use.

This target photo had been taken at iso 1250, 1/320s and f/5.6 at 600mm on the 5DIII.
 

Attachments

  • 5DIII_1250iso_320_f5.6_6871.jpg
    5DIII_1250iso_320_f5.6_6871.jpg
    351 KB · Views: 4,599

AlanF

Stay at home
CR Pro
Aug 16, 2012
8,890
12,206
The first set of tests were taken in lowish light and somewhat under exposed. These conditions do tend to emphasize the noise, which is useful for comparison.

The first are 70D above, 7D below, both at f/2.8, iso 320 and 1/160s. The 70D is less noisy. The sharpness is pretty similar, but judge for yourself.
 

Attachments

  • 70D_320iso_160_F2.80514.jpg
    70D_320iso_160_F2.80514.jpg
    507.2 KB · Views: 2,290
  • 7D_320iso_160_f2.8_2300.jpg
    7D_320iso_160_f2.8_2300.jpg
    381.6 KB · Views: 2,342

AlanF

Stay at home
CR Pro
Aug 16, 2012
8,890
12,206
The next are at iso 640 and 1/320 s.
 

Attachments

  • 70D_640iso_320_F2.8_0509.jpg
    70D_640iso_320_F2.8_0509.jpg
    497.4 KB · Views: 2,234
  • 7D_640iso_320_f2.8_2294.jpg
    7D_640iso_320_f2.8_2294.jpg
    482.4 KB · Views: 2,272

AlanF

Stay at home
CR Pro
Aug 16, 2012
8,890
12,206
Finally, at iso 1250 and 1/640 s. The superiority of the 70D noise is more obvious.
 

Attachments

  • 70D_1250_640iso_F2.8_0514.jpg
    70D_1250_640iso_F2.8_0514.jpg
    511.5 KB · Views: 2,228
  • 7D_1250iso_640_F2.8_2292.jpg
    7D_1250iso_640_F2.8_2292.jpg
    515.5 KB · Views: 2,307

AlanF

Stay at home
CR Pro
Aug 16, 2012
8,890
12,206
Later on, the sun came out and brightened the room, with dramatic effects on the noise.

At iso160 and 1/3200 s exposure at f/2.8, there is very little noise and the auto exposure was better. The photos of the print are approaching the quality of the original jpeg from the 5DIII.

I should have mentioned that 7D and 70D images are jpegs from RAW with no noise reduction or sharpening, but the original 5DIII image was cleaned up.
 

Attachments

  • 7D_160iso_3200_f2.8_2305.jpg
    7D_160iso_3200_f2.8_2305.jpg
    496.5 KB · Views: 2,196
  • 70D_160iso_3200_f2.8_0524.jpg
    70D_160iso_3200_f2.8_0524.jpg
    527.2 KB · Views: 2,243

AlanF

Stay at home
CR Pro
Aug 16, 2012
8,890
12,206
Finally, for the 70D at iso 320, 1/4000 s and f/2.8 (top) and iso 1250, 1/4000 s and f/5.6, the noise levels are still excellent. In bright light, the 70D is not noisy. The same number of photons are hitting the sensor to the noise must be sensor noise, not photons. The quality of the images are excellent, compared with the original.
 

Attachments

  • 70D_320iso_4000_f2.8_0517.jpg
    70D_320iso_4000_f2.8_0517.jpg
    504.1 KB · Views: 2,159
  • 70D_1250iso_4000_f5.6_0518.jpg
    70D_1250iso_4000_f5.6_0518.jpg
    504.1 KB · Views: 2,214

AlanF

Stay at home
CR Pro
Aug 16, 2012
8,890
12,206
You need to download the images to analyse them. My 7D does tend to under-expose and you can see it relative to the 70D.

My take on all of this is that the 70D is superior for my purposes to the 7D. It focusses better, faster and more consistently. I think it is sharper. The noise levels are better. It is a great camera by itself and I think that in good light it will have genuinely longer reach than the 5DIII. The live view focussing is simply awesome.
 

Grant Atkinson

EOS 90D
Jan 6, 2014
191
29
Hi there Alan F

I am new to this forum but not new to Canon dslrs and wildlife photography. I have also just completed a comparison of the two cameras, 70D and 7D. My experiences were similar to yours in that I found the 70D to have better image quality all around, and focus at least as well, if not with a little more stability, especially when long lenses are used. I owned a couple of 7D bodies for over a year and have been using the 70D in the field for the last two months. Here is the link to my comparison:http://www.grantatkinson.com/blog/canon-eos-70d-and-canon-eos-7d-compared

Thanks for posting your kingfisher image, it is a good one
Grant
 

MichaelHodges

EOS RP
Nov 7, 2012
423
2
Thanks for the review. I have been in a love/hate relationship with my 7d for several years now. When everything works as it should, it's great. But it's so unforgiving and inconsistent with telephoto lenses. And, IMHO, even the sky and water in ISO 100 shots look noisy.

You need good light to get the best of it, which IMHO defeats the entire purpose of being a good wildlife camera. Most wildlife is out and about when the light is low, or darting into forests with uneven light.

I'd classify the 7D as a zoo camera, portrait and sports camera, and would not recommend it for wildlife shooting except for afternoons. The images are just too sludgy when not in perfect light.
 

AlanF

Stay at home
CR Pro
Aug 16, 2012
8,890
12,206
Grant Atkinson said:
Hi there Alan F

I am new to this forum but not new to Canon dslrs and wildlife photography. I have also just completed a comparison of the two cameras, 70D and 7D. My experiences were similar to yours in that I found the 70D to have better image quality all around, and focus at least as well, if not with a little more stability, especially when long lenses are used. I owned a couple of 7D bodies for over a year and have been using the 70D in the field for the last two months. Here is the link to my comparison:http://www.grantatkinson.com/blog/canon-eos-70d-and-canon-eos-7d-compared

Thanks for posting your kingfisher image, it is a good one
Grant

Grant
Yours is one of the best reviews I have ever seen. Thanks so much for the link. Where my modest contribution overlaps with yours, there is complete agreement, even to the 7D's relative degree of under-exposure. I also learned a few tricks from you.

Welcome to the forum!
Alan
 
Grant, welcome to CR. I found your site some time ago when I was researching the 300 2.8 IS II and found your insights on it and all of the other gear you've reviewed really helpful. I really like your work, particularly the lion shots, and I'm glad to see you posting here.
 

AlanF

Stay at home
CR Pro
Aug 16, 2012
8,890
12,206
Marsu42 said:
Btw (hint, hint :) ... unless shooting jpeg only use full iso stops, iso 320 is really iso 400 underexposed, iso 640 is iso 800 underexposed, you gain nothing by this but only lose dynamic range.

This often crops up. iso 160, 320, 640 and 1250 are local minima in the jagged profile of noise vs iso for the 5DIII and 7D. Other isos might give you better DR. For what I am doing, I prefer lower noise.
 
AlanF said:
Marsu42 said:
Btw (hint, hint :) ... unless shooting jpeg only use full iso stops, iso 320 is really iso 400 underexposed, iso 640 is iso 800 underexposed, you gain nothing by this but only lose dynamic range.

This often crops up. iso 160, 320, 640 and 1250 are local minima in the jagged profile of noise vs iso for the 5DIII and 7D. Other isos might give you better DR. For what I am doing, I prefer lower noise.
Alan, not to wade into this, but I don't think those ISOs make any difference in noise when shooting RAW, but do affect DR. I haven't tested this, but I think Marsu is correct in saying that they only reduce noise for JPEG shooting. If I have time, I'll try to do some tests with DxO (PRIME) and Adobe ACR at some point to test this theory. I don't have any Victorian bricks, but I'll see what I can do :).

Those are interesting insights into the 70D vs. 7D. It seems like the build quality, viewfinder, and burst rate are the only reasons to choose the 7D and probably aren't significant enough to make a big difference.
 

Grant Atkinson

EOS 90D
Jan 6, 2014
191
29
Thanks for the welcome here Alan F, and Mackguyver,
I have followed along here and there the relevant posts on CR forum for a while now just viewing, but thought I would join in. Forums like this are good places to learn, confirm and share especially :)
Alan, I was a little surprized to find the exposure difference between the 7D and 70D, so much so that I actually went back and re-took all my indoor comparison shots.
Thanks again for the welcome
cheers
Grant
 

neuroanatomist

I post too Much on Here!!
CR Pro
Jul 21, 2010
26,123
4,720
Alan, thanks for posting these results!

Marsu42 said:
Btw (hint, hint :) ... unless shooting jpeg only use full iso stops, iso 320 is really iso 400 underexposed, iso 640 is iso 800 underexposed, you gain nothing by this but only lose dynamic range.

Do you have data to support that assertion? You didn't respond previously

Btw (hint, hint), Bill Claff's data suggest that you're wrong, and that using ISO 160-320-640 gives less noise (as AlanF indicates) and also more DR than the 1/3-stop higher full stop increments of ISO 200-400-800.

As I stated in the other thread, it seems that from a noise as well as a DR standpoint, you're better off using multiples of ISO 160 than ISO 100 (unless you're using a 1D X, which has smooth noise and DR curves, unlike the jagged ones of other recent Canon sensors).
 

Attachments

  • Claff ISO Noise.png
    Claff ISO Noise.png
    76.7 KB · Views: 10,421
  • Claff ISO DR.png
    Claff ISO DR.png
    76.6 KB · Views: 10,229

DanielW

EOS RP
Mar 13, 2012
255
0
43
neuroanatomist said:
Alan, thanks for posting these results!

Marsu42 said:
Btw (hint, hint :) ... unless shooting jpeg only use full iso stops, iso 320 is really iso 400 underexposed, iso 640 is iso 800 underexposed, you gain nothing by this but only lose dynamic range.

Do you have data to support that assertion? You didn't respond previously

Btw (hint, hint), Bill Claff's data suggest that you're wrong, and that using ISO 160-320-640 gives less noise (as AlanF indicates) and also more DR than the 1/3-stop higher full stop increments of ISO 200-400-800.

As I stated in the other thread, it seems that from a noise as well as a DR standpoint, you're better off using multiples of ISO 160 than ISO 100 (unless you're using a 1D X, which has smooth noise and DR curves, unlike the jagged ones of other recent Canon sensors).

Very helpful and supported!

mackguyver said:
Neuro, this item definitely deserves some research - I'll try to shoot some high DR test shots this weekend at the various ISOs (on my 5DIII) and then run them through a battery of tests with the tools I have.

Waiting anxiously on the results!
 

Marsu42

Canon Pride.
Feb 7, 2012
6,314
0
Berlin
der-tierfotograf.de
neuroanatomist said:
Alan, thanks for posting these results!
Marsu42 said:
Btw (hint, hint :) ... unless shooting jpeg only use full iso stops, iso 320 is really iso 400 underexposed, iso 640 is iso 800 underexposed, you gain nothing by this but only lose dynamic range.
Do you have data to support that assertion? You didn't respond previously …

Sorry, I am a CR regular, but I don't spot every post, so I missed that. My source are tests are the Magic Lantern devs: http://magiclantern.wikia.com/wiki/ISO

SO 160, ISO 200 and ISO 250 are identical in RAW. Proof: try it yourself with dcraw. This means that all of them are obtained by different digital processing of the same RAW data. In all 3 cases, the analog circuitry is configured at the same parameters. Some (or all) of these ISOs are either pushed or pulled digitally.

However they also say that the *ML* version if iso160 multiples has better dr, but not the Canon one - however their wiki which afaik is based on the 5d2 might be outdated and Canon has improved their digital pull/push on newer cameras, if so that would be very valuable information and I'll have a look into the information you provided.

Are ISO 160 multiples the best to use?
NO. They have harsh highlight rolloff and intentionally clipped details in highlights. I have no idea why.

Are ISO 100 multiples the best to use?
NO. While they do have smooth highlight rolloff, they are digitally pushed by a small amount (the exact value depends on picture style and other settings). What does this mean: a small amount of raw data, which actually has the best SNR possible, is simply thrown away.

Then, what is the best ISO?
To the best of my knowledge, the best ISOs are the ones available in recent Magic Lantern versions (April 2012 or later), obtained from ISO 100 multiples adjusted with a small amount of negative digital gain:

ISO 85, 175, 350, 700, 1400, 2800 - best for Neutral -4 and other low-contrast styles.
ISO 80, 160, 320, 640, 1250, 2500 - good for a wide range of situations.
ISO 70, 140, 280, 560, 1100, 2200 - best for high-contrast styles.
 

neuroanatomist

I post too Much on Here!!
CR Pro
Jul 21, 2010
26,123
4,720
Marsu42 said:
My source are tests are the Magic Lantern devs: http://magiclantern.wikia.com/wiki/ISO

SO 160, ISO 200 and ISO 250 are identical in RAW. Proof: try it yourself with dcraw. This means that all of them are obtained by different digital processing of the same RAW data. In all 3 cases, the analog circuitry is configured at the same parameters. Some (or all) of these ISOs are either pushed or pulled digitally.

However they also say that the *ML* version if iso160 multiples has better dr, but not the Canon one - however their wiki which afaik is based on the 5d2 might be outdated and Canon has improved their digital pull/push on newer cameras, if so that would be very valuable information and I'll have a look into the information you provided.

According the Bill Claff's data, the 5DII is pretty close to the 5DIII in terms of DR at various ISO values. The only blip seems to be specifically at ISO 160, where the 5DIII has significantly better DR (better than ISO 200, whereas with the 5DII, ISO 160 is slightly lower, but ISO 320 has more DR than ISO 400, and ISO 640 is better than ISO 800.

I'm not sure if this represents a difference specific to ML, a difference in testing methodologies, or a difference in video vs. still processing (if the ML information is based on video, but maybe it's not).

I agree that it seems sets of three ISOs are digital push and pull from a single analog exposure (for cameras except the 1D X, which doesn't have that jagged response). But…is the triplet 160/pull–200/none–250/push, or is it 100-pull–160/none–200-push, etc.? Or none of them? DxO reports 'base ISO' values of 80 for some cameras, 95 for others, etc., meaning that the ML statement, "ISO 100 multiples adjusted with a small amount of negative digital gain," may be closest to the truth, with the amount of negative digital gain simply offsetting the positive digital gain applied by the camera (and that 'small amount' will be different for different cameras).
 

Attachments

  • 5DII DR.png
    5DII DR.png
    64.4 KB · Views: 529

Marsu42

Canon Pride.
Feb 7, 2012
6,314
0
Berlin
der-tierfotograf.de
neuroanatomist said:
I'm not sure if this represents a difference specific to ML, a difference in testing methodologies, or a difference in video vs. still processing (if the ML information is based on video, but maybe it's not).

It's certainly based on stills raw shooting, ML doesn't only do video :) ... to be sure what's it all about I crossposted the issue on their forum, I hope their devs will have a look at your DxO data: http://www.magiclantern.fm/forum/index.php?topic=9867.msg94974#msg94974