70D Image Quality Worse than T3i?

[cpderic]

Hey there,

I recently purchased a Canon 70D, an upgrade from my previous camera, the Rebel T3i. But after doing some shooting, I've found the image quality on the 70D to actually be a downgrade from the T3i. I'm not sure if anyone else has noticed this, but I wanted to get a second opinion.

As a test, I took a few shot out front of my house. Both cameras were set in manual mode with the following settings:

ISO: 400

Shutter Speed: 1/250

Apature: f/9

These shots were both taken on the Canon 18-55 EF-S kit lens.

I opened the images up in Photoshop at 100% resolution, and compared the leaves and branches on a tree.

These were my results.

It seems clear to me that the T3i produces a much sharper and clearer image, which is a dissapointment. With the 70D being a higher price class, I was expecting the image quality to be at least as good as the Rebel series, if not better.

I'd love to hear a second opinion on this issue, or hear from anyone who might have had a similar issue.

Thanks,

Eric

[Skirball]

---

@cpderic wrote:

Part of the issue is, even though the 70D may have focused on the pine tree to the left, there is not enough depth of field that would cause everything else in the shot to become blurred. So even if my two shots aren't the best example, I still think there's something wrong with the camera. If you look at the full images and compare any area of the image, the T3i looks better no matter what.

---

The 70D photo isn't in focus.  Plain and simple.  I don't know what you focused on, but it wasn't the tree.  If it is, then your camera has serious focus issues.  Test charts are of limited use, but this is why we have them.  Indoors, controlled light, flat static subject.  Manual camera settings, manual focus (unless you're testing the AF).  You need to reduce as many variables as possible to really compare two cameras. 

I don't know what the issue is, and it's possible that you have a lemon.  But there's absolutely no way the 70D is that soft compared to any other of its crop sensors.

[cpderic]

---

@Skirball wrote:

The 70D photo isn't in focus.  Plain and simple.  I don't know what you focused on, but it wasn't the tree.  If it is, then your camera has serious focus issues.  Test charts are of limited use, but this is why we have them.  Indoors, controlled light, flat static subject.  Manual camera settings, manual focus (unless you're testing the AF).  You need to reduce as many variables as possible to really compare two cameras. 

I don't know what the issue is, and it's possible that you have a lemon.  But there's absolutely no way the 70D is that soft compared to any other of its crop sensors.

---

In a way, I am testing the AF. While I respect the fact that I could do much more controlled testing, I also would like a camera that is capable of preforming consistant autofocus while run-and-gunning. And if I've never had any autofocus issues with the T3i, I simply don't understand why I'm having them with the 70D.

[cuda719]

In a way, I am testing the AF. While I respect the fact that I could do much more controlled testing, I also would like a camera that is capable of preforming consistant autofocus while run-and-gunning. And if I've never had any autofocus issues with the T3i, I simply don't understand why I'm having them with the 70D. Because your letting the camera select the the focus point not you. You need to learn how the camera operates, has something all salesman leave out when purchasing a new upgraded camera. You need to but the cameras on a tripod use a image that has a lot of contrast. Have both cameras in manual select the focus point. Compose your shot then take two picures one with each camera. Then compare. As far as the run and gun statement. We first learn how to crawl before we learn how to run.

[TCampbell]

The 70D has 19 AF points ... all cross-type.  The T3i has 9 AF points... only one cross-type.

As with all Canon DSLR cameras... you can let the camera use any AF point it wants to lock focus.... OR you can pick the focus point you want the camera to use.    If you let the camera pick the AF point, then you need to be aware that the camera will ALWAYS pick the AF point which is able to lock focus at the CLOSEST focusing distance to the camera.

If a camera is in "One Shot" mode it will insist on locking focus before it shoots.  If it is in "AI Servo" mode then it WILL take the shot when you fully press the shutter button whether or not it had time to lock focus.

We know that the 70D image is soft... but we don't know why.  

Unfortunately the EXIF data is incomplete in these images... I cannot check the AF point selection or the focus mode, etc. 

It's entirely possible this is the camera's fault.  But it's also entirely possible this is the photographer's fault.  There's not enough info here.

In any case, whenever you want to test camera focus performance, you REALLY need to use controlled test conditions.  The camera is just a tool.  I can buy a top-end violin... but as I don't know how to play a violin, it will still just sound like I'm torturing cats.  

If you don't own a commercial test chart, download a free chart... like this one:  http://regex.info/blog/photo-tech/focus-chart

You'll print that, follow instructions, and force the camera to use just one selected AF point (your choice), but you'll lock that point on the center of the test chart with the chart on an angle.  You'll then inspect to determine if the optimum focus is spot on where you want it... or if it's closer... or if it's farther (that's why the chart is leaning back on an angle.)

You need to do many tests (not just one... one is not a representative sample size.)  I deliberately de-focus the lens before each shot... doign many shots where I deliberately run focus all the way, as well as another series where I run focus all the way out.  This is becuase it's possible for a lens to nail focus in one direction... and yet miss focus in the other direction.

If the camera is missing focus consistently and always in the same direction... you're in luck.  The 70D has AF Microadjustment.  Meaning you can tell the camera that it consistently misses focus in some direction ... for example suppose it always focuses a bit too far away... you can tell it to always focus a bit closer then it's calculated focus position to compensate.    See:  http://www.learn.usa.canon.com/resources/articles/2011/af_microadjustment_article.shtml

[cpderic]

If the camera is missing focus consistently and always in the same direction... you're in luck.  The 70D has AF Microadjustment.  Meaning you can tell the camera that it consistently misses focus in some direction ... for example suppose it always focuses a bit too far away... you can tell it to always focus a bit closer then it's calculated focus position to compensate.    See:  http://www.learn.usa.canon.com/resources/articles/2011/af_microadjustment_article.shtml

---

I have done some other tests, and found that my images seem to improve if I set my af microadjustment to +10. However, I still don't feel comfortable knowing that it is a brand new camera and needs this much of an adjustment. Is it normal for a camera to need such large AF microadjustments right out of the box?

[TCampbell]

Every lens/body combination will be different.

See:  http://www.lensrentals.com/blog/2008/12/this-lens-is-soft-and-other-myths

[70Downer]

@TCAMPBELL

I downloaded his pictures then brought them into DPP. Both are AF one shot, auto  (all points) selection . But if you look at his full size pictures for the T3i he locked the center point half way up on to the skinny tree. For the 70D he pointed it at the green tree 20+ feet to the left. Of which the camera locked on to the small clump of weeds  on the ground, the closest high contrast target.

Then he took 100% crops of the skinny tree (the one he locked the t3i on) to prove his 70D has an AF problem.

I am not saying he does or doesnt have a AF problem. But this test will not help one way or another 🙂

[hsbn]

Here Campbell

Right Click to view larger size. Camera not only focus on a different object, but each use different focus mode also.

[TCampbell]

Ahh, yes I missed the post where the full size images were linked.  I downloaded those and opened them in Canon Digital Photo Professional.

I see the T3i was in AI Servo mode and the 70D was in One Shot mode.  

As yoiu've both pointed out, the cameras were locking on different subjects.

The 70D still looks like it front focused a bit but if there was a focus-and-recompose then that could explain that.

We can't really draw any conclusions from these shots because the focus targets were not on the same subject.