Do I have a 70D AF problem or don't I?

tony22

Forum Enthusiast
Messages
480
Reaction score
21
Location
US
I used my 50mm f/1.4 Canon lens to do the AF test as documented here


I placed my new refurb 70D on my Manfrotto tripod, used my electronic beam level to make sure the front plane of the lens (against the front edge of the filter ring) and the rotation of the camera were both square to the target. Focus target for both viewfinder and Live View were dead on the center widget of the target.

These pics are JPG SOOC so no cropping, Levels, or anything were applied. I wasn't sure if I could apply Levels and not skew the result, so these are a little dim. Raining today and I could only get so much light in the room where I took the shots. I was 2.5 meters away from the target. I know this makes it tough, but I'd like to see if people who know about this problem can provide any insight.




Viewfinder focus




Live View




MAF +6
 

Attachments

  • 3086934.jpg
    3086934.jpg
    5 MB · Views: 0
  • 3086935.jpg
    3086935.jpg
    5.1 MB · Views: 0
  • 3086936.jpg
    3086936.jpg
    5.3 MB · Views: 0
That was the big flaw of that video you linked to, the author didn't seem to understand the real problem that some 70Ds had.

The real AF problem that was experienced by some (primarily early production models sold in Germany) with the 70D was a difference in focus between the center AF point when using the center f/2.8 AF point, and the focus when using the surrounding f/5.6 AF points.
I thought the 70D problem was that (on the cams that have this problem and lenses with max aperture of f/2.8 and lower) with single point center focus the viewfinder focus was "off" when compared with the Live View AF.
NO!!! That is perfectly NORMAL. (see "This lens is soft" and other myths)

It is if the f/2.8 AF point is giving you different focus than the other PDAF (viewfinder) AF points. That is not normal and is an issue.
Yes, I am familiar with the German video that identified the issue.
Rest assured I do plan to use the DotTune method for MFA.
--
All statements in my posts represent my interpretation of data, research opinion or viewpoints.
The opinions expressed are not representations of fact, and are subject to change without notice.
All images are used for educational purposes.
I am very confused reading your responses. I understand the initial issue you mentioned. The one I have heard about and that many people are taking about is.. With fast lenses and single point auto focus (center), the images taken using the viewfinder are out of focus compared to if the same shot is taken with the LCD. If someone does a controlled test (tripod and shutter release/10 second timer) and they focus on the same spot very carefully..images take with viewfinder should be just as sharp as they are when taken using the LCD. Unless I missed something you posted I don't get how someone could disagree that that's an issue.
 
That was the big flaw of that video you linked to, the author didn't seem to understand the real problem that some 70Ds had.

The real AF problem that was experienced by some (primarily early production models sold in Germany) with the 70D was a difference in focus between the center AF point when using the center f/2.8 AF point, and the focus when using the surrounding f/5.6 AF points.
I thought the 70D problem was that (on the cams that have this problem and lenses with max aperture of f/2.8 and lower) with single point center focus the viewfinder focus was "off" when compared with the Live View AF.
NO!!! That is perfectly NORMAL. (see "This lens is soft" and other myths)

It is if the f/2.8 AF point is giving you different focus than the other PDAF (viewfinder) AF points. That is not normal and is an issue.
Yes, I am familiar with the German video that identified the issue.
Rest assured I do plan to use the DotTune method for MFA.
 
That was the big flaw of that video you linked to, the author didn't seem to understand the real problem that some 70Ds had.

The real AF problem that was experienced by some (primarily early production models sold in Germany) with the 70D was a difference in focus between the center AF point when using the center f/2.8 AF point, and the focus when using the surrounding f/5.6 AF points.
I thought the 70D problem was that (on the cams that have this problem and lenses with max aperture of f/2.8 and lower) with single point center focus the viewfinder focus was "off" when compared with the Live View AF.
NO!!! That is perfectly NORMAL. (see "This lens is soft" and other myths)

It is if the f/2.8 AF point is giving you different focus than the other PDAF (viewfinder) AF points. That is not normal and is an issue.
Yes, I am familiar with the German video that identified the issue.
Rest assured I do plan to use the DotTune method for MFA.
--
All statements in my posts represent my interpretation of data, research opinion or viewpoints.
The opinions expressed are not representations of fact, and are subject to change without notice.
All images are used for educational purposes.
I am very confused reading your responses. I understand the initial issue you mentioned. The one I have heard about and that many people are taking about is.. With fast lenses and single point auto focus (center), the images taken using the viewfinder are out of focus compared to if the same shot is taken with the LCD. If someone does a controlled test (tripod and shutter release/10 second timer) and they focus on the same spot very carefully..images take with viewfinder should be just as sharp as they are when taken using the LCD. Unless I missed something you posted I don't get how someone could disagree that that's an issue.
Because it is NORMAL as long as it is correctable using micro focus adjustment.

Please read this article to understand why: "This lens is soft" and other myths

--
All statements in my posts represent my interpretation of data, research opinion or viewpoints.
The opinions expressed are not representations of fact, and are subject to change without notice.
All images are used for educational purposes.
I understand and I just read that article..the 2 concerns I have are.

I don't know about this so correct me if I am wrong. If you take 2 pics. The focus from the live view one is spot on and then the one from the view finder is soft or off...how is that normal? If you MFA is that not affecting the live view focus..which is already spot on

The second thing. If there is something from above that I'm wrong about..and you do MFA...and the focus is still different between the view finder and live view then that's an issue.

That is what most people are referring to. People are doing controlled tests after MFA and getting the same results. Good focus with love view and bad with view finder.
 
Last edited:
That was the big flaw of that video you linked to, the author didn't seem to understand the real problem that some 70Ds had.

The real AF problem that was experienced by some (primarily early production models sold in Germany) with the 70D was a difference in focus between the center AF point when using the center f/2.8 AF point, and the focus when using the surrounding f/5.6 AF points.
I thought the 70D problem was that (on the cams that have this problem and lenses with max aperture of f/2.8 and lower) with single point center focus the viewfinder focus was "off" when compared with the Live View AF.
NO!!! That is perfectly NORMAL. (see "This lens is soft" and other myths)

It is if the f/2.8 AF point is giving you different focus than the other PDAF (viewfinder) AF points. That is not normal and is an issue.
Yes, I am familiar with the German video that identified the issue.
Rest assured I do plan to use the DotTune method for MFA.
 
Again, I highly recommend the DotTune method. And keep in mind that MFA should be done in the light you normally use the lens in, and at the distances you normally shoot at.
Just finished it for my 50mm f/1.8. It's a great method! Thanks for the link TTM. I ran it a number of times and got +2 / +2.5 every time.

I feel like I'm good to go. My single point AF yields the same results as Zone, and my MFA on the first lens was repeatable and not far off nominal at all. Good old Canon refurb (this is my second one over many years of using different Canon cams).
 
That was the big flaw of that video you linked to, the author didn't seem to understand the real problem that some 70Ds had.

The real AF problem that was experienced by some (primarily early production models sold in Germany) with the 70D was a difference in focus between the center AF point when using the center f/2.8 AF point, and the focus when using the surrounding f/5.6 AF points.
I thought the 70D problem was that (on the cams that have this problem and lenses with max aperture of f/2.8 and lower) with single point center focus the viewfinder focus was "off" when compared with the Live View AF.
NO!!! That is perfectly NORMAL. (see "This lens is soft" and other myths)

It is if the f/2.8 AF point is giving you different focus than the other PDAF (viewfinder) AF points. That is not normal and is an issue.
Yes, I am familiar with the German video that identified the issue.
Rest assured I do plan to use the DotTune method for MFA.
 
That was the big flaw of that video you linked to, the author didn't seem to understand the real problem that some 70Ds had.

The real AF problem that was experienced by some (primarily early production models sold in Germany) with the 70D was a difference in focus between the center AF point when using the center f/2.8 AF point, and the focus when using the surrounding f/5.6 AF points.
I thought the 70D problem was that (on the cams that have this problem and lenses with max aperture of f/2.8 and lower) with single point center focus the viewfinder focus was "off" when compared with the Live View AF.
NO!!! That is perfectly NORMAL. (see "This lens is soft" and other myths)

It is if the f/2.8 AF point is giving you different focus than the other PDAF (viewfinder) AF points. That is not normal and is an issue.
Yes, I am familiar with the German video that identified the issue.
Rest assured I do plan to use the DotTune method for MFA.
I am very confused reading your responses. I understand the initial issue you mentioned. The one I have heard about and that many people are taking about is.. With fast lenses and single point auto focus (center), the images taken using the viewfinder are out of focus compared to if the same shot is taken with the LCD. If someone does a controlled test (tripod and shutter release/10 second timer) and they focus on the same spot very carefully..images take with viewfinder should be just as sharp as they are when taken using the LCD. Unless I missed something you posted I don't get how someone could disagree that that's an issue.
Because it is NORMAL as long as it is correctable using micro focus adjustment.

Please read this article to understand why: "This lens is soft" and other myths
I understand and I just read that article..the 2 concerns I have are. I don't know about this so correct me if I am wrong. If you take 2 pics. The focus from the live view one is spot on and then the one from the view finder is soft or off...how is that normal?
Because they are using two different focus systems. When looking through the viewfinder the camera uses a separate phase detect AF sensor to determine the distance to the subject, and tells the lens to focus at that distance.

When using the LCD the focus is being done by the image sensor itself.

It is perfectly normal for the PDAF system to be slightly off. In the past the only way to correct this was to send the camera into Canon for calibration. But, with a camera that has micro focus adjustment, you can perform the calibration yourself.

By performing the calibration yourself variations from lens to lens can be adjusted for rather than sending every lens and the camera into Canon for calibration.
If you MFA is that not affecting the live view focus..which is already spot on
No MFA only affects the phase detect (viewfinder) auto focus system
The second thing. If there is something from above that I'm wrong about..and you do MFA...and the focus is still different between the view finder and live view then that's an issue.
If you do MFA adjustment properly there should no longer be a difference between the LCD (image sensor) focus, and the viewfinder focus.
That is what most people are referring to. People are doing controlled tests after MFA and getting the same results. Good focus with love view and bad with view finder.
Then they most likely didn't do the MFA properly.

Again, I highly recommend the DotTune method. And keep in mind that MFA should be done in the light you normally use the lens in, and at the distances you normally shoot at.
Thank you for the quick replies and you cleared up the MFA questions about live view vs LCD...but are you saying that every single person that tests this issue is doing something wrong?
No, but, I am saying that people who think that there is something wrong with the camera because you have to use MFA are wrong.
If someone with experience does MFA and still has the issue you still think it's them and there is no chance of it being the camera?
There is always a chance that it is the camera and not the person.
I don't believe that at all.

Sorry, that person didn't understand the actual issue, and as I recall after micro focus adjusting his lenses he eliminated the issue in most of the lenses. Any remaining issues with a lens or two were most likely contributed to the lenses and not the camera.
It's not that hard to take 2 pictures. If you MFA and still have bad focus with the view finder that is 100% not normal.
No, it is not normal. Done properly MFA should eliminate any variation between the LCD (DPAF sensor) AF and the viewfinder (PDAF) AF.
Just another random thought. You explained that the view finder and the LCD each use a different focus system....Ok so how come with everyones testing it's the view finder that has the issue.
Because the sensor AF uses a combination of the dual pixel AF on the image sensor to determine the distance, and then a final contrast detect (looking at the sharpness of the image on the image sensor) for focus.
It's never live view focus causing problems. You said that MFA is for the view finder but how can the live view focus not need it?
See above.
how can it be spot on with all lenses right from the factory but the view finder is not.
See above.

--
All statements in my posts represent my interpretation of data, research opinion or viewpoints.
The opinions expressed are not representations of fact, and are subject to change without notice.
All images are used for educational purposes.
 
Last edited:
That was the big flaw of that video you linked to, the author didn't seem to understand the real problem that some 70Ds had.

The real AF problem that was experienced by some (primarily early production models sold in Germany) with the 70D was a difference in focus between the center AF point when using the center f/2.8 AF point, and the focus when using the surrounding f/5.6 AF points.
I thought the 70D problem was that (on the cams that have this problem and lenses with max aperture of f/2.8 and lower) with single point center focus the viewfinder focus was "off" when compared with the Live View AF.
NO!!! That is perfectly NORMAL. (see "This lens is soft" and other myths)

It is if the f/2.8 AF point is giving you different focus than the other PDAF (viewfinder) AF points. That is not normal and is an issue.
Yes, I am familiar with the German video that identified the issue.
Rest assured I do plan to use the DotTune method for MFA.
I am very confused reading your responses. I understand the initial issue you mentioned. The one I have heard about and that many people are taking about is.. With fast lenses and single point auto focus (center), the images taken using the viewfinder are out of focus compared to if the same shot is taken with the LCD. If someone does a controlled test (tripod and shutter release/10 second timer) and they focus on the same spot very carefully..images take with viewfinder should be just as sharp as they are when taken using the LCD. Unless I missed something you posted I don't get how someone could disagree that that's an issue.
Because it is NORMAL as long as it is correctable using micro focus adjustment.

Please read this article to understand why: "This lens is soft" and other myths
I understand and I just read that article..the 2 concerns I have are. I don't know about this so correct me if I am wrong. If you take 2 pics. The focus from the live view one is spot on and then the one from the view finder is soft or off...how is that normal?
Because they are using two different focus systems. When looking through the viewfinder the camera uses a separate phase detect AF sensor to determine the distance to the subject, and tells the lens to focus at that distance.

When using the LCD the focus is being done by the image sensor itself.

It is perfectly normal for the PDAF system to be slightly off. In the past the only way to correct this was to send the camera into Canon for calibration. But, with a camera that has micro focus adjustment, you can perform the calibration yourself.

By performing the calibration yourself variations from lens to lens can be adjusted for rather than sending every lens and the camera into Canon for calibration.
If you MFA is that not affecting the live view focus..which is already spot on
No MFA only affects the phase detect (viewfinder) auto focus system
The second thing. If there is something from above that I'm wrong about..and you do MFA...and the focus is still different between the view finder and live view then that's an issue.
If you do MFA adjustment properly there should no longer be a difference between the LCD (image sensor) focus, and the viewfinder focus.
That is what most people are referring to. People are doing controlled tests after MFA and getting the same results. Good focus with love view and bad with view finder.
Then they most likely didn't do the MFA properly.

Again, I highly recommend the DotTune method. And keep in mind that MFA should be done in the light you normally use the lens in, and at the distances you normally shoot at.
Thank you for the quick replies and you cleared up the MFA questions about live view vs LCD...but are you saying that every single person that tests this issue is doing something wrong?
No, but, I am saying that people who think that there is something wrong with the camera because you have to use MFA are wrong.
If someone with experience does MFA and still has the issue you still think it's them and there is no chance of it being the camera?
There is always a chance that it is the camera and not the person.
I don't believe that at all.

Sorry, that person didn't understand the actual issue, and as I recall after micro focus adjusting his lenses he eliminated the issue in most of the lenses. Any remaining issues with a lens or two were most likely contributed to the lenses and not the camera.
It's not that hard to take 2 pictures. If you MFA and still have bad focus with the view finder that is 100% not normal.
No, it is not normal. Done properly MFA should eliminate any variation between the LCD (DPAF sensor) AF and the viewfinder (PDAF) AF.
Just another random thought. You explained that the view finder and the LCD each use a different focus system....Ok so how come with everyones testing it's the view finder that has the issue.
Because the sensor AF uses a combination of the dual pixel AF on the image sensor to determine the distance, and then a final contrast detect (looking at the sharpness of the image on the image sensor) for focus.
It's never live view focus causing problems. You said that MFA is for the view finder but how can the live view focus not need it?
See above.
how can it be spot on with all lenses right from the factory but the view finder is not.
See above.

--
All statements in my posts represent my interpretation of data, research opinion or viewpoints.
The opinions expressed are not representations of fact, and are subject to change without notice.
All images are used for educational purposes.
If you follow up with him and read comments on various sites you will see his issue was never fixed. It got slightly better with MFA but it still very much an issue that he said he just works around. I totally understand that MFA is needed once again I am fine with that. If you do the research you will see that MANY well respected photographers have done MFA and still experience the issue. People have even sent their camera and lenses to canon and still have the issue. EVERY person who buys this camera should do the research..test..MFA..and then test again........ and like I said before I understand live view and view finder use different focus mechanisms.. My question was. Why does live view focus come spot on all the time from every camera maker..why does it not need some type of MFA
 
That was the big flaw of that video you linked to, the author didn't seem to understand the real problem that some 70Ds had.

The real AF problem that was experienced by some (primarily early production models sold in Germany) with the 70D was a difference in focus between the center AF point when using the center f/2.8 AF point, and the focus when using the surrounding f/5.6 AF points.
I thought the 70D problem was that (on the cams that have this problem and lenses with max aperture of f/2.8 and lower) with single point center focus the viewfinder focus was "off" when compared with the Live View AF.
NO!!! That is perfectly NORMAL. (see "This lens is soft" and other myths)

It is if the f/2.8 AF point is giving you different focus than the other PDAF (viewfinder) AF points. That is not normal and is an issue.
Yes, I am familiar with the German video that identified the issue.
Rest assured I do plan to use the DotTune method for MFA.
I am very confused reading your responses. I understand the initial issue you mentioned. The one I have heard about and that many people are taking about is.. With fast lenses and single point auto focus (center), the images taken using the viewfinder are out of focus compared to if the same shot is taken with the LCD. If someone does a controlled test (tripod and shutter release/10 second timer) and they focus on the same spot very carefully..images take with viewfinder should be just as sharp as they are when taken using the LCD. Unless I missed something you posted I don't get how someone could disagree that that's an issue.
Because it is NORMAL as long as it is correctable using micro focus adjustment.

Please read this article to understand why: "This lens is soft" and other myths
I understand and I just read that article..the 2 concerns I have are. I don't know about this so correct me if I am wrong. If you take 2 pics. The focus from the live view one is spot on and then the one from the view finder is soft or off...how is that normal?
Because they are using two different focus systems. When looking through the viewfinder the camera uses a separate phase detect AF sensor to determine the distance to the subject, and tells the lens to focus at that distance.

When using the LCD the focus is being done by the image sensor itself.

It is perfectly normal for the PDAF system to be slightly off. In the past the only way to correct this was to send the camera into Canon for calibration. But, with a camera that has micro focus adjustment, you can perform the calibration yourself.

By performing the calibration yourself variations from lens to lens can be adjusted for rather than sending every lens and the camera into Canon for calibration.
If you MFA is that not affecting the live view focus..which is already spot on
No MFA only affects the phase detect (viewfinder) auto focus system
The second thing. If there is something from above that I'm wrong about..and you do MFA...and the focus is still different between the view finder and live view then that's an issue.
If you do MFA adjustment properly there should no longer be a difference between the LCD (image sensor) focus, and the viewfinder focus.
That is what most people are referring to. People are doing controlled tests after MFA and getting the same results. Good focus with love view and bad with view finder.
Then they most likely didn't do the MFA properly.

Again, I highly recommend the DotTune method. And keep in mind that MFA should be done in the light you normally use the lens in, and at the distances you normally shoot at.
Thank you for the quick replies and you cleared up the MFA questions about live view vs LCD...but are you saying that every single person that tests this issue is doing something wrong?
No, but, I am saying that people who think that there is something wrong with the camera because you have to use MFA are wrong.
If someone with experience does MFA and still has the issue you still think it's them and there is no chance of it being the camera?
There is always a chance that it is the camera and not the person.
I don't believe that at all.

Sorry, that person didn't understand the actual issue, and as I recall after micro focus adjusting his lenses he eliminated the issue in most of the lenses. Any remaining issues with a lens or two were most likely contributed to the lenses and not the camera.
It's not that hard to take 2 pictures. If you MFA and still have bad focus with the view finder that is 100% not normal.
No, it is not normal. Done properly MFA should eliminate any variation between the LCD (DPAF sensor) AF and the viewfinder (PDAF) AF.
Just another random thought. You explained that the view finder and the LCD each use a different focus system....Ok so how come with everyones testing it's the view finder that has the issue.
Because the sensor AF uses a combination of the dual pixel AF on the image sensor to determine the distance, and then a final contrast detect (looking at the sharpness of the image on the image sensor) for focus.
It's never live view focus causing problems. You said that MFA is for the view finder but how can the live view focus not need it?
See above.
how can it be spot on with all lenses right from the factory but the view finder is not.
See above.

--
All statements in my posts represent my interpretation of data, research opinion or viewpoints.
The opinions expressed are not representations of fact, and are subject to change without notice.
All images are used for educational purposes.
If you follow up with him and read comments on various sites you will see his issue was never fixed. It got slightly better with MFA but it still very much an issue that he said he just works around. I totally understand that MFA is needed once again I am fine with that. If you do the research you will see that MANY well respected photographers have done MFA and still experience the issue. People have even sent their camera and lenses to canon and still have the issue. EVERY person who buys this camera should do the research..test..MFA..and then test again........ and like I said before I understand live view and view finder use different focus mechanisms.. My question was. Why does live view focus come spot on all the time from every camera maker..why does it not need some type of MFA
 
That was the big flaw of that video you linked to, the author didn't seem to understand the real problem that some 70Ds had.

The real AF problem that was experienced by some (primarily early production models sold in Germany) with the 70D was a difference in focus between the center AF point when using the center f/2.8 AF point, and the focus when using the surrounding f/5.6 AF points.
I thought the 70D problem was that (on the cams that have this problem and lenses with max aperture of f/2.8 and lower) with single point center focus the viewfinder focus was "off" when compared with the Live View AF.
NO!!! That is perfectly NORMAL. (see "This lens is soft" and other myths)

It is if the f/2.8 AF point is giving you different focus than the other PDAF (viewfinder) AF points. That is not normal and is an issue.
Yes, I am familiar with the German video that identified the issue.
Rest assured I do plan to use the DotTune method for MFA.
I am very confused reading your responses. I understand the initial issue you mentioned. The one I have heard about and that many people are taking about is.. With fast lenses and single point auto focus (center), the images taken using the viewfinder are out of focus compared to if the same shot is taken with the LCD. If someone does a controlled test (tripod and shutter release/10 second timer) and they focus on the same spot very carefully..images take with viewfinder should be just as sharp as they are when taken using the LCD. Unless I missed something you posted I don't get how someone could disagree that that's an issue.
Because it is NORMAL as long as it is correctable using micro focus adjustment.

Please read this article to understand why: "This lens is soft" and other myths
I understand and I just read that article..the 2 concerns I have are. I don't know about this so correct me if I am wrong. If you take 2 pics. The focus from the live view one is spot on and then the one from the view finder is soft or off...how is that normal?
Because they are using two different focus systems. When looking through the viewfinder the camera uses a separate phase detect AF sensor to determine the distance to the subject, and tells the lens to focus at that distance.

When using the LCD the focus is being done by the image sensor itself.

It is perfectly normal for the PDAF system to be slightly off. In the past the only way to correct this was to send the camera into Canon for calibration. But, with a camera that has micro focus adjustment, you can perform the calibration yourself.

By performing the calibration yourself variations from lens to lens can be adjusted for rather than sending every lens and the camera into Canon for calibration.
If you MFA is that not affecting the live view focus..which is already spot on
No MFA only affects the phase detect (viewfinder) auto focus system
The second thing. If there is something from above that I'm wrong about..and you do MFA...and the focus is still different between the view finder and live view then that's an issue.
If you do MFA adjustment properly there should no longer be a difference between the LCD (image sensor) focus, and the viewfinder focus.
That is what most people are referring to. People are doing controlled tests after MFA and getting the same results. Good focus with love view and bad with view finder.
Then they most likely didn't do the MFA properly.

Again, I highly recommend the DotTune method. And keep in mind that MFA should be done in the light you normally use the lens in, and at the distances you normally shoot at.
Thank you for the quick replies and you cleared up the MFA questions about live view vs LCD...but are you saying that every single person that tests this issue is doing something wrong?
No, but, I am saying that people who think that there is something wrong with the camera because you have to use MFA are wrong.
If someone with experience does MFA and still has the issue you still think it's them and there is no chance of it being the camera?
There is always a chance that it is the camera and not the person.
I don't believe that at all.

Sorry, that person didn't understand the actual issue, and as I recall after micro focus adjusting his lenses he eliminated the issue in most of the lenses. Any remaining issues with a lens or two were most likely contributed to the lenses and not the camera.
It's not that hard to take 2 pictures. If you MFA and still have bad focus with the view finder that is 100% not normal.
No, it is not normal. Done properly MFA should eliminate any variation between the LCD (DPAF sensor) AF and the viewfinder (PDAF) AF.
Just another random thought. You explained that the view finder and the LCD each use a different focus system....Ok so how come with everyones testing it's the view finder that has the issue.
Because the sensor AF uses a combination of the dual pixel AF on the image sensor to determine the distance, and then a final contrast detect (looking at the sharpness of the image on the image sensor) for focus.
It's never live view focus causing problems. You said that MFA is for the view finder but how can the live view focus not need it?
See above.
how can it be spot on with all lenses right from the factory but the view finder is not.
See above.
 
Why does live view focus come spot on all the time from every camera maker..why does it not need some type of MFA
Because that process tunes the focus to give the highest contrast on the image sensor. The PD AF process aligns two images on a separate sensor, taking account of a fudge factor provided by the lens that should give the best focus on a perfect body. Many mass produced (uncalibrated) bodies are imperfect, as are many lenses. Microadjustment lets you adjust for the imperfections.
 
I know this makes it tough, but I'd like to see if people who know about this problem can provide any insight.
From these three pics you are fine. The third image shows slightly higher resolution, but the 50/1.8 often needs MFA anyways (plus it's a squirrely focuser to begin with).

You are free to head out shooting. Have fun with the new toy!

R2

--
R2's commented on many of my AF issue threads. I've also got feedback from Canon's technical department over here and quite frankly, as yet they haven't been able to get the lens to do what I expected it to be able to do. MFA with the 50mm f/1.8 is only going to change the range that the AF varies. I don't think you can get consistent results with that lens without live-view.

For portraits, expect to shoot at f/4 and have pentagonal busy bokeh - or wide-open in live-view.

The 50mm f/1.4 by contrast can be set to take shallow DOF portraits reliably with MFA adjusted for each distance, light spectrum and aperture prevailing on the day. When you do this often, it becomes second nature.

To get into this habit, 3 times a day I'd take a walk-around and try and get maximum sharpness shots at different apertures, taking notes of my results.

Again, Live-view is the easiest work-around here - but for moving subjects, you can get by.

Remember, you're probably looking at extracting the very most from the lens in terms of sharpness and low light capability and the lens is a very old design.

What I resort to is a 2 step adjustment. My experience tells me that it's going to be adequately sharp at +14 / +/- 3 most of the time so I'm normally set for +14 - typically it's focusing short so I bump it to +17 and the +19 if it's still not adequate - THEN I RETURN IT TO +14. If I made a mistake, and +17 isn't sharper, I drop to +11 and worst case +9.

I can do all of the above in 60 seconds.

cb
 
That was the big flaw of that video you linked to, the author didn't seem to understand the real problem that some 70Ds had.

The real AF problem that was experienced by some (primarily early production models sold in Germany) with the 70D was a difference in focus between the center AF point when using the center f/2.8 AF point, and the focus when using the surrounding f/5.6 AF points.
I thought the 70D problem was that (on the cams that have this problem and lenses with max aperture of f/2.8 and lower) with single point center focus the viewfinder focus was "off" when compared with the Live View AF.
NO!!! That is perfectly NORMAL. (see "This lens is soft" and other myths)

It is if the f/2.8 AF point is giving you different focus than the other PDAF (viewfinder) AF points. That is not normal and is an issue.
Yes, I am familiar with the German video that identified the issue.
Rest assured I do plan to use the DotTune method for MFA.
I am very confused reading your responses. I understand the initial issue you mentioned. The one I have heard about and that many people are taking about is.. With fast lenses and single point auto focus (center), the images taken using the viewfinder are out of focus compared to if the same shot is taken with the LCD. If someone does a controlled test (tripod and shutter release/10 second timer) and they focus on the same spot very carefully..images take with viewfinder should be just as sharp as they are when taken using the LCD. Unless I missed something you posted I don't get how someone could disagree that that's an issue.
Because it is NORMAL as long as it is correctable using micro focus adjustment.

Please read this article to understand why: "This lens is soft" and other myths
I understand and I just read that article..the 2 concerns I have are. I don't know about this so correct me if I am wrong. If you take 2 pics. The focus from the live view one is spot on and then the one from the view finder is soft or off...how is that normal?
Because they are using two different focus systems. When looking through the viewfinder the camera uses a separate phase detect AF sensor to determine the distance to the subject, and tells the lens to focus at that distance.

When using the LCD the focus is being done by the image sensor itself.

It is perfectly normal for the PDAF system to be slightly off. In the past the only way to correct this was to send the camera into Canon for calibration. But, with a camera that has micro focus adjustment, you can perform the calibration yourself.

By performing the calibration yourself variations from lens to lens can be adjusted for rather than sending every lens and the camera into Canon for calibration.
If you MFA is that not affecting the live view focus..which is already spot on
No MFA only affects the phase detect (viewfinder) auto focus system
The second thing. If there is something from above that I'm wrong about..and you do MFA...and the focus is still different between the view finder and live view then that's an issue.
If you do MFA adjustment properly there should no longer be a difference between the LCD (image sensor) focus, and the viewfinder focus.
That is what most people are referring to. People are doing controlled tests after MFA and getting the same results. Good focus with love view and bad with view finder.
Then they most likely didn't do the MFA properly.

Again, I highly recommend the DotTune method. And keep in mind that MFA should be done in the light you normally use the lens in, and at the distances you normally shoot at.
Thank you for the quick replies and you cleared up the MFA questions about live view vs LCD...but are you saying that every single person that tests this issue is doing something wrong?
No, but, I am saying that people who think that there is something wrong with the camera because you have to use MFA are wrong.
If someone with experience does MFA and still has the issue you still think it's them and there is no chance of it being the camera?
There is always a chance that it is the camera and not the person.
I don't believe that at all.

Sorry, that person didn't understand the actual issue, and as I recall after micro focus adjusting his lenses he eliminated the issue in most of the lenses. Any remaining issues with a lens or two were most likely contributed to the lenses and not the camera.
It's not that hard to take 2 pictures. If you MFA and still have bad focus with the view finder that is 100% not normal.
No, it is not normal. Done properly MFA should eliminate any variation between the LCD (DPAF sensor) AF and the viewfinder (PDAF) AF.
Just another random thought. You explained that the view finder and the LCD each use a different focus system....Ok so how come with everyones testing it's the view finder that has the issue.
Because the sensor AF uses a combination of the dual pixel AF on the image sensor to determine the distance, and then a final contrast detect (looking at the sharpness of the image on the image sensor) for focus.
It's never live view focus causing problems. You said that MFA is for the view finder but how can the live view focus not need it?
See above.
how can it be spot on with all lenses right from the factory but the view finder is not.
See above.
. . . and like I said before I understand live view and view finder use different focus mechanisms.. My question was. Why does live view focus come spot on all the time from every camera maker..why does it not need some type of MFA
Because LiveView focus is done right off the image sensor that actually takes the photo so there is no need to calibrate it.

Viewfinder focus is done by a separate focus sensor with an image that follows a different path. So it has to be calibrated to align with the image sensor itself.

Image from photographylife.com

Image from photographylife.com



--
All statements in my posts represent my interpretation of data, research opinion or viewpoints.
The opinions expressed are not representations of fact, and are subject to change without notice.
All images are used for educational purposes.
 
I was seriously considering a 70D, because of the touch screen, the great new live view focusing,and especially what seems to be a much more user friendly system than my Nikon D7100. However, this and another thread, as well as some youtube videos, have scared me. I'm very disappointed, because I was really looking forward to a new toy. I'm certainly glad I learned what appears to be a fiasco now, and not after I pulled the trigger. I don't need headaches and uncertainty and questions. When I get a poor result, I want to be sure it was due to user error, not the camera.
 
I was seriously considering a 70D, because of the touch screen, the great new live view focusing,and especially what seems to be a much more user friendly system than my Nikon D7100. However, this and another thread, as well as some youtube videos, have scared me. I'm very disappointed, because I was really looking forward to a new toy. I'm certainly glad I learned what appears to be a fiasco now, and not after I pulled the trigger. I don't need headaches and uncertainty and questions. When I get a poor result, I want to be sure it was due to user error, not the camera.
As another poster pointed out, Canon has sold 100's of thousands of 70D's, yes there are going to be a few cameras with issues, just as I'm sure there are some D7100's with issues.

Buy whatever camera you are comfortable with, but, personally I would have no hesitation buying a 70D. I know from past experience if in the extremely rare chance there is something wrong with the camera Canon will take care of it quickly and without hassle.

Roger Cicala who is in charge of maintaining hundreds of cameras and lenses at LensRentals.com, chose a Canon for his personal camera and their support was part of the reason.

Roger Buys a Camera System: Finally ~ by Roger Cicala, LensRentals.com

I mentioned in the first post that I'm a fanatic about customer service and repairs. That's a big edge to Canon USA compare to Nikon USA right now (it's different in different countries), and right now is when I'm making my decision. Fanboys can go off as much as they want, but I handled several thousand repairs last year. Nikon takes, on average, three times as long at double the cost. (Lensrentals insider joke: What do you call a D800 with a scratched sensor? Parts. Because at $1,800 for a sensor replacement . . . )

I haven't had many Pentax repairs so I looked into their service and got good news and bad. The good was a really nice customer support system with live chat and quick, knowledgeable phone support. The bad news was when I asked how long repair turnaround time would be the answer was 20 to 30 days. So I ranked them between Canon and Nikon.

My Personal Rant

This may not bother you at all, but it does me, at least a bit. If I buy Nikon right now, I'm in effect saying it's fine that you stopped selling parts, made most of the independent shops stop working on Nikon, upped your repair prices, slowed your repair service, and lowered quality control. You were right to do so, because here's my money.


--
All statements in my posts represent my interpretation of data, research opinion or viewpoints.
The opinions expressed are not representations of fact, and are subject to change without notice.
All images are used for educational purposes.
 
Last edited:
With 50mm 1.8 it is impossible to get precise focus wide open. I tried 3 copies, all of them were the same story. Depending you go from infinity or close focus, you get different focus. Just try this experiment to be convinced. Open live view, magnify to 100%, choose quick focus , (so that camera focuses with mirror) and focus on the target. Then every time you get different focus, because the focus motor and mechanism wiggles sufficient enough to create back/front focus issue.
 
When I bought my 50mm f/1.8 I went through three copies for exactly this problem. The third one did not exhibit this variability so I kept it. While this was a very common problem with these lenses, not every single one had it.
 
Probably, but just worth trying with different lenses that has >2.8 max aperture. If you do not have friends to borrow, just rent usm-motored one.
 
With 50mm 1.8 it is impossible to get precise focus wide open. I tried 3 copies, all of them were the same story. Depending you go from infinity or close focus, you get different focus. Just try this experiment to be convinced. Open live view, magnify to 100%, choose quick focus , (so that camera focuses with mirror) and focus on the target. Then every time you get different focus,
Correct observation (of typical behaviour).
because the focus motor and mechanism wiggles sufficient enough to create back/front focus issue.
Incorrect interpretation. It's an optical problem - see The Canon EF 50mm f/1.8 II Has a Crude Focus Mechanism in Busted! Digital Photography Myths.
 
That makes sense! Thanks
 

Keyboard shortcuts

Back
Top