silver blade
Member
Not sure if such problem is possible. For example, focus on the eye, but the rope was focused instead? I am facing similar issue with the E-M1. Now being looked at, at Olympus sg.
OMG! Look at the EM-1 aka the now official E-M1 Thread
- Thread starter hammeryou
- Start date
tomcat
Senior Member
If the focal plane of the rope is slightly in front of the focal plane of the eyes, then the rope could be in focus despite the user focusing on the eyes if the lens has a front focusing problem. This could be easily confirmed if the image shows the focusing square is on the eye but the eyes are slightly out of focus but the rope is sharply focused.
You do need to bring both the camera body and the lens in question down to the Service Centre if you want to check for back/front focusing problems.
Last edited:
silver blade
Member
Thanks.
I have similar problem eg. focus on the eye but eye is out of focus but slightly further down the neck focused. Shall wait to hear from Olympus.
I have similar problem eg. focus on the eye but eye is out of focus but slightly further down the neck focused. Shall wait to hear from Olympus.
NazgulKing
Senior Member
That's assuming the camera figured properly where best to focus. My experience with the EM5 was that the AF prediction is pretty retarded.
tomcat
Senior Member
Front & back focusing is a lens issue affecting only that particular lens and that particular camera body usually due to slight variations in tolerances between different lenses of the same make & model. The problem is endemic for some brands and rare in others. Don't think it has anything to do with what kind of auto focusing method is employed.
The fact that Olympus has included autofocus micro-adjustment function in their cameras means that this problem could exist for these cameras.
The fact that Olympus has included autofocus micro-adjustment function in their cameras means that this problem could exist for these cameras.
Last edited:
Bro, this lies in the software right ? Not a lens issue ???
Bro thanks for sharing. I was thinking the cdaf, focus will lock when the lens reach the best contrast.
always thought The micro adjustment will only be applicable to mfd or infinity problem
Didn't know lens will cause af issue on cdaf. But good to learn.
Actually I think Ayuready is right. CDAF shouldn't have back/front focus issues if I understand the way it works correctly. It's analysing the images for highest contrast and stops when it achieves this.
PDAF in traditional DSLRs have a different optical path to the sensor as ayuready've pointed out, the module is separate so alignment issues cause slight AF discrepancies.
On-sensor PDAF should mitigate this problem since the sensor is now on the sensor plane and not elsewhere but as Tomcat pointed out, there are still lens-side AF accuracy issues.
I believe the micro-adjustment on the EM1 is for the PDAF sensors only, not CDAF. Exactly how the on-sensor PDAF works with the 4/3 lens to produce slight inaccuracies I'm not exactly sure though.
PDAF in traditional DSLRs have a different optical path to the sensor as ayuready've pointed out, the module is separate so alignment issues cause slight AF discrepancies.
On-sensor PDAF should mitigate this problem since the sensor is now on the sensor plane and not elsewhere but as Tomcat pointed out, there are still lens-side AF accuracy issues.
I believe the micro-adjustment on the EM1 is for the PDAF sensors only, not CDAF. Exactly how the on-sensor PDAF works with the 4/3 lens to produce slight inaccuracies I'm not exactly sure though.
tomcat
Senior Member
If that's true, then there should be no front/back focusing issues with the pdaf of the E-M1 too since the pdaf detection sensors are pixels on the sensor itself and not in a separate module away from the sensor. Front/back focusing problem is caused by slight variations in the distance between the lens and the sensor surface. In any case, it is very easy to determine if one's lens has this problem. Just get and shoot the relatively cheap Spyder LensCal which is a miniature version of the setup used by various camera serve centers to calibrate lenses with front/back focusing problems. Alternatively just shoot a ruler inclined at 45 deg angle to check for front/back focusing problems and satisfy yourself once and for all one way or another.
Yep.. I'm sure the slight variations from the back of the lens is a factor too and it's still consistent with what I wrote.
That variation results in an optical path distance to the PDAF module (in a DSLR) that is different to the distance to the sensor when the mirror flips up. So even a very well aligned AF module in a DSLR can result in slight discrepancies due to the lens. But a poorly aligned PDAF module makes things quite bad, in fact this is very likely the cause of the D800 left side AF issues.
And I also agree that on-sensor PDAF should really solve this issue since, as you said the pixels are the sensors (ie. same plane). Incidentally on-sensor PDAF should also solve any back of lens to sensor variation problems too since readings are taken off the sensor plane now.
It seems there's more to how PDAF work with PDAF lenses that results in these small discrepancies but I can only speculate why this is. It is also doubtful current on-sensor PDAF are as good as their dedicated module predecessors so that could also be a factor. There just isn't enough literature on the on-sensor PDAF designs yet.
silver blade
Member
To update. The out of focus issue is almost certain to be a problem with the specific body (A) and not a lens issue. Tired a couple other lens with body A and several shots were out of focus. Use those same lens on body B (also an E-M1)and every shots were accurately focus.
Yet to heard from Olympus what is wrong with body A.
tomcat
Senior Member
That's the same symptoms for those cameras & lenses with front/back focusing problems.
A particular lens may exhibit front/back focusing problem with Camera Body X but can be completely OK with Camera Body Y. Every camera body and lens have an acceptable design/construction tolerance error of +/- x%. When a camera and lens is used together, the combination error may be magnified if the magnitude and direction (as in + or -) are similar. Almost every Sigma lens suffers from some front/back focusing problem from my experience. Nowadays if I buy any Sigma lenses, I would check the extent of the problem with my Spyder LensCal and correct for them with the camera's AF micro-adjustment function or send the camera and lens in to the Service Centre for calibration.
Typically a camera body may not exhibit front/back focusing problems with every lens used but if it does like in your case, then the tolerance error may be too high to the extent that it affects all the lenses you tested and should be brought in to the Service Centre for repair or even replacement.
May I ask what lens did you try? specifically m43 or 43 lens?
silver blade
Member
All the lens I tried on it were M4/3 lens eg. 17mm f1.8, 25mm f1.4, 12-40mm f2.8, etc.
Ok on E-M5 & another E-M1 body.
Ok on E-M5 & another E-M1 body.
Last edited:
silver blade
Member
Thank you for the detailed explanation.
Thank you for the detailed explanation.
Sounds like there's a whole lot of engineers here
microcosm
Senior Member
Sounds like there's a whole lot of engineers here
Olympus attracts geeks.
=.= I'm majored in pure math and physics ... So there's no way I'm disagreeing with u here
