front focus issue with *manual* focus

I can't tell if this is the same problem others refer to as front focusing or not, but given that a debug menu setting is usually suggested as a fix, I'm guessing not, because unless this setting physically moves the sensor further froward or backward, I can't imagine how any setting would affect what I'm seeing.

What I am finding, after a few years with my *ist DS, is that manual focus is less reliable than I thought. I first became aware of it checking out an M50/1.4 in a shop this morning, and have since been experimenting with my other lenses (including M50/1.7 and DA40/2.8) and finding that it's pervasive, although not to as dramatic an effect as it was with the 1.4 (presumably, because shallow DOF magnifies this effect).

I have my diopter set correctly (the framing grid is in focus). I focus on a subject. My eyes tell me it is in focus, and the focus indicator agrees. The picture taken tells me otherwise. The area of sharpest focus is in front of the subject (and shooting a newspaper or ruled or the floorboards at an angle verifies this). The extent of the discrepancy varies from an inch or two to several depending on subject distance and aperture.

In the case of the M50/1.4 wide open, the DOF is so shallow that the discrepancy rendered my test pictures useless. With a subject about 10 feet away, the actual plane of sharpest focus was around 2 feet in front of the subject, putting the subject hopelessly out of focus. I'm thinking that with other (slower) lenses, the DOF was deep enough such that I had not noticed this before, but it was pretty extreme in this case, and I have no reason to assume that this is new behavior, but again, I can detect it with any reasonably fast lens.

I've done some searching around, and other threads in other forums have pretty much either expressed bewilderment that it was even possible for manual focus to exhibit these types of focus problems (with an implied "you just don't know how to focus properly), to suggests that a different type of focus screen would help, or else resignation that this is just how things are with fast lenses.

So sorry if this is old news here, but can anyone here able to shed more light on this?

--
Marc Sabatella
http://www.marcsabatella.com/photo/

And back focus also. This phenomena (manual lens, viewfinder in focus, huge patch of focus behind the desired subject) is the low point of my existence as a photographer.

I know I haven't contributed much to the conversation with this reply, but I too am keen to hear what others have experienced.

http://www.flickr.com/photos/red3photography

Pretty much got the same issue here, albeit back focusing. Before I corrected the focus in Debug mode, when it looked in focus to me, the AF locked as well but the image was back-focused. Now that I adjusted it, the AF is bang on, but the image does not appear to my eye as being in focus. Therefore, I always use AF assist in Manual Focus mode to confirm my focus point. Not certain if the viewfinder (or sensor front and back position) can be calibrated so the AF and MF are in synch or not. I'd certainly would very much like to hear if anyone had this issue solved by Pentax service dept. Before I'd send it though, I'd like to have someone with good eyes (I wear glasses) confirm the issue. It could be my eye glasses, in spite of trying to compensate with the diopter.

Here a new theory... are you near sighted? I am far sighted and got back focus and if you are near sighted and others as well with the same issue, perhaps this is an issue amongst other thing.

--
Roger

Hi Marc,

I think you are experiencing an example of bad quality control. I also think this is a choice made by the manufacturer.

When I used high quality manual 35mm film SLR cameras I never had a problem with front- or back focusing. My eyesight was better back then but this is not about eyesight.

The viewfinders were more important those days as they were for made for both framing and focusing. So, they were checked with care. I have owned five dSLR cameras. Two of them have been good, one a bit so-so and two clearly off (with regards to having the focusing screen at the exact same optical distance from the lens as the sensor. Two of them have been tilted (the E-510 the worst one) to that.

The focusing indicator is sloppy and can give you a green light over a pretty wide area. It is not to be trusted at all.

If you have got the dioptre setting correctly (aiming towards a white wall with the lens unfocused and the viewfinder information clear and sharp without any eye strain) you have done your part. I attached a viewfinder enlarger to the cameras I suspected to be out of alignment and that confirmed it. This is not a necessary step though.

If you are very interested in improve the focusing you can try using shims (or thin tape at the edges) at the viewfinder screen trying to force it to be repositioned a tiny bit.

Another option (if you are in the USA) is to send the camera to KatzEye and ask for one of their viewfinder screens. They will adjust "the focusing" when switching the screen for you.

If you are in Sweden there is no use complaining to Pentax Scandinavia. They found my ugly VPN and viewfinder misalignment to be within specifikations and even enclosured a printed picture when they returned the camera. (A ruler with a mark clearly in focus "Taken with our reference lens, a 50mm macro".)

Well, that's about it. With my E-510 I just live with the problem and use Live View when critical focusing is necessary.

my 2 cents,

--
Jonas

Marc Sabatella said:

-snip-
I've done some searching around, and other threads in other forums
have pretty much either expressed bewilderment that it was even
possible for manual focus to exhibit these types of focus problems
(with an implied "you just don't know how to focus properly), to
suggests that a different type of focus screen would help, or else
resignation that this is just how things are with fast lenses.

So sorry if this is old news here, but can anyone here able to shed
more light on this?

Click to expand...

FWIW I never have had that problem with my DS, it has always been right on.

Tom

'The difference between genius and stupidity is genius has its limits.'

• Albert Einstein

I had the same problem with my istD.

Perfect AF but wrong MF.

I noticed this at a wedding shoot. The AF pics came out fine, when I used MF they were slightly out of focus, although they looked right in the viewfinder (I have good eyes, don't wear glasses).

Again the AF ones, did not look perfectly in focus in the viewfinder, although they were in reality!

Then it dawned on me that there was no misalignment of the AF sensors, there was no calibration needed, I just had to reposition slightly the focusing screen.

I put a shim (2 stripes of paper tape on the edge of the short sides of the screen) above the focusing screen and this fixed the problem 100%. I don't remember if I had MF front focusing or back focusing. You might have to put the shim under the screen.

Don't worry, you'll fix it alright.

Greetings

--
Myon
i-s-t-D, grip, three sixty flash

DA fourty mm, FA seventy - two hundred ten, DA eighteen - fifty five, DA ten - seventeen, A fifty f/1.7, DA fourteen mm

Well I am near sighted, shooting a 100mm macro manually I almost always back focus on my K10, but compensating for that has been really hard. I'll have to try the shims idea.

I've been tempted to muss with the debug menu. Autofocus seems to be ok most days, though sometimes I don't know if its my eyes or the camera, but I do wonder if a Katz would help.

-Dave

I noticed some minor but strange focussing issues recently with a manual 50mm 1.4 lens at night. Viewfinder looked in focus to my eyes but many pictures were out of focus. My autofocus lenses produce well focussed images generally, after a little adjustment to my DS2 via debug. I do notice however that when i use autofocus lenses they don't necessarily appear correctly focused in the viewfinder. I have set the diopter correctly but now I am wondering if I adjust the dipoter just a little in one directrion or another could this be a way to compensate for the slight inconsistency I am noticing between AF & viewfinder . Don't have my camera at work to try it but does this make any sense to try? (Probably will try it anyway as it can't hurt anything)
--
C Evans

You won't be able to correct the problem with the diopter adjustment. All you'll accomplish is making it more blurry.

The reason is pretty simple when you think about it. Looking through the viewfinder, you're actually looking at a little rear projection screen (the focus screen), which sits under the pentaprism/pentamirror. The diopter adjustment helps to focus your view of this screen from the front side; the problem is that the lens projecting onto the screen's back side isn't properly focused. It's exactly equivalent to going to a movie theater where the projectionist has not properly focused the projector and attempting to use a wrong eyeglass prescription to compensate. You'd only succeed in seeing an even more out-of-focus movie (which, for some films, might not be such a bad thing after all).
--
--DrewE

Thanks! Makes lots of sense.
--
C Evans

Myon said:

Then it dawned on me that there was no misalignment of the AF
sensors, there was no calibration needed, I just had to reposition
slightly the focusing screen.

I put a shim (2 stripes of paper tape on the edge of the short sides
of the screen) above the focusing screen and this fixed the problem
100%. I don't remember if I had MF front focusing or back focusing.
You might have to put the shim under the screen.

Don't worry, you'll fix it alright.

Click to expand...

Thanks, this is very encouraging. I realized that the issue must be a difference in relative distance from the lens to the focus screen versus the sensor, which is why I knew a debug menu setting wasn't going to cut it. Several people have now suggested repositioning the screen with shims, and this seems perfectly logical.

Actually, I should have mentioned (!) that the night before I discovered the problem, I had removed the screen to clean the viewfinder. There were darks spots on the LCD display at the bottom, but not to the point where one whole LCD segment was dark. I can't say that i figured out where the problem was, but a few puffs of air definitely fixed it. It is entirely possible, if not likely, that the screen was mispositioned slightly when I closed it up. I actually did open it up again to check shortly after discovering the problem, and didn't see anything amiss, but didn't really change anything before closing it and doing more tests.

BTW, I have an M135/3.5 on order from KEH, should be arriving tomorrow. I know it's not the best 135 ever made, and only a stop faster than my DA50-200 at that aperture, but I'm still looking forward to having it - I've been leaving my 135/2.8 at home much of the time because I don't feel like carrying that extra pound everywhere I go.

Anyhow, any thoughts on whether the DOF of the 135/3.5 will be smaller (because of the longer focal length) than my 50/1.7? The smaller the DOF, the easier to test my focus issues.

Marc Sabatella said:

-snip-
Anyhow, any thoughts on whether the DOF of the 135/3.5 will be
smaller (because of the longer focal length) than my 50/1.7? The
smaller the DOF, the easier to test my focus issues.

Click to expand...

You don't need anything else but your 50/1.7 to check the focusing.

--
Jonas

Never had the problem with my DS (especially with the Katz-eye), but while the DS is in the shop I've been using a loaner DL which has a few strikes against it:

1. No Katz-Eye! (Sob)
2. Penta-mirror viewfinder is smaller and darker

3. No matter how careful I am setting up my manual focus, I'm always "actually" focused a few inches behind what looks sharp on the screen.

If it were my own camera and not a loaner, I'd have it into the shop in a heartbeat to have an expert line things up. As it is, I'm just counting the days that my "real" camera comes back from being repaired!

-Charles

FWIW, I have been trying the shim solution, without much luck. It appears to be the case that shims employed to put the focus screen further from the prism (closer to the mirror) would be an effective fix to a backfocus problem, but for frontfocus, it seems I need to get the screen closer to the prism (further from the mirror), and a shim between the screen and the tray that holds it is not having that effect - this distance appears more fixed. There does seem to be some wiggle room in there where the tray is hinged and where it latches, and I might eventually be able to sort that out. But at this point, it might be time to stop obsessing about this (!) and go back to using the camera as is to see if the problem starts bothering me in practice - as I said, it never did before, and I suspect nothing has actually changed.

As mention earlier I had a back-focus issue with the k10d, particularly with tele lenses (a 300/2.8 practically lives on the camera body). I put some shims (strips of plastic down the short sides of the focus screen) in place, effectively moving the focus screen closer to the mirror (as the previous message mentions), and the back focus has largely disappeared. At least at F4.

Jonas B said:

Hi Marc,

I think you are experiencing an example of bad quality control. I
also think this is a choice made by the manufacturer.

When I used high quality manual 35mm film SLR cameras I never had a
problem with front- or back focusing. My eyesight was better back
then but this is not about eyesight.
The viewfinders were more important those days as they were for made
for both framing and focusing. So, they were checked with care. I
have owned five dSLR cameras. Two of them have been good, one a bit
so-so and two clearly off (with regards to having the focusing screen
at the exact same optical distance from the lens as the sensor. Two
of them have been tilted (the E-510 the worst one) to that.

The focusing indicator is sloppy and can give you a green light over
a pretty wide area. It is not to be trusted at all.

If you have got the dioptre setting correctly (aiming towards a white
wall with the lens unfocused and the viewfinder information clear and
sharp without any eye strain) you have done your part. I attached a
viewfinder enlarger to the cameras I suspected to be out of alignment
and that confirmed it. This is not a necessary step though.

If you are very interested in improve the focusing you can try using
shims (or thin tape at the edges) at the viewfinder screen trying to
force it to be repositioned a tiny bit.

Another option (if you are in the USA) is to send the camera to
KatzEye and ask for one of their viewfinder screens. They will adjust
"the focusing" when switching the screen for you.

Click to expand...

hi Jonas,
thanks for your reply, I've been thinking about this problem too

--
kind regards
Sune

There sure must be a better way to adjust either the viewscreen position, mirror or the sensor front-back position. This must be adding pressure on the viewscreen frame, and I'd be concerned of what would/will happen if/when the camera heats up in the summer.

Does anyone had Pentax adjusting either of these ? i.e. are these things set in place with varying degree of tolerance, or can they be adjusted ? (either viewscreen, mirror, and/or sensor)

My AF is dead accurate; albeit after I applied a 140 correction, but MF really bad.

--
Roger

Funny that some people think live view cannot possibly be useful on a DSLR.

I cannot stand not seeing the focus and exact field of view on my G9 camera's viewfinder so I use live view whenever I can.

When you think of it even tiny pocket cameras have through the lens view (via live view LCD on the back) which used to be one of the big advantage of SLRs. Now it is mostly being able to choose lenses and fast shot to shot operation. I don't understand why fast shot-to-shot has to be reserved for DSLRs - my G9 not having frame buffers makes no sense for a $500 camera.

Jonas B wrote:
...

Jonas B said:

Well, that's about it. With my E-510 I just live with the problem and
use Live View when critical focusing is necessary.

Click to expand...

I've not heard of this before. Did it happen much with the *ist series?

Ben

--



http://www.luzartphotography.com
http://luzart.blogspot.com

guyv said:

Funny that some people think live view cannot possibly be useful on a
DSLR.
I cannot stand not seeing the focus and exact field of view on my G9
camera's viewfinder so I use live view whenever I can.

Click to expand...

Well, yeah, but the fact that the viewfinder is so small on the G9 - even if bigger than many other non-DSLR cameras - pretty much guarantees people will resort to live view.

On the other hand, sure, I thought of this - live view would help. But still, given the choice, I'd much rather have an accurate viewfinder.

--
Marc Sabatella
http://www.marcsabatella.com/photo/