Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
01-12-2009, 05:02 AM   #31
Pentaxian
wildman's Avatar

Join Date: Jun 2008
Location: Flyover America
Posts: 4,310
QuoteOriginally posted by Class A Quote
Did you intend to write "MF" instead of "AF"? Surely it was an AF problem.


Really? Can I read that somewhere? That sounds too simplistic for an fault that persisted embarrassingly long. Have you read about the 10/10 cameras that didn't work for a reviewer? Then another hand selected 2/2?
No what I was talking about was a MF problem only.

This problem only effected a few camera bodies I think. I believe the problem I'm talking about is apart from what you are talking about.

I know the guy that had the problem when using his 600mm Canon glass. He was just not getting the quality in MF that he was used to with his old body.

In fairness to Canon if memory serves...
he called Canon directly and they said they were aware of this possible problem but that it was difficult to track down which bodies were bad and which not because of the high level of hand work on these bodies. There was no serial run of bad bodies as would be the case with automated production. It just seemed to pop up infrequently and randomly as you would expect with hand assembly. Canon sent him a new body overnight air express no questions asked. Just send them back the old body when convenient.

So far as I know he's a happy camper now.


Last edited by wildman; 01-12-2009 at 05:12 AM.
01-12-2009, 05:31 AM   #32
Pentaxian




Join Date: Apr 2007
Location: Sweden
Posts: 1,883
QuoteOriginally posted by Marc Sabatella Quote
The body turns the screw. But the lens has to respond to that screw-turning correctly. The body does an initial check, determines how far focus is off and in which direction, then calculates how many turns of the screw it needs to fix it. If the lens is "off" in some way, it won't be in focus after the "correct" number of turns.



Kind of, sort of, but it doesn't really check that hard, it seems - at least, not once it gets enough info to decide how much adjustment is needed.
The thing is though that if you press AF again it will not move in to correct focus, its happy with the incorrect focus it got the first time. Also, if you manually correct it and then press AF, the motor will move it out of focus and then light up the focus confirmation.

I had this problem with a Sigma 17-70 but only at the wide end, sent it in and they corrected it.
01-12-2009, 11:44 AM   #33
Pentaxian
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,686
QuoteOriginally posted by Gimbal Quote
The thing is though that if you press AF again it will not move in to correct focus, its happy with the incorrect focus it got the first time. Also, if you manually correct it and then press AF, the motor will move it out of focus and then light up the focus confirmation.

I had this problem with a Sigma 17-70 but only at the wide end, sent it in and they corrected it.
OK, but that, then, would not be the specific type of defect I was talking about.
01-12-2009, 11:54 AM   #34
Pentaxian
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,686
QuoteOriginally posted by wildman Quote
Something I have always wondered about:

I hardly ever shoot AF. Most all of my shooting is through 500mm and more scopes using manual focus.

Considering that the light path when focusing through the viewfinder is different from the light path to the sensor when the shutter is released has there been any reports of mis-calibration, however slight, between the two light paths in the past?
Oh yes, absolutely. It's why shims are often needed with third party focusing screens.

For a while - before I realized the problem was really that the DOF in the viewfinder was greater than that of the image for wide apertures - I was convinced my DS had a misaligned focus screen, as I kept getting FF with manual focus on my M50/1.7. After unsuccessful attempts to shim/realign the focus screen myself, I realized I could correct a BF issues this way but not FF. So I sent the camera in for service. They sent it back saying there was nothing wrong. Luckily, in the interim I had bought a K200D and done a ton more testing of every camera I could get my hands on, and enough reading to finally understand the ramifications of the DOF issue. And it didn't take but a few minutes of testing on getting the camera back to realize the service folks were right - there was nothing wrong with the camera at all.

Anyhow, my camera may not have actually had a problem, but I sure thought it did, and at the time found other discussions of the issue (on dpreview, mostly). But you're right, it is not discussed very often - in fact, it's surprising how often people take for granted that focusing manually is always a foolproof workaround for issues with BF/FF.

01-12-2009, 12:48 PM   #35
Senior Member




Join Date: Dec 2008
Location: Quebec
Photos: Gallery | Albums
Posts: 210
Original Poster
Well... well... well... You guys just lost me. Anyhow, since I posted this question, I used my DA35 macro limited a few times and found out that on average, it focuses just as well as my other lens. I did not try the focus chart again and don't plan to do so either. I guess I figured out it's much more satisfying to go out and take shots than stay inside and take pictures of focus chart. Now, I just hope you did not spend too much of your time answering it!

Thanks again for all the good information. I sure learned a lot from you guys.
01-12-2009, 01:20 PM   #36
Veteran Member
Ben_Edict's Avatar

Join Date: Jul 2007
Location: SouthWest "Regio"
Photos: Albums
Posts: 3,303
QuoteOriginally posted by Class A Quote
Only in the sense that it would stop working before it got (too) inaccurate.

If I understand the split-prism system correctly (whose principle is also used in AF systems) then any deviation from a "tuned" frequency, will result in an error. By narrowing the frequency you only make it stop earlier and allow for better phase matching as there isn't as much overlap of signals (assuming the AF sensors themselves cannot distinguish between colours).
No, it is not that easy. Common photographic lenses (indeed any common lens for imaging purposes like binoculars or refracting telescopes) are optimized to have the least colour aberration at around 550nm wavelength. This is the wavelength, where at least two aberration curves (usually blue and red) meet at a common focus (achromatic lenses). Apochromatic lenses add a third line there (green or yellow). So around 550nm focus is optimized and it makes sense to also optimize the AF-system's response for that wavelength.

QuoteOriginally posted by Class A Quote
Everything is small here, isn't it? I'm sure the difference in phase will be very, very small even for considerable misfocus.
That is too true. But measuring a phase change or match is easier, than any mechanical adjustment (i.e. lens movements, falnge distance etc.)

QuoteOriginally posted by Class A Quote
I haven't heard anyone complaining about slow focusing under tungsten light. Under low light, yes, but the common problem with tungsten light is BF, not speed or inconsistency.
Oh, this variable FF/BF is exactly the problem that FINN reports in this thread. With "slow" focusing I mean hunting around, going forwards and backwards several times. I think, you will find, that this low light level issue will be more pronounced under Tungsten lighting.

QuoteOriginally posted by Class A Quote
Fine, but the usable portion of the spectrum would still be far enough away from the tuned frequency to cause an error. At least that's a possiblitiy.
I don't think so. Longer wavelengthes (if we assume that the Fuji AF sensor's spectral response is to a certain degree representative of AF-sensors in general) are not only suppressed to a 60 % sensititvity level, but we should also take into account, that these longer wavelengthes are less energetic, than shorter wavelengthes. So the AF-sensor might (it is all only assumption) be less excited by the reds.

I would think, that first and foremost the Tungsten light is too red and therfor too dim, to allow reliable AF-function. Secondary it might be possible, that at wide apertures, the far reds will be not focused correctly (with a standard achromatic lens), as the AF-sensor refers to a peak wavelength of around 550nm for focusing. As the reds are dominant under Tungsten, the whole image will thus look misfocused.

It would be interesting to see, whether this misfocusing will persist, when using an apochromatic or even better super-apochromatic lens. As Apos usually are of longer focal length (because under normal circumstances achromatic lenses are sufficiently corrected at all focal lengthes up to around 200mm fl, I can't so this test myself. My lenses with "Apo" designation are of limited use indoors (the Sigma 70-200/2.8 being the shortest) and anyway I don't trust such a designation...

QuoteOriginally posted by Class A Quote
Why would it without taking prism dispersion into account? What else explains a consistent BF? I don't see how a loss of intensity (usable light for AF purposes) does that.
First: I have never seen this consistent BF, certainly not with the lenses and cameras I have used so far. And FINN farther up the thread also reports inconsistent, variable BF7FF issues under Tungsten lighting. I will not say, that this rules out any influence by the prisms, but it at least asks for other explanations, too.

Secondly: I cannot decide, whether the prisms influence is as strong or weaker or even more pronounced, than the influence of CAs by standard achromatic lenses. That is why I left that question open. I think, that lens influence is highly variable and might be the more decisive factor - but in any case much less important than the colour temperature.

Ben
01-12-2009, 02:21 PM   #37
Pentaxian
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 8,934
QuoteOriginally posted by Marc Sabatella Quote
OK, but that, then, would not be the specific type of defect I was talking about.
I'm not so sure. I believe he was describing the scenario where a lens has so much of FF that even when focused spot on (say through manual adjustment), it will move out of focus to achieve its FF (what it thinks is perfect focus).

I believe this will actually occur and is not a symptom of a different problem but would be evidence of the fact that the camera lens combination actively and willingly sets the incorrect focus because it determined it to be the supposedly perfect one.
01-12-2009, 02:32 PM   #38
Pentaxian
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 8,934
Ben, thanks for your further input.

QuoteOriginally posted by Ben_Edict Quote
Secondary it might be possible, that at wide apertures, the far reds will be not focused correctly (with a standard achromatic lens), as the AF-sensor refers to a peak wavelength of around 550nm for focusing. As the reds are dominant under Tungsten, the whole image will thus look misfocused.
That's an interesting thought. In other words, the AF system focussed correctly for an image with a standard colour temperature, but the image is so red that the lens doesn't exactly project it onto the sensor plane.

Doing the test with an APO lens should indeed valid this hypothesis.

QuoteOriginally posted by Ben_Edict Quote
First: I have never seen this consistent BF, certainly not with the lenses and cameras I have used so far.
Other forum members have. Most prominently RiceHigh has devoted part of his blog to this issue. It is actually FF (not BF). Sorry for mixing this up earlier.

QuoteOriginally posted by Ben_Edict Quote
I cannot decide, whether the prisms influence is as strong or weaker or even more pronounced, than the influence of CAs by standard achromatic lenses.
I cannot either and I sometimes wonder why I'm allowing myself to get occupied with thinking about this issues without having proper access to the technical details and/or means to validate hypotheses. Well, I hope I'm just curious.

01-12-2009, 03:02 PM   #39
Pentaxian
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,686
QuoteOriginally posted by Class A Quote
I'm not so sure. I believe he was describing the scenario where a lens has so much of FF that even when focused spot on (say through manual adjustment), it will move out of focus to achieve its FF (what it thinks is perfect focus).
At this point, there is enough water under the bridge that I can't be exactly sure what point is being made in which post without going back and re-reading everything. However, the point I think was trying to make is that *some* focus errors are a matter of the phase detect system correctly figuring out how much the focus is off and calculating the number of turns to give the focus screw, but some sort of mechanical issue causing that to have the wrong effect. This *particular* problem would not cause the system to deliberately *defocus* an image that was already focused correctly - the phase detect system shold figure out there is no work to be done and leave well enough alone. Since this didn't happen, I can only assume it is due to a *different* cause.

Kind of hard for me to imagine how it could be lens related, though. If the phase detect system thinks it is OOF when it is actually in, that scream misaligned AF sensor to me. Maybe only an issue at certain (wide) apertures or focal lengths, which is why it might be *noticed* with only certain lenses. But I'm stumped as to how any service to the lens itself would cause the phase detect system to *not* try to turn the screw in cases where the image is actually in focus.

QuoteQuote:
I believe this will actually occur and is not a symptom of a different problem but would be evidence of the fact that the camera lens combination actively and willingly sets the incorrect focus because it determined it to be the supposedly perfect one.
Right - which is a *different* problem than I was previous describing. You are talking about the phase detect system being wrong about whether the image is in focus, and I agree, that's what seems to be the case here. But I had been talking about a case where the phase detect system was correct, but the turning of the focus screw did not have the desired effect, and the second check didn't fix the problem.
01-12-2009, 04:04 PM   #40
Pentaxian
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 8,934
QuoteOriginally posted by Marc Sabatella Quote
At this point, there is enough water under the bridge that I can't be exactly sure what point is being made in which post without going back and re-reading everything.
I think what we have referred to as a "problem" is FF/BF.
You offered an interesting explanation for it (lens not executing properly) but Gimbal and I challenged it with two arguments.

Gimbal's argument: The AF system is actually happy with the achieved misfocus and will actively restore it when disturbed by manual spot on focussing.

My argument: The AF system would have a regrettable high tolerance for the check after the lens has executed the specified number of turns.

I think your explanation is still valid but I think just for the cases where the AF did an acceptable job but varies a little within tolerances.

I think Gimbal's argument is a good one to suggest that the majority of noticeable BF/FF problems can not be explained by your hypothesis.

QuoteOriginally posted by Marc Sabatella Quote
If the phase detect system thinks it is OOF when it is actually in, that scream misaligned AF sensor to me.
Not necessarily, since many users report different lenses to require different AF adjustment values. So the body could be OK, but combination with a lens yields BF/FF respectively. As to why this can happen is one of the lens mysteries to me. Perhaps it can be explained by differing chromatic aberrations of lenses (which throw off the AF system).

QuoteOriginally posted by Marc Sabatella Quote
But I'm stumped as to how any service to the lens itself would cause the phase detect system to *not* try to turn the screw in cases where the image is actually in focus.
Same here. I don't understand how servicing of a lens can fix a body/lens combination FF/BF problem, unless the FF/BF is really caused by lens aberrations which can be fixed in the servicing.
01-12-2009, 07:31 PM   #41
Pentaxian
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,686
QuoteOriginally posted by Class A Quote
I think what we have referred to as a "problem" is FF/BF.
Oh, well, sure, that's the symptom of a general umbrella category of problems being discussed - but I was discussing one specific problem that might lead to BF/FF. There are, of course, other specific problems that could also lead to BF/FF. It's a mistake to assume all cases come from the same underlying causes. Cases that came from the cause I was talking about would *not* produce errors if the lens was prefocused, so if any lens is showing BF/FF even when prefocused - the camera actually moves the lens out of focus - then that *particular* case is not caused by the particular problem I was talking about.

QuoteQuote:
QuoteQuote:
If the phase detect system thinks it is OOF when it is actually in, that scream misaligned AF sensor to me.
Not necessarily, since many users report different lenses to require different AF adjustment values.
I don't see that as inconsistent at all. If you're using the AF adjust menu as a way of working around a misaligned sensor, then the specific amount by which the system needs to adjusts its own idea of what seems to be in focus could well vary by focal length and focusing aperture.
01-13-2009, 07:19 AM   #42
Pentaxian




Join Date: Apr 2007
Location: Sweden
Posts: 1,883
Could it be that a modern lens tells the camera that at a certain focal length there should be a certain phase error (for some unknown optical reason) in order to achieve focus? That would explain quite a lot.
01-13-2009, 01:03 PM   #43
Pentaxian
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 8,934
QuoteOriginally posted by Marc Sabatella Quote
If you're using the AF adjust menu as a way of working around a misaligned sensor, then the specific amount by which the system needs to adjusts its own idea of what seems to be in focus could well vary by focal length and focusing aperture.
I cannot exclude this possibility but it seems unlikely to me. If the error is with the camera then adjusting it at the camera should fix the error for all lenses. Why would the hardware fix correct the problem for all lenses and the software fix be dependent on the lens? Doesn't quite sense to me as I'm assuming the software lens is trying to compensate for the hardware problem. It should be possible to compensate for a misaligned AF sensor by not matching for a 0 degree phase but another phase deviation amount that corresponds to the misalignment. This should fix all lenses in the same way the hardware fix would have done. I could be wrong about this, though.


QuoteOriginally posted by Gimbal Quote
Could it be that a modern lens tells the camera that at a certain focal length there should be a certain phase error (for some unknown optical reason) in order to achieve focus? That would explain quite a lot.
That would indeed explain a lot. It begs the question as to why this should happen; why not simply let the camera adjust focus until it is OK? A potential explanation would be that lenses have chromatic aberrations that depend on the focal length. The AF system then thinks it got the right focus but because of the aberration it did not (regarding the image on the sensor). In this case the lens could tell the camera how to compensate for this effect. If this data doesn't match the lens characteristic then there will be FF/BF.

I have no idea though whether lens aberrations would be sufficiently high enough to either a) throw off the AF system or b) let the AF system do its work properly but cause the lens to project the image not quite on the sensor. This is just guesswork on my behalf and I wish I had access to some technical explanation of these phenomena.

Last edited by Class A; 01-13-2009 at 01:08 PM.
01-13-2009, 01:34 PM   #44
Pentaxian
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,686
QuoteOriginally posted by Class A Quote
I cannot exclude this possibility but it seems unlikely to me. If the error is with the camera then adjusting it at the camera should fix the error for all lenses. Why would the hardware fix correct the problem for all lenses and the software fix be dependent on the lens?
Let's say the error in phase caused by the misaligned sensor (and I'm being extremely non-technical here, because I don't pretend to understand the process) is 1mm. That is, the misalignment of the sensor is causing the phase to read as correct when it is actually off by 1mm, or and conversely, it causes the phase to be off by 1mm when it is actually focused correctly. Well, the fact that the phase difference on the lens is 1mm is one thing, but what does this translate into in terms of how far the actul focus of the lens is> 1mm of phase difference might be 1 foot of difference in focus on one lens, 2 feet on another, 3 inches on another, etc. This is something the camera would normally figure out (presumably) based on its communication with the lens, and would potentially depend on focal length, focusing distance, and focusing aperture. In a correctly algined system, all is well, and the calculations go smoothly. But if the sensor is misaligned, I could easily see it throwing off those calculation such that the "usual" way of calculation how many turns of the screw to give for a given focal length, focus distance and focusing aperture might be off. In fact, I'd be *shocked* if the math happened to work out such that a single global adjustment actually fixed all lenses for all focus distances.
01-13-2009, 02:15 PM   #45
Veteran Member
Ben_Edict's Avatar

Join Date: Jul 2007
Location: SouthWest "Regio"
Photos: Albums
Posts: 3,303
For once, I don't think the lenses provbide that much intelligence - of if they do, it does not seem to be the cause of the AF problems. Why? Because we also have really old F-lenses arround. These do sure not contain this detailed information (phase-correction), but focus just as any modern lens.

Measuring the phase or phase-shift with the AF-module is one thing (phase analyzers are off-the-shelf products), but including correction parameters into highly mechanical devices (like a lens) simply won't work. Especially in the cheaper zooms, the mechanical play between all the tubes is way too much to justify any inbuilt correction parameters. I really think, the AF processing is done by the AF module on its own - at least with Pentax.

Ben
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
da35, ff, focus, k-mount, lens, pentax, pentax lens, question, slr lens, spot
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Can a manual-focus lens have back focus problems? PocketPixels Pentax DSLR Discussion 33 02-13-2011 05:37 AM
Suggestion How about adding Focus Throw to lens details in the Lens database? brecklundin Site Suggestions and Help 2 12-08-2009 05:49 PM
Front focus/ back focus on Manual lens. Possible? WangJianWei Troubleshooting and Beginner Help 6 04-20-2009 07:50 PM
Colour varies between applications nobbsie Digital Processing, Software, and Printing 0 09-23-2008 04:05 AM



All times are GMT -7. The time now is 02:01 PM. | See also: NikonForums.com, part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top