Is CDAF really derived from in-cam JPG adjustments?

Started May 27, 2012 | Discussions
Detail Man
Forum ProPosts: 14,950
Like?
Is CDAF really derived from in-cam JPG adjustments?
May 27, 2012

This is not to doubt or question those who are finding that an Olympus "Vivid" in-camera JPG mode seems to improve C-AF results (or even S-AF results). That's fine, and I do mean to "rain on your parades" (so to speak). I presume that such assumtions are based upon adequate experimental "controls" (that is, trying the very same thing when not in the "Vivid" mode, etc.) ...

This is about the general implications regarding the Auto Focus "system architecture" of various brands and camera models - because it clearly has important implications for the performance characteristics of CD Auto Focus systems in general (as it affects not only JPG, but also RAW, recording). The typical speed of CDAF with many M43 systems makes these things hard to determine via time measurements - but the issue of auto-focus integrity remains paramount indeed

It seems reasonable to assume that some (possibly abbreviated for speed) form of de-mosaicing of the photo-site data occurs prior to evaluation by a CDAF system implemented in a camera system.

We know that most/all systems widen the aperture to it's maximum diameter when auto-focusing (and at least in some cases when manual focusing). This makes good sense. Maximizing the absolute value of the (de-mosaiced) pixel-data (assuming that photo-site saturation does not occur as a result, that is) should result in better performance of the image-data utilized in CDAF systems.

It makes sense that in-camera JPG adjustments that directly and most significantly affect the amplitude of the image-data anywhere in the signal pathway preceding the CDAF system would/should be taken into account (so as to avoid non-linear "clipping" at any point). I'll buy that.

But, what about some of the other in-camera parameters that are user-adjustable ? " Contrast " ? In my Lumix FZ30, FZ50, and GH2, the "Contrast" adjustment appears to modify the scaling of the tone-curve below the maximum point of the tone-curve transfer-function - but does not alter the set-point of the maximum point itself (so as not cause large changes in the overal "brightness"). Nevertheless, it's hard to understand why camera designers would want "Contrast" adjustments to significantly affect the CDAF system (causing variations in CDAF performance based upon settings).

" Saturation " ? It seems unclear whether the CDAF should (or does) care about relative intensity of the color data. CDAF would rightly seem to concern itself with composite luminance - and allowing color saturation adjustments to alter CDAF functionality would (again) only seem to cause unwanted variations in CDAF performance based upon settings. Again, this does not sound wise ...

" Sharpness " ? Here again we have a situation (probably the most dramatic case of all discussed here) where the operational integrity of the CDAF system would become hostage to user-settings.

It makes zero sense to me that camera designers would prefer a situation where users become forced to use specific "modes" (such as Pana "Film Modes", or the Oly equivalent seen in modes such as "Vivid") in order to try to coax better (or even adequate ) CDAF system performance out of their cameras. Most users would miss such subtleties altogether - and simply conclude that a camera model has "lousy AF". A tiny, tiny fraction of users who lurk and post in fora such as this would endlessly trade in "groovy" in-camera JPG user-settings that might allow them to actually achieve varying ( possibly better) levels of CDAF system performance. A subjective "total mess" ...

Nothing have to do with common sense (engineering, or otherwise) would imply that such a cloudy and variable situation where it comes to CDAF system performance would be a desirable or a prudent one where it comes to camera design and utilization. RAW shooters (who thought that they had blissfully avoided all the in-camera JPG settings "stuff" and worries) would be forced to join the fray of confused and non-scientific speculation as well - simply to try to ensure that their cameras' CDAF systems would/will function in an effective and acceptable manner . It is (sort of, in some cases) well understood that in-camera JPG settings affect previews/histograms/blinkies, and any other matter surrounding camera metering, (and the RAW-level scaling) - hence, "UniWB" ...

But, for camera designers to extend such a bona-fide "total mess" (surrounding in-camera JPG settings) beyond (camera-setting determined) "exposure values", and into matters of the operational characteristics/integrity of CDAF sounds like a prescription for trouble/disappointment ...

A few posters have in the past boldy claimed that because "preview" modes have to operate quickly , then it would follow that it would seem to be unlikely that the image-data that CDAF systems operate upon could itself be anything other than what the user (and histograms/blinkies) "see" in LCD/EVF displays. Yet, we are talking about proprietary system architectures that are very closely held by camera manufacturers (of which we can only merely speculate, and try to somehow evaluate by some sort of observation of how the systems as a whole operate). In one case, a poster who merely speculates that the above-described situation is likely the case also commonly asserts that the chip-sets in camera hardware-processors are very complex, capable, able to simultaneously perform a number of operational tasks simultaneously . Thus, much remains unclear.

So, a number of things about these matters simply "do not add up" - on an engineering design level, or on a practical user level. "Experimental observations" conducted which do not implement adequate "controls" (images also recorded without such "special settings") seemed doomed to the realms of potential (and signifcant) subjectivity . Can "techno-wizards in the know" enlighten us ???

DM ...

Panasonic Lumix DMC-FZ50 Panasonic Lumix DMC-GH2
If you believe there are incorrect tags, please send us this post using our feedback form.
markintosh13
Senior MemberPosts: 1,413
Like?
Re: Is CDAF really derived from in-cam JPG adjustments ?
In reply to Detail Man, May 27, 2012

Why not run some tests that you would consider adequately detailed and controlled and report back?

Asking the great unwashed will likely not produce the answer to your stated question.

Though they may produce the results you desire.

Reply   Reply with quote   Complain
bgalb
Senior MemberPosts: 1,244Gear list
Like?
Re: Is CDAF really derived from in-cam JPG adjustments ?
In reply to markintosh13, May 27, 2012

Yep! Another 150 post thread.

 bgalb's gear list:bgalb's gear list
Olympus OM-D E-M5 Olympus E-M1 Olympus Zuiko Digital ED 12-60mm 1:2.8-4.0 SWD Olympus Zuiko Digital ED 50-200mm 1:2.8-3.5 SWD Olympus Zuiko Digital 1.4x Teleconverter EC-14 +5 more
Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,950
Like?
Re: Is CDAF really derived from in-cam JPG adjustments ?
In reply to markintosh13, May 27, 2012

markintosh13 wrote:

Why not run some tests that you would consider adequately detailed and controlled and report back?

One good reason that one might elect to do so is that such operational characteristics are likely exceeding hard to actually perform valid tests regarding those characteristics. It is fine if others may draw their own personal conclusions from this or that - but (unless the methodology is very involved, and includes adequate controls), I would not endeavor to make such assertions one way or the other. One thing that seems relatively clear (I would think to most) is that a CDAF system-architecture that is held hostage to specifc in-camera JPG settings would be a less than ideal thing.

If I were to present "findings" lacking in solid basis, such efforts would (rightly) be questionable ...

Asking the great unwashed will likely not produce the answer to your stated question.

I don't know how the readership will appreciate (or, perhaps, take some personal offence) to your assessment above. There are, in fact, some fairly intelligent and knowledgeable people who read these threads/posts. I am assuming that you might tend to count yourself within such ranks ? No ?

Though they may produce the results you desire.

You seem to have aquired the fora syndrome/disease of "criticize and cast dispersions first, maybe think about an actual technical issue later". Best wishes for a full recovery from such snarkiness.

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,950
Like?
Re: Is CDAF really derived from in-cam JPG adjustments ?
In reply to bgalb, May 27, 2012

bgalb wrote:

Yep! Another 150 post thread.

Well, if a raised question is in itself an interesting as well as largely undiscovered area of common knowledge, it might seem that others may take a keen interest in such matters - and possibly "throw in their own two-bits" ? I (myself) would rather ponder such subjects than the likes of:

"What is the cutest fur-lined case for my little woobie ?";

"What do other clueless blowhards and idiots like me divine that huge multi-national corporations will decide to do six months from today's date ?";

"Is the new object of material worship simply "perfect", or could it be that it is "super-perfect ?";

"There seem to be a lot of a-holes around here, how shall we manage to, by lynching them en masse, manage to eradicate all evil and mental incompetence from planet Earth once and for all ?";

... Or the classic fool's errand from word one:

"There is only one proper way for others to themselves subjectively perceive the "proper" amount of DOF/exposure/contrast/color-saturation/sharpness - and that would, (of course), be my way !"

Useless exercises such as the above hit the 150 post limit every day. In fact, the volume of responses seems roughly inversely proportional to the actual rational merit of the original post ...

Thus, since the original post of this thread seems (to me) to inquire about an important as well as quite interesting subject, it may indeed be much more likely to "sink like a stone" beneath wisdom.

DM ...

Reply   Reply with quote   Complain
Timur Born
Senior MemberPosts: 3,753
Like?
Re: Is CDAF really derived from in-cam JPG adjustments?
In reply to Detail Man, May 27, 2012

Let me hint you into some direction. Try aiming at the green leaves of a tree while half-pressing shutter in AF-C. Aim around the camera and hold it still repeatedly while watching the screen closely.

  • Gradations and Frame Rate (normal vs. fast) has an impact on AF-S performance (including initial focus acquisition with AF-C).

  • Whenever the camera has to perform a distinct focus acquisition action it changes contrast and saturation to extreme settings not available manually, brightness/gradations also changes.

  • When the camera is held still in AF-C with shutter half pressed you get different contrast and saturation compared to aiming around. One of these conditions may be affected by manual saturation, contrast, sharpness and color mode settings, or it may be not. Personally I could not reproduce any big difference yet.

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,950
Like?
Re: Is CDAF really derived from in-cam JPG adjustments?
In reply to Timur Born, May 27, 2012

Timur ,

Thank you for providing some specific (and what do indeed seem like useful) observations surrounding the (brand/model unstated, but assumed to be) Olympus OM-D E-M5 camera model

Timur Born wrote:

Let me hint you into some direction. Try aiming at the green leaves of a tree while half-pressing shutter in AF-C. Aim around the camera and hold it still repeatedly while watching the screen closely.

  • Gradations and Frame Rate (normal vs. fast) has an impact on AF-S performance (including initial focus acquisition with AF-C).

"Gradations" relating to the shape of the in-camera JPG RGB tone-curve transfer-functions. "Frame Rate" relating to the speed of image-recording (as well as to the refresh-rate of user-previewing ?)

  • Whenever the camera has to perform a distinct focus acquisition action it changes contrast and saturation to extreme settings not available manually, brightness/gradations also changes.

Hmm. Very interesting. That would make sense - and argues against more simplistic "explanations".

  • When the camera is held still in AF-C with shutter half pressed you get different contrast and saturation compared to aiming around. One of these conditions may be affected by manual saturation, contrast, sharpness and color mode settings, or it may be not. Personally I could not reproduce any big difference yet.

Hmm. Very interesting. That would make sense - and argues against more simplistic "explanations".

Thank you for your input !

DM ...

Reply   Reply with quote   Complain
Timur Born
Senior MemberPosts: 3,753
Like?
Re: Is CDAF really derived from in-cam JPG adjustments?
In reply to Detail Man, May 27, 2012

Yes and yes to your questions.

Impact of both depend on what you are aiming at and on light conditions. Especially Frame Rate leads to considerably faster hunting (double focus motor movement speed I'd say) in some lower light conditions. But only to a point where its lower screen exposure may not focus in darkest shadows at all anymore while Normal frame rate still focuses slow but reliably.

The latter was mentions in a short sentence in Richard's and mine E-M5 article.

Reply   Reply with quote   Complain
Timur Born
Senior MemberPosts: 3,753
Like?
Re: Is CDAF really derived from in-cam JPG adjustments?
In reply to Timur Born, May 27, 2012

Just to mention:

  • Each color mode saves/uses its own settings for gradations, contrast, saturations and sharpness. So some people might not notice that switching to Vivid may also change one of those settings, too.

  • Fast frame rate is not available in that "i" mode (forgot the name and have no cam to check atm) or any Art filter modes. It will automatically switch to normal in these modes and switch back to fast when leaving them.

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,950
Like?
Correction to OP: Is CDAF really derived from in-cam JPG adjustments?
In reply to Detail Man, May 27, 2012

Detail Man wrote:

This is not to doubt or question those who are finding that an Olympus "Vivid" in-camera JPG mode seems to improve C-AF results (or even S-AF results). That's fine, and I do mean to "rain on your parades" (so to speak). I presume that such assumtions are based upon adequate experimental "controls" (that is, trying the very same thing when not in the "Vivid" mode, etc.) ...

WHICH WAS ORIGINALLY INTENDED TO READ :

This is not to doubt or question those who are finding that an Olympus "Vivid" in-camera JPG mode seems to improve C-AF results (or even S-AF results). That's fine, and I do not mean to "rain on your parades" (so to speak). I presume that such assumtions are based upon adequate experimental "controls" (that is, trying the very same thing when not in the "Vivid" mode, etc.) ...

My sincere apologies to any readers who may have taken "umbrage" due to my unfortunate typo .

DM ...

Reply   Reply with quote   Complain
markintosh13
Senior MemberPosts: 1,413
Like?
Re: Is CDAF really derived from in-cam JPG adjustments ?
In reply to Detail Man, May 27, 2012

If your stated hypothesis is correct, perhaps it is a limitation of the affected cameras inability to process images separately for the purposes of focus contrast edge detection and image capture and a desire of the designers quest to minimize shutter lag / improve AF speed. I would have thought that a high pass edge detection approach would be better than relying on the actual high res image, but I am not a designer of CDAF AF systems.

That said, do we really have anyone here who would be intimate enough with the design of m4/3 AF design to credibly comment?

Have we ever had a forum member claim to be?

I would suggest that unless they were a current and active member of Panasonic or Olympus's camera engineering teams, anyone else's contribution would be mere speculation.

This is how myths and half-truths are created.

I think that given your very precise nature, and unwillingness to accept individual anecdotal qualitative findings, you still remain the best person to investigate this particular topic as it applies to you and your specific equipment, and that a careful investigation might shed some light on this currently unproven and somewhat believable / somewhat incredible conjecture.

I do not doubt your ability to formulate and document the problem statement, the methodology, the data collection technique and method, collect and analyze and present the results in a way that would make it understandable and replicable by others with different equipment. Thus, additional data points could be collected and contribute a greater understanding to this topic.

Reply   Reply with quote   Complain
markintosh13
Senior MemberPosts: 1,413
Like?
Re: Is CDAF really derived from in-cam JPG adjustments ?
In reply to markintosh13, May 27, 2012

Oh wait.

So this isn't CDAF in general, it's CDAF as it pertains specifically to one camera?

I thought this phenomenon was thought to apply to several cameras.

Reply   Reply with quote   Complain
markintosh13
Senior MemberPosts: 1,413
Like?
Re: Is CDAF really derived from in-cam JPG adjustments ?
In reply to markintosh13, May 28, 2012

What's the Panasonic / GH2 equivalent to Vivid?

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,950
Like?
Re: Is CDAF really derived from in-cam JPG adjustments ?
In reply to markintosh13, May 28, 2012

markintosh13 wrote:

Oh wait.

So this isn't CDAF in general, it's CDAF as it pertains specifically to one camera?

I thought this phenomenon was thought to apply to several cameras.

Such phenomena may or may not . Any other assumptions exist only in your profoundly gifted mind, oh "snarkitola" of all that is righteous, good, and evidently worth your "pearls of wisdom".

What's the Panasonic / GH2 equivalent to Vivid?

Why not ... RTFM, Einstein ?

Reply   Reply with quote   Complain
markintosh13
Senior MemberPosts: 1,413
Like?
Re: Is CDAF really derived from in-cam JPG adjustments ?
In reply to Detail Man, May 28, 2012

Sorry if the question seemed impolite.

I'd like to think we could converse without the swearing though, even in acronym form though.

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,950
Like?
Re: Is CDAF really derived from in-cam JPG adjustments ?
In reply to markintosh13, May 28, 2012

markintosh13 wrote:

Sorry if the question seemed impolite.

I'd like to think we could converse without the swearing though, even in acronym form though.

Oh, OK. I get the feeling that you (so far) have held little interest in the actual subject-matter of this thread. You seem to be trying to merely, snark, dig at, and (attempt to) bait me. Perhaps not "obscene" in the literal sense - but (so far) not very substantive or informing, either. What I think that I'll do is just not respond to your posts on this thread unless they offer something substantive

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,950
Like?
Re: Is CDAF really derived from in-cam JPG adjustments ?
In reply to markintosh13, May 28, 2012

Sorry, I do seem to have missed this post of yours (which seems to have some things to say).

markintosh13 wrote:

If your stated hypothesis is correct, perhaps it is a limitation of the affected cameras inability to process images separately for the purposes of focus contrast edge detection and image capture and a desire of the designers quest to minimize shutter lag / improve AF speed.

I guess that you are trying to say that: finding themselves unable to acheive the highest quality results from (wisely) maintaining a high degree of separation between the operational perfomance of CDAF systems and in-camera JPG user-settings, that (at least some) camera designers have elected to explore (as well as to perhaps implement), "avenues" which intermingle these (ideally maximally separated) functionalities in their otherwise non-productive quest for CDAF performance?

I would have thought that a high pass edge detection approach would be better than relying on the actual high res image, but I am not a designer of CDAF AF systems.

In fact, and measurement of "contrast" seems to (implicitly) amount to taking some sort of "derivative" (mathematical difference calculation). This, in itself, removes any numerical constant values that may exist in the image-data to be analyzed - and (in essence) amounts to what is termed as a spatial-frequency "high-pass" ("edge-detection") filtering. This would be true (of CDAF) regardless of the origin or nature of the image-data that is analyzed by the CDAF system.

That said, do we really have anyone here who would be intimate enough with the design of m4/3 AF design to credibly comment?

As you may have noticed, that is what the original post starting this thread attempts to ask (by asking the questions themselves). If it is true that such questions cannot be answered as a result of the discovery of existing system-architecture knowledge that may possibly be know and here disclosed, then we are left with the sobering possibility of having only anectdotal accounts from (likely, understandably so) non-rigorous experimentations on the part of individuals as information.

The assessment of the integrity of focus seems no less ridden with numerous variables (vastly complicated if in-camera JPG settings are indeed included) that are assessments of the efficacy of image-stabilization systems. Some very similar unknowns (in both subject-areas) exist. How deep was the DOF ? What does "acceptably sharp" happen to mean to "Tom, Dick, and Harry" - and how are we to somehow enter their pre-optic areas in order to actually have knowledge of such things)? What viewing medium , viewing size, viewing distance, and visual acuity accompanied such observations, etc. ? Assessing "focus" seems only slightly less complex than camera-stability, IS, etc

Have we ever had a forum member claim to be?

Yes, but (I) have found such persons to range into the dubious, "sweeping hothead" categories.

I would suggest that unless they were a current and active member of Panasonic or Olympus's camera engineering teams, anyone else's contribution would be mere speculation.

Well, it is rather hard to find much of any first-hand knowledge-claims on this forum that are not .

This is how myths and half-truths are created.

The whole idea of this thread to is gather what reliable information may exist to be known - and, in doing so, place other knowledge claims within the perspective that they are worthy of. The idea is not to invalidate the personal nature of such averments - simply to point out their limitations ...

I think that given your very precise nature, and unwillingness to accept individual anecdotal qualitative findings, you still remain the best person to investigate this particular topic as it applies to you and your specific equipment, and that a careful investigation might shed some light on this currently unproven and somewhat believable / somewhat incredible conjecture.

Not sure what you are referring to as "somewhat believable / somewhat incredible conjecture". You seem to have in one phrase contradicted yourself. Whatever you are referring to could be one or the other - but not both . If in the interest of "diplomacy" coherence is lost, what is gained?

I do not doubt your ability to formulate and document the problem statement, the methodology, the data collection technique and method, collect and analyze and present the results in a way that would make it understandable and replicable by others with different equipment. Thus, additional data points could be collected and contribute a greater understanding to this topic.

Well, I don't think that it might not sometimes to be a reasonable proposition to ask a question of others (in the hopes of any resident knowledge, or even some reasoned speculation coming forth) as opposed to personally endeavoring upon a major research project. If questions and rational speculations regarding possible answers were "not allowed", then all persons would silently perform vast amounts of personal research to formulate certain knowledge. Then, if/when others might ask if they would share their knowledge, should the response be, "is not that something which you, yourself surely must be ready, willing, and able to discover yourself ?". I appreciate your recognition of my "bold and foxy" standards - but question why such things might perhaps not be substantive subject-matter to discuss. Why would the cogence of such conversations be in doubt ? If the reason might be that few (if any) who lurk/post here are in a position to make concrete knowledge claims regarding these matters, then (it seems to me) that in itself "says something"

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,950
Like?
Re: Is CDAF really derived from in-cam JPG adjustments ?
In reply to bgalb, May 28, 2012

bgalb wrote:

Yep! Another 150 post thread.

As I've predicted, the original post would probably need to be more brief, bellicose, bone-headed in nature to resonate in the primal lizard-brain with the common frothing "sham rage" responses

Reply   Reply with quote   Complain
Lights
Veteran MemberPosts: 3,270Gear list
Like?
Re: Correction to OP: Is CDAF really derived from in-cam JPG adjustments?
In reply to Detail Man, May 28, 2012

I'm glad you brought this up. I couldn't figure out why, if this is indeed true about the faster focusing method (which I don't know if it is or isn't) why wouldn't the camera always focus in this mode, but process the output data in the selected mode? (jpg)

Which it does anyway, if considering Raw files. It would seem all the varying modes would do is leave small markers (which they do - since even in Raw, Olympus viewer is aware of the shot modes. You can ignore them and change to another mode in software)
--

http://www.pbase.com/madlights
http://barriolson.aminus3.com/

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,950
Like?
Re: Correction to OP: Is CDAF really derived from in-cam JPG adjustments?
In reply to Lights, May 28, 2012

Lights wrote:

I'm glad you brought this up. I couldn't figure out why, if this is indeed true about the faster focusing method (which I don't know if it is or isn't) why wouldn't the camera always focus in this mode, but process the output data in the selected mode? (jpg)

You make the very good point (which has occured to me, as well) - that is, if it is so "robust and groovy" when in Vivid mode, why would they ever purposely compromise such "perfection" ???

If you read Timur Born 's post on this thread, it appears from what he reports that what the E-M5 is actually doing whenever CDAF is being performed seems to be rather mysterious, indeed ...

Which it does anyway, if considering Raw files. It would seem all the varying modes would do is leave small markers (which they do - since even in Raw, Olympus viewer is aware of the shot modes. You can ignore them and change to another mode in software)

Hmmm. Olympus is very likely smart enough to not reveal their "trade secrets" via the meta-data informational "tags". That would be "too easy" to crack. What goes on is probably not in there ...

Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads