walkaround: Wilba, great article that answered several questions I had. Thanks for posting it!
You're welcome. Things are not always as they seem (or as everyone says they are). :- )
ctlow: Having read many comments, maybe Marom's images have foreground and maybe they don't. (They do ... or the rest of the photos would be the entire photos. He's termed it "unforeground" but it could be "muted foreground" or some other term or your liking) But regardless, it's a great reminder to pay attention to the foreground when you you feel that your subject needs some underpinning. I see lots of amateur photos with ugly foregrounds. These simply need to be cropped out. If considering an "unforeground", as in these images, consider whether you're including it because it's just there or because it does actually add to the image appeal.Either way, Marom has demonstrated some stunning photographs, and explained how he conceptualized the processes, and that's been instructive to me.
He's not talking about images having no foreground, he's talking about no foreground _element_ - something that attracts the viewer to look at it, by being "usually... massive... particularly detailed and interesting and also separated from the other image components". In other words, nothing that stands out in the foreground. He often says "foreground" when he means "foreground element", so keep that in mind as you read the article.
Winner of mini-challenge 412 - Sunsets
mostlyboringphotog: @WilbaW (18 hours ago)
My most sincere apology and deep regret that I did not notice your post that already suggested the same.
At least, I hope you appreciate that a great mind is imitated by the lesser at later time.
I wish I had read the earlier posts before I jumped on my soap box first.
Again, my sincere apology.
No worries. Just an occasional "(c) Wilba, 2014" when you use terms of the form "DOF-equivalent aperture" will suffice. :-D
Richard Butler: It's been suggested to me that people would be happier if we used the term 'Equivalent F-number' rather than 'Equivalent Aperture.' Is this the case?
I don't think he meant you could only ever specify one aspect of equivalence at a time (I certainly didn't), just that if you're making a point about DOF (or diffraction, or noise...), you can explicitly emphasise it in that way. Similarly for multiple aspects. It's about naming the relevant aspects so people don't think it's about something else. But you're right, once you've done that a few times everyone should know that "equivalent" implies all aspects. :-)
I wonder if we just need to be more explicit, like "DOF-equivalent aperture" (similarly - FOV-equivalent FL, diffraction-equivalent f-ratio, noise-equivalent entrance pupil...), to make sure no-one thinks we might be talking about metering.
mostlyboringphotog: If you want better IQ (less noise) get large sensor camera and deal with bigger mostly everything. If you want to carry your camera in you pocket, get a small sensor camera and live with less IQ photo.Just to prove the rather obvious in the above, you folks comes up with "equivalence" as if these were not understood (this may be the my misreading) before you folks pointed out.So now we have "f2=f2=f2, no if and buts" vs. "f2=f2=f2 is BS". Mama mia!!It comes down to this at least for me:Which is less confusing? 100mm on FF and 50mm on mFT will provide equivalent FOV or 100mm FF and 50mm mFT are equivalent.Which is less confusing? F/4 on FF and F/2 on mFT will provide equivalent DOF or F/4 FF and F/2 mFT are equivalent.And please no Quantum Effects, there may some chance that the photo of your Aunt will turn out to be equivalent to a photo of Beyoncé but not likely.And please don't start out if someone does not like this article, explore CoC; the objection may just be the font.
@MBP "Which is less confusing? 100mm on FF and 50mm on mFT will provide equivalent FOV or 100mm FF and 50mm mFT are equivalent."
Depends on context. If someone is talking explicitly about FOV and says, "100mm FF and 50mm mFT are equivalent", there is no confusion, except perhaps for an inattentive reader. But if they say it out of the blue, we need to know in what way are they equivalent, so in isolation it should be said like, "100mm FF and 50mm mFT are equivalent for FOV". As long as the communication is clear and explicit, neither form is confusing in any way. In the context of a discussion of equivalence, there is no other implication to "100mm FF and 50mm mFT are equivalent" than that it's about FOV.
"Which is less confusing? F/4 on FF and F/2 on mFT will provide equivalent DOF or F/4 FF and F/2 mFT are equivalent."
I'm not sure what your point is, but I think it might become clear if you were to answer that question yourself.
WilbaW: I have a hunch that those who don't get equivalence also don't get DOF - both are all and only about the viewer's perception of an image, and in the case of equivalence, it's a comparison of same size images viewed from the same distance.
If that confronts you, I suggest exploring the derivation of COC and how that flows through to DOF calculation, then come back to the article and see if it makes a difference to your understanding of how equivalence delivers the same DOF from different formats.
What does 37 comments by you on just the current first page tell us about how much you care?
Ah, okay. I've never thought exposure would be a relevant metric in cross-format comparisons, so no challenge for me! :-)
Mm, there sure does seem to be a reaction as if equivalence is telling us which format is better, which one to to buy, or how to take a shot with it (which for the record, of course, it doesn't in any way).
I'm not sure we have to unlearn anything... f/2 = f/2 = f/2 _for_exposure_ (both technically and for image brightness). Then you just have to add f/2 = f/3.2 = f/4 _for_DOF_. That's no harder to understand than 50mm = 50mm = 50mm for the lens, and 50mm = 65mm = 100mm for framing. Since we already accept "equivalent focal length" without distress, I can't understand why "equivalent aperture" applies torsion to undergarments.
I wonder what those who reject "equivalence" as undeserving would accept in its place, or OTOH, what other visual properties would have to be included?
I don't disagree with anything you said, I just have a feeling reading through the comments that the guys who struggle are the ones who think COC is a fixed property of the format (and therefore of the sensor and the camera it's in), rather than a variable property of the viewing of an image.
Total light is a leap, for sure. Maybe it's as simple as whether you really do follow the physics?
I have a hunch that those who don't get equivalence also don't get DOF - both are all and only about the viewer's perception of an image, and in the case of equivalence, it's a comparison of same size images viewed from the same distance.
ProfHankD: Different aspect ratios don't really have straightforward crop factors, although using the diagonal is common. It gets even more confusing when one, for example, quotes a 2X crop factor for micro4/3 and then shoots 16:9 video on it... because the crop factor jumps up a bit more than it does from a 3:2 aspect. There are also perversions like Canon's slightly-small-APS-C (1.6X vs. 1.5X crop).
The problem with these notions of "equivalence" in general is that the equivalence only holds for one attribute at a time. For example, DoF equivalences don't imply equivalent exposure times or field of view. For that matter, DoF really is a function of allowable circle of confusion, which is a function of system resolution (combined effect of pixel count, CFA interpolation, etc.), NOT format or sensor size. In sum, there's really no substitute for actually understanding what each measure means.
Well, "DOF is... NOT [a function] of format or sensor size", says there is no mathematical term related to sensor size (or magnification) in the derivation of COC. But I suppose if you use "function" in some vague meaningless way in some unspecified context you might get away with it. :-)
I don't see how sensor resolution attributes could be the determining factor in "normal" viewing (from a distance of the order of the diagonal of the image) of typical images from contemporary cameras, so we're basically taking about two very different things. As long as you're clear about the context in which your statement applies, I don't have a problem with it, but the implication doesn't apply in the normal context under which DOF and COC were defined.
Not a specific print size and viewing distance but a simple relationship between the two (though we do tend to refer to the classic example).
Yes, CoC is a moving target, but it can be derived for a given level of acuity and an image size and viewing distance relationship. To say it has nothing to do with relative magnification (format, IOW), is obviously wrong.
From where are you taking your definition of allowable circle of confusion?
WilbaW: "A 100mm equivalent lens on a small-sensor camera will give the same framing and perspective as an actual 100mm lens does on a full-frame camera..."
I don't have time to check all 1155 preceding comments to see if this has been mentioned already, so apologies if it has... please stop perpetuating the misunderstanding that perspective is related to framing or angle of view. It only depends on the point of view (where the camera is in space).
I think we're using different definitions of "framing". A true unambiguous statement would be like, "a 100mm equivalent lens on a small-sensor camera will give the same framing and perspective as an actual 100mm lens does on a full-frame camera from the same position and orientation...". Same position and orientation needs to be explicit if that's the intended meaning. Without it, the original statement implies that perspective depends on equivalent focal length.
It's only true if same position is specified. For instance, "a 100mm equivalent lens on a small-sensor camera will give the same framing as an actual 100mm lens does on a full-frame camera, and from the same position will give the same perspective..." (In that case the perspective bit is self-evident and redundant, so it's better to just say, "a 100mm equivalent lens on a small-sensor camera will give the same framing as an actual 100mm lens does on a full-frame camera".)
The author failed to do that, so the statement is at best misleading, and perpetuates the misunderstanding that perspective depends on AOV.
"A 100mm equivalent lens on a small-sensor camera will give the same framing and perspective as an actual 100mm lens does on a full-frame camera..."
Subtitle: "Are we well camouflaged or what?"
alex.di: The 50/1.8 doesn't detect small changes in focus position. That's why it's 'crude.' You can shift position until you're on the trailing edge of the available DOF, or even just outside it, and nothing will happen if you attempt to refocus. Whether or not the physical mechanism is to blame, it's an important characteristic of the lens to know.
You don't use AI Servo with stationary subjects because it doesn't check focus like One Shot. It also defaults to release-priority. That aside, my 40D constantly adjusted some lenses in AIS even on a tripod.
You don't move the focus ring in Auto mode with non-FTM lenses partly because the gears weren't designed to be stressed backwards, but also because it generates power. It's unlikely the camera filters that input, however small.
Exposure Compensation in Manual is in the context of Auto-ISO. That's the issue; it has nothing to do with a semantic argument over what the words mean, nor interpretations that define Manual as 'a fixed exposure.'
I think you're confusing DOF with the span between the two points of focus that PD AF gives you with the 50/1.8. The DOF surrounds one or other of the two points. You can move the subject out of that DOF without moving it beyond the other point. It's only when you go outside the two points that you get a refocus.
If you can't SHOW that AI Servo doesn't check focus like One-shot, how can you KNOW that AI Servo doesn't check focus like One-shot?
"I've never even seen that argument arise..." - check the link in the article to a "2013 discussion of Exposure Compensation".
I'm not aware of a myth about why there isn't EC in TAv mode, so I haven't written about it here.