[Ratmagiclady]

Perhaps you're the guy with 20/5 vision that can tell from looking at an 8x10 from 6x6cm negative that's front-focused by .3 cm. If so, then obviously you're going to have higher level requirements. DOF for you will be much narrower than for me (with mere 20/15 vision). Lenses that are exceptional for nearly everyone will be merely adequate for you. And digital isn't going to be much fun, as you'll see the blurriness of digital 300 DPI prints. But you're not going to be happy with FF, either.

Hee, that can be kind of a curse, too. My youthful eyesight was pretty ridiculous, (More like 20/2 or just plain off the chart ) (And you bet I wanted to be shooting Zeiss, too: someone brought a Contax into my store and I was in love. ) Nothing had enough edge on it. I'm actually really trying to loosen up about that, now that I'm in glasses, tempting as it is to magnify stuff on screens and all. )

Even so, my idea of acceptable DOF is kind of narrower, anyway: it's one reason I don't mind so much if you often end up without being able to make it razor-thin. Since I do a lot of low-light work, my old fast lenses would more often be giving me *less* DOF than I'd have liked than it being useful. (Apart from having been a manual-focus holdout till a few years ago. Narrow DOF is very good for that, and the real reason I love the fast glass. )

[jstevewhite]

Hee, that can be kind of a curse, too. My youthful eyesight was pretty ridiculous, (More like 20/2 or just plain off the chart ) (And you bet I wanted to be shooting Zeiss, too: someone brought a Contax into my store and I was in love. ) Nothing had enough edge on it. I'm actually really trying to loosen up about that, now that I'm in glasses, tempting as it is to magnify stuff on screens and all. )

Even so, my idea of acceptable DOF is kind of narrower, anyway: it's one reason I don't mind so much if you often end up without being able to make it razor-thin. Since I do a lot of low-light work, my old fast lenses would more often be giving me *less* DOF than I'd have liked than it being useful. (Apart from having been a manual-focus holdout till a few years ago. Narrow DOF is very good for that, and the real reason I love the fast glass. )

I worked in a camera shop with a guy like that. I'm not typically given to envy, but I envied him his vision and his memory (both were scary sharp). We would tape a magazine page to the wall (we were in our early twenties LOL ) and then walk up on it and stop as soon as we could read it. There would be most of the store's staff, 2-3 feet away; I'd be standing there five or six feet away, and Rick would be halfway across the room.

My DOF requirements are... Mercurial. Sometimes I literally don't care. As you say, though, I can't recall the last time I said "I need less DOF", but I have many times stopped down for more. I tend to shoot for the best MTF of the lens.

But, being old school, I've been known to use my 10-20mm at hyperfocal @f11 as a high-quality, wide-angle point-and-shoot, just like I used to do with my Canon 24mm.

[Wheatfield]

There's people in this forum who seriously claims there's no reason for Hoya to begin FF production (true FF, not Front Focus due to absence of QC) since crappy cameras provides the same IQ.

I'm wondering what you define as a crappy camera?
Do you actually think a 12mp 135 format camera is going to automatically have better "IQ" than a 16mp APS-C camera simply because it has a bigger sensor?
I'm sure there may be some conditions where it does, but there will be more conditions where 30% more pixels will give the smaller sensor camera the edge (such as resolving fine detail, which IIRC, is a major part of IQ)
I've pretty much concluded that you are even more of a troll than I am.

[jstevewhite]

I'm wondering what you define as a crappy camera?
Do you actually think a 12mp 135 format camera is going to automatically have better "IQ" than a 16mp APS-C camera simply because it has a bigger sensor?
I'm sure there may be some conditions where it does, but there will be more conditions where 30% more pixels will give the smaller sensor camera the edge (such as resolving fine detail, which IIRC, is a major part of IQ)
I've pretty much concluded that you are even more of a troll than I am.

I was going to ask if I get to call his FF a "crappy camera" because I shot 6x6cm in film, or if he's decided arbitrarily that 135 is the lower limit for good and APS-c the upper limit for crappy. It's actually *that* sort of hyperbole that got me started on the FF vs APS-c tirade.

[Wheatfield]

I was going to ask if I get to call his FF a "crappy camera" because I shot 6x6cm in film, or if he's decided arbitrarily that 135 is the lower limit for good and APS-c the upper limit for crappy. It's actually *that* sort of hyperbole that got me started on the FF vs APS-c tirade.

Similarly because I also use a Tachihara....

[Emacs]

I'm wondering what you define as a crappy camera?
Do you actually think a 12mp 135 format camera is going to automatically have better "IQ" than a 16mp APS-C camera simply because it has a bigger sensor?
I'm sure there may be some conditions where it does, but there will be more conditions where 30% more pixels will give the smaller sensor camera the edge (such as resolving fine detail, which IIRC, is a major part of IQ)
I've pretty much concluded that you are even more of a troll than I am.

Yes, it is. The mps are don't matter significantly. The difference is how lenses should work. When compare lenses, you should compare similar FOV, and difference will become huge there: the 31Ltd can be very good, but it loses to any good 47mm lens on FF when working as 47mm one on APS-C. Just because it provides 1.5 times less details than those 47mm lenses

[jsherman999]

Absolutely so. But you're not comparing power tools and hand tools; you're comparing different brands of power tools, because you could certainly produce the same volume of images with APS-c that you can with FF, right?

Workflow throughput would be affected if you needed to run all images from a session through an additional NR step to match the FF cleanliness, or if you needed to do some multi-image mashups to approximate a more shallow DOF for the same FOV. More likely, though, the AF module is just going to bring a higher keeper ratio (AF is not format-specific but comes with the bodies.) So yes, as a tool, it's easier to use generally.

A guy shows up to build your kid a swingset, do you care if he's using a 3/4 horsepower DeWalt or a 5/8 horsepower Makita drill to drill the mounting holes? *HE* may care.

Bingo! Steve, if you want your metaphorical analogy to follow mine, then keep the protagonist the same - you are the worker here, not the guy having his swingset built. You are choosing the tools that make your job easier, quicker, more enjoyable.

He may say that the DeWalt fits his hand better. He may say that the 1/8 horsepower more he gets really makes his life easier. He might say that his Dewalt is BETTER than the Makita. Consumer reports may say that the Makita has 5% fewer failures, so the other guy might claim his Makita is better.

I'd say DeWalt vs. Makita is more K5 vs D7000, or 5DmII vs D700 than aps-c vs FF.

The workmen care about their tools because they live with them.

Bingo++.

Metaphors are fun.

They are. But you have to make sure you don't get so excited you lose the gist of the preceding metaphor you're trying to refute!

Fair enough. But if isolation is my goal, I can step back and shoot a longer lens. The question of "better" presumes that one would shoot the same image with both cameras. Do you carry 'em both and shoot identical images with both of 'em?

Maybe I should talk about my typical shooting situations. Most of my shots are kid/family shots, I'm basically documenting their lives as much as practical, for us and extended family. Before FF I lived with a 35mm prime on all the time, I just found the FL so useful for that. Outside it was the K20D + DA 35ltd, inside usually the D90 + 35 1.8G. I also liked the 100mm & 135mm fl on aps-c a lot, for outdoor kid shots and just general telephoto stuff.

With the D700, I swapped the 35 for a $109 50 1.8, and fell in love with that FL on FF, and that lens performed very well on that camera. I noticed that right away I could get the f/1.8-like DOF at my typical subject distances (which I liked) at f/2.8 on the 50 + D700 - but f/2.8 on the 50 was sharper than f/1.8 on the 35 (and had less CA.) It was win-win, except the D700 combo was a bit bigger, so really, WIN-WIN-lose.


(50mm f/1.8, 35mm f/1.2 aps-c equiv, and 50mm f/2.2 ISO 2200, 35mm f/1.5 equiv)



The 180 2.8 (at $500, one of my most expensive Nikon lenses!) arrived in the mail, and it became my 'new' 100 and 135, with about a 120mm FL equivalent. But it was a 120mm f/1.8 with regards to DOF - not something I always took advantage of, but it was there. But again, the D700 + 180 2.8 was much larger than the K20D + M 100 2.8 or 135 3.5.


(both 180mm f/2.8, 120mm f/1.8 aps-c equiv)



So... for me, the whole DOF thing did work out to be nice. Not something I'd buy into FF for alone, but something that has been fun to have in my typical shooting situations. I fully understand if it doesn't ring everyone's bell.

(and nothing has yet defeated the 77ltd, the DA 35ltd is still the best 35 (or FF 50 ) I've ever shot, and I'm using the 15ltd more than anything on either mount, now. So Pentax aps-c is staying in my bag as long as the company stays afloat. )

Well, I know we've "been over this", but when I looked up the math, it came out like i expected in regard to hyperfocal length. So stopping down the longer lens simply does *not* give you the same DOF as APS-c, even with reduced COC. I'd be happy to consider factual rebuttals; I just googled the math and punched in numbers. So according to that math and my current understanding of it, it's not "more control of DOF", it's "less DOF".

I saw the example you gave to Marc, a 35mm lens at f/16 had an encroaching hyperfocal line... ?? I'm not sure how that applies to my example or to most shooting situations where you're shooting at subject-isolation apertures.

My 'stop down to match' example equates to:

35mm f/2.8 on aps-c = hyperfocal line at 21.8 meters
50mm f/4.5 on FF = hyperfocal line at 18.5 meters

Subject distance was about 3 meters. That's a long way from hyperfocal in both scenarios. Bob Atkins mentions that each bullet point applies as long as your subject is a significant distance from hyperfocal. Can you tell me how your hyperfocal argument applies to anything here?



.

---

Last edited by jsherman999; 05-17-2011 at 11:14 PM.

[Marc Sabatella]

Hyperfocal distance is independent of format and dependent on focal length and aperture. Thus, the hyperfocal distance for, say, a 30mm lens at f16 is found by:

H is approximately (30)^2/(16)(.03) or 1875mm from the lens.

OK, then it's *not* the same basic calculation as DOF.

I'm certainly willing to read the material you'll provide that demonstrates that math to be inaccurate.

The readings I refer to were about ordinary DOF calculations, not hyperfocal distance. Sounds like there really is a difference here.

[jsherman999]

Hyperfocal distance is independent of format and dependent on focal length and aperture. .

Except it's not independent of format - at least not according to Bob Atkin's DOF/hyperfocal calculator here.


.

[Marc Sabatella]

if isolation is my goal, I can step back and shoot a longer lens.

Well, sure, but then you have a different shot. And of course that's also an option on FF. The point being that FF gives you *more* options.

when I looked up the math, it came out like i expected in regard to hyperfocal length. So stopping down the longer lens simply does *not* give you the same DOF as APS-c, even with reduced COC. I'd be happy to consider factual rebuttals; I just googled the math and punched in numbers. So according to that math and my current understanding of it, it's not "more control of DOF", it's "less DOF".

No, it's less DOF in the very rare special case of someone trying to stop down as far as possible and then shooting at hyperfocal distance (assuming your math is indeed correct). In all other cases, you have the option of the same DOF, but also the option of less DOF.

[Marc Sabatella]

Except it's not independent of format - at least not according to Bob Atkin's DOF/hyperfocal calculator here.

Indeed. I just plugged in the numbers, and sure enough, hyperfocal distance for a 35mm lens at f/16 on APS-C is exactly the same as for a 50mm lens at f/22 on FF, just as one might expect. Whatever this special magic "different behavior as you approach hyperfocal distance" thing might be that allegedly makes the hyperforcal distance closer for APS-C than FF, it doesn't seem to apply here. The ordinary concept of equivalence seems to apply just as would one expect, according to the calculator on Atkins' site, which most definitely *does* include format as one of its input parameters.

So I'm back to being completely unconvinced there is anything whatsoever to that claim.

[TOUGEFC]

The workflow in Photoshop takes too much time; I've started using AutoPano Giga. There's a trial - check it out!

Generally, I import the images in Lightroom. I select the set I want, and export them to 16 bit TIFFs (so as not to lose the RAW dynamic range!), then open the folder in APG; click "detect", and somewhere between a few seconds (~5 images) and a minute (~42 images) a window opens that has them all aligned and masked. You can zoom in, noodle alignment, all kinds of stuff (most of which I don't have to do, and you won't either, if you're careful shooting). Then you click on the gear button and export to 16 bit TIFF.

Then I import THAT image into Photoshop or Lightroom and do post processing.

I did a landscape pano recently as a test with three-shot bracket; I am trying to discover the workflow for HDR->Pano (or Pano->HDR)... Not crazy tone mapping; just to bring out the sky better.

Shooting the stitching set... use manual focus and manual exposure and manual ISO. Make sure there's good overlap. I've ruined whole stitches before by missing a picture in the middle

+1 rep, thanks mate i will make a note of it and check it out

[jstevewhite]

Yes, it is. The mps are don't matter significantly. The difference is how lenses should work. When compare lenses, you should compare similar FOV, and difference will become huge there: the 31Ltd can be very good, but it loses to any good 47mm lens on FF when working as 47mm one on APS-C. Just because it provides 1.5 times less details than those 47mm lenses

You're taking the general, ideal case and projecting it onto real-world situations that don't approach the ideal case.

The wider the image circle relative to the focal length, the harder (and more expensive it is) to design sharp lenses; for instance, Nikon made a lens for a chip fab that resolved 1200 lp/mm; but it was in the 2mm paraxial region. So if the same targets were applied to APS-c that are applied to FF, we could have much sharper and faster lenses than we do. The FA 31 1.8, for instance, could probably be a 31mm 1.4 or 1.2 for the same - or less - money on APS-c, with higher resolution.

It's true that a lens can be much poorer and out-resolve a FF sensor (beyond which point additional sharpness becomes moot). But we're not talking about poor lenses. Even on the K-5, my 35mm F2 at least matches the sensor resolution. So the K5 will record more detail - and more information - than a 12MP FF.

Of course, if the FF camera has the same pixel density as the APS-c, this is no longer true, but it gives up some of the advantages of the large sensor sites (signal to noise, for instance).

[jstevewhite]

Workflow throughput would be affected if you needed to run all images from a session through an additional NR step to match the FF cleanliness, or if you needed to do some multi-image mashups to approximate a more shallow DOF for the same FOV. More likely, though, the AF module is just going to bring a higher keeper ratio (AF is not format-specific but comes with the bodies.) So yes, as a tool, it's easier to use generally.

Yeah, but I don't have to do any NR processing. (Beautiful images, by the way. I do like your work!)

Bingo! Steve, if you want your metaphorical analogy to follow mine, then keep the protagonist the same - you are the worker here, not the guy having his swingset built. You are choosing the tools that make your job easier, quicker, more enjoyable.

Heeh... We don't and have never disagreed here; if you LIKE it better, it's worth it. No argument. As I've also said, I LOVE the viewfinder and miss it horribly. If my eyes get much worse as I get older, I might switch on that basis alone... LOL

Maybe I should talk about my typical shooting situations. Most of my shots are kid/family shots, I'm basically documenting their lives as much as practical, for us and extended family. Before FF I lived with a 35mm prime on all the time, I just found the FL so useful for that. Outside it was the K20D + DA 35ltd, inside usually the D90 + 35 1.8G. I also liked the 100mm & 135mm fl on aps-c a lot, for outdoor kid shots and just general telephoto stuff.

With the D700, I swapped the 35 for a $109 50 1.8, and fell in love with that FL on FF, and that lens performed very well on that camera. I noticed that right away I could get the f/1.8-like DOF at my typical subject distances (which I liked) at f/2.8 on the 50 + D700 - but f/2.8 on the 50 was sharper than f/1.8 on the 35 (and had less CA.) It was win-win, except the D700 combo was a bit bigger, so really, WIN-WIN-lose.

So FF matches your vision. I completely support the idea that the tools that make it easier for you to achieve your vision are the right ones.

I saw the example you gave to Marc, a 35mm lens at f/16 had an encroaching hyperfocal line... ?? I'm not sure how that applies to my example or to most shooting situations where you're shooting at subject-isolation apertures.

Not specifically to your arguments. Hyperfocal distance - back in the day - was something every photographer learned to use; it was our "autofocus". A 24mm @f8 in the sun was in focus from ~1m to inf, if memory serves. I still use it with the 10-20 sometimes for pickups and candids.

[jstevewhite]

Except it's not independent of format - at least not according to Bob Atkin's DOF/hyperfocal calculator here.


.

I see that his DOF calculator does indeed suggest that a 50mm FF has the same hyperfocal distance as a 35mm on 1.5 where f=f+1. Which I find a bit suspect, as the number he throws around (and most people arguing for FF superiority ) is 1.3 stops, not 1 stop. Furthermore, I looked up the actual math, not a canned calculator, and put in the numbers myself, and got a different result. Feel free to check my math, or present a source for an equation that gives the results Bob's calculator does with a reason to believe it's authoritative.

The formula I have for hyperfocal is H=(f^2)/Nc, where f=focal length, N=f-number, and c= acceptable COC (which varies with enlargement).