August 20, 2014, 03:33:05 AM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - jrista

Pages: 1 ... 96 97 [98] 99 100 ... 263
1456
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 11, 2014, 05:19:52 PM »
Well, certainly. Literal ISO is always different. It isn't even consistent within the same brand, though. For example, according to DXO, "measured" ISO 100 on the 5D III is 80, measured ISO 200 is 160, where as "measured" ISO 100 on the 70D is 93 and ISO 200 is 160. The 6D measures 80 for ISO 100, but 153 for ISO 200.

My point was that when you select ISO 100 on a Canon camera, you have selected the cameras actual native base ISO (whether it is 80 or 93 or 100 in actuality under the scenes.) Whether the measured ISO is 80 or 100 doesn't change anything, though. Instead of ISO 125 for the first 1/3rd stop push, you would have a measured ISO 105, and for the first 1/3rd stop pull you would have a measured ISO 140. Same difference in the end...you are clipping either blacks or whites, and baking in that loss of information into the final output that is actually recorded into the RAW file.

So DxO measures an ISO less than 100 when ISO 100 is selected.  But DxO only measures full stops. It would seem they have the capability to measure 1/3-stops, but they don't.  Perhaps it's just too much work.  Or perhaps they do measure them, use them in Optics Pro, but don't publish them. I wonder, if they actually measured ISO 160, perhaps the real measurement would be closer to the setting.  Canon's internal processing isn't necessarily as one would expect, as shown by the clandestine ISO boost with very wide aperture lenses.

I don't know why DXO doesn't measure third stops. Or half stops, for that matter. Either way, I don't really think it matters. All I know is that the way Canon manages their non-full-stop ISO settings irks me. :P I wish they would just use electronic gain at the pixel for every ISO setting (up until those very high ISO settings...seems whatever they do at very very high ISO results in better output than simple gain off the pixel.)

Anyway, I don't suspect Canon will be changing their MO any time soon. Even if they improve DR, I expect to be saddled with quirky third-stops for a long time to come (however they work, whatever their consequences.)

1457
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 11, 2014, 04:41:01 PM »
For ISO 160, Canon does a downstream third-stop "pull", which again costs you a third of a stop dynamic range (crushing blacks). Since it is a post-read push, it reduces read noise by a third of a stop (hence, the notion that ISO 160 is "cleaner" than ISO 100...it is, by a minuscule amount.)

Bill Claff's data show the 'jagged' relationship of read noise vs. ISO, where the 'valleys' of noise are 160-320-640-.  But, his data show the same jagged plot for DR vs. ISO, except the 'peaks' are the inverse of noise, with the greatest DR at 160-320-640.  Your statement of a loss of 1/3-stop ISO at 'pulled' settings doesn't jive with his data that ISO 320 has greater DR than ISO 400 as well as less noise.

You express certainty that ISO 100 is the true base, but I think a model where the real base ISO was somewhat lower than 100, such that multiples of 100 are a light push, multiples of 125 are a harder push, and multiples of 160 are a light pull, would explain Claff's data.

Any idea what's going on there?

I am not exactly sure how Bill Claff derives his data. Logically, I am unsure how one could gain dynamic range by shifting the signal after the pixels are read. Think of what Canon does as a digital post-process exposure boost. If you utilize 100% of the sensor's available dynamic range at ISO 400, you have information at every level from 0 through 65535. If you "push", you shift the whole histogram to the right...and clip the highlights. If you "pull", you shift the whole histogram to the left...and clip the shadows. Now, in post, your probably working with a tool like Lightroom, in which case your actual RAW data is never changed, and you can shift the digital signal around to your hearts content without loss of precision or data. However with third-stop ISO settings in a Canon camera, the results after the push or pull are baked in. If you push, anything that clipped past pure white is permanently pure white. You cannot recover it. Similarly, if you pull, anything clipped past pure black is permanently pure black.

I'm wondering if the apparent reduction in the read noise floor is what gives Claff's results this third-stop "pull" boost to DR. If he is ignoring the fact that the highlights shift down along with the rest of the exposure, and simply uses a the 5D III FWC as a constant for the upper limit for the signal, then I can see how dynamic range would theoretically increase. The read noise floor is indeed lower, however because of the pull, you actually destroyed data in the shadows, and the camera itself is not actually utilizing the headroom gained by shifting highlights down (since this all occurs AFTER the sensor has been read)...hence the "loss" of a third of a stop dynamic range.

Yes, I get the effects of in-camera push and pull.  Where does your certainty that ISO 100 is the true base come from?  As I stated, Claff's data could be explained if base ISO is slightly lower than 100.  On the ML forums, one if their devs determined that using ISO 100 multiples with 'a little bit of negative gain' yielded the highest DR, and that's also consistent with base ISO being less than 100.

Well, certainly. Literal ISO is always different. It isn't even consistent within the same brand, though. For example, according to DXO, "measured" ISO 100 on the 5D III is 80, measured ISO 200 is 160, where as "measured" ISO 100 on the 70D is 93 and ISO 200 is 160. The 6D measures 80 for ISO 100, but 153 for ISO 200.

My point was that when you select ISO 100 on a Canon camera, you have selected the cameras actual native base ISO (whether it is 80 or 93 or 100 in actuality under the scenes.) Whether the measured ISO is 80 or 100 doesn't change anything, though. Instead of ISO 125 for the first 1/3rd stop push, you would have a measured ISO 105, and for the first 1/3rd stop pull you would have a measured ISO 140. Same difference in the end...you are clipping either blacks or whites, and baking in that loss of information into the final output that is actually recorded into the RAW file. 

1458
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 11, 2014, 03:17:09 PM »
I'm quite content with the DR of my 5D3 ;D




Lovely shot. But is it only me who feels that a bit more detail in the burnt out sun area would be nicer? A grad filter or change in lighting. Just wondering... I know the hot spot is interesting but JUST A BIT MORE DETAIL perhaps?


Personally, I like it how it is. I might actually increase the glare just a bit. Not every region of a photo needs more detail, sometimes lower detail and less contrast is exactly what you want.


It's a lovely shot and one I really like...but that sky is blown out. An ND grad would render the sky darker and probably lost detail in the darker sky areas. It would have increased contrast where it wasn't wanted. The only way to have fixed this here is to have taken a 2nd photograph but at a 2-3 stop darker exposure and then blended the highlight areas carefully in photoshop using a layer. Shadows can be pulled but clipped highlights are not recoverable. It's also important to render the sky brighter than the foreground, another error I regularly see where ND grads are employed. If an ND grad was used with the above photo, the sky would have been darker than the foreground and wouldn't look right.

Many people here are talking about the D800's extra DR, but the truth is that it's only in the shadows...or rather it's the push-ability in the shadows during post production with low iso noise is really what is being talked about. Highlight clipping / blown highlights occurs at pretty much the same between the 5DIII and D800. So it's not really any extra DR, just better iso thresholds in the deep black areas.


You wouldn't consider the blown out sky to be artistically desirable? Personally, I think the sky is exactly how it should be (maybe even more blown out, covering more of the corner, would be even better.) With our own eyes, we couldn't see this scene without the sky being much brighter than it is depicted here in this photo. I think it speaks to realism that V8 captured it the way he did, and I think it maintains a certain artistic flare.

Not everything in photography is nor should be about recovering every last scrap of detail, so much so that you could see sunspots in the glaring sun...I think that would result in a grave imbalance in a shot like this.

1459
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 11, 2014, 03:14:16 PM »
High ISO...I'd call that ISO 800 or above. ISO 100, 200, and 400 I consider low ISO, although ISO 400 is kind of in the middle there, and others might have a different opinion.

I don't have a specific study. It's a simple observation, however on that I have been making over the last four years or so. (FYI, I've moderated photo.stackexchange.com since 2010, and have encountered and chatted with quite a number of photographers over the last four years from a wide range of photographic endeavors.) How many white Canon lenses do you see at pretty much every sporting event around the globe? Hundreds to thousands at each and every event. Canon dominates sports, hands down, no question. They really dominate action, not just sports. I spend a lot of time out in nature, and meet a fair number of nature photographers. The very vast majority of the people I've met out in the wilderness, including both wildlife and bird photographers as well as landscape photographers, overwhelmingly have Canon equipment. Canon 1D IV, Canon 5D III, and Canon 1D X are becoming almost ubiquitous in the wildlife and bird world. Canon great white lenses, 300s, 500s, and 600s, are extremely common (particularly the 500 Mark Is...lot of wildlifers and birders use that lens, guess it's at a sweet spot of weight and cost). I've met a few who have Nikon equipment, two of whom use D800's for bird photography. I know of one (now a good friend) who uses Pentax and Nikon. I also know and have encountered/chatted with a decent number of wedding & portrait photographers. Most use the Canon 5D II. A few still use the 5DC (they don't seem to care about resolution). Some use the 5D III (and all of the 5D line wedding photographers had one consistent complaint before the 5D III: Sucky AF.) I know of several wedding photographers who use Nikon and other brands (some have gone to mirrorless as of late, with a variety of brands.) I know two wedding and portrait photographers who use Nikon exclusively. One uses a D800 and D3, the other uses a D7000, with a D800 planned for very soon. I would say that Nikon seems to have a growing following in the strait portraiture arena...not so much for DR, but for the sheer amount of detail the D800 or D600 bring to the table...seems that ridiculous, razor-sharp detail that brings out every single pore is really "in" right now, and there is no question that the D800 offers that in spades.

So, sorry, I don't have an official study for you, but it really isn't a difficult observation to make. Just look around.  It's a very well-educated guess. The number of cameras and lenses that you can spot in the world that say "Canon" on them vastly outnumber  any other brand. Of those, the biggest group that uses the most cohesive set of camera features are the action shooters. Sports/Olympics, Wildlife, Birds...and you can throw in car racing, air shows (know a few guys who do this, damn good at it too), kayaking, boat racing, pretty much anything you could remotely call a sport, or has moving subjects...the camera is going to be at a higher ISO setting, and is probably a 5D III or a 1D X. The next two biggest groups would be Wedding & Portrait, and Landscapes. Not sure which is bigger...seems pretty evenly split here in Colorado, but if you hit larger metropolitan areas, I would make the educated guess that Wedding and Portrait photographers would end up significantly out-pacing the Landscape photographers (and I mean real landscape photographers...I know more people than I can count who use entry level cameras, Canon, Nikon, Sony, Pentax, everything...and call themselves landscape photographers, but their work wouldn't land on anyones walls (no offense to anyone like this, but)...blown clouds, random people in the frame, lack of interesting composition, effectively point-and-shoot mountain peaks and a few scattered rocks or stubby evergreen trees here and there, never any post processing, thrown up on Imgur, PhotoBucket, or Facebook.)

I honestly don't have all that much knowledge about studio photographers. I can't really say how big a customer segment studio photographers might be for Canon...but I guess big enough for them to create the 1Ds line in the past. What I DO know about studio photography, it seems to lean medium format (or maybe Leica S-system) a lot more than it leans Canon, Nikon or Sony. Phase One also seems to be the brand I hear about most from the studio photogs I do know or have crossed paths with.


If you present something like some sort of statistical fact, you need to prove it. If it’s just an educated guess based on personal observation, it’s not a very strong base to build an argument on because it has no more value than any other personal observation claiming the opposite unless you have some proven authority on the subject. I have no way of knowing how well-educated your guess is. Please point me to your scientific publications in this particular field of research because moderating a website about photography doesn’t make someone a statistical expert. Ask yourself if the people visiting the website you moderate are even close to a cross section of camera users worldwide (that needs to be true if you want to extrapolate). 


Alright, fair enough.

Anyway, if iso 800 and above is high iso, just look at the facts as they are presented by DxO mark. I’m not looking for (yet another) debate about DxO sensor scores. Just look at their measurement.


 
ISO 800 (high iso by your standard) on the Nikon D800 is as good as ISO 100 on the 5D3 (that’s a 3 stop ISO difference).
At ISO 3200 and above the 5D3 gets me marginally better results.

The reason I think “playing catchup” in the DR department is true for Canon is because they are not purposely crippling the 5D3. It’s not like the autofocus or fps on the 5D2. Canon doesn’t have a senor in production with the low iso DR capabilities of other manufacturers and at some point they need to catchup.

But don’t get me wrong, I’m not arguing the D800 is a better camera or Nikon has better camera systems. I know there is more to photography than dynamic range and sensor resolution. I’m happy with my 5D3 and my Canon TS-E 24mm f/3.5L II is much better than the Nikon 24mm PC-E but as Canon users we probably will be better off if we encourage Canon to catchup in the DR department.


Sure, I don't disagree that Canon lags behind in the low ISO arena. I just don't think having that extra two stops of dynamic range is actually quite as important as may photographers thing. It improves editing latitude, that's all it really does. One way or another, the things we view our photographs on...screen and print...have considerably less dynamic range. Screens average 8 stops, print averages 6-7 stops (or even as little as 5 stops). If anyone has tried to print on a low dMax paper with a natural white L*, they would understand how difficult it can be to compress even 11 stops of dynamic range into less than half that. It is not an easy task, and one has to be careful to avoid posterization and color fidelity issues in the shadows, where the available data (even after you remove noise and recover DR in the case of Canon sensors) is "sparse" compared to the midtones and highlights.

Compressing 13 stops of native dynamic range into 8 stops of screen DR is similar. You can lift the shadows, but beyond a certain point (around 3-4 stops) you run into the same issues...information in those low signal strength areas is sparse...discrete level transitions are often harsh, and ironically, the addition of noise is often the only way to combat posterization. Don't get me wrong, having more DR is useful, but until we have more available dynamic range on screen and in print (both of which I think are coming), being able to push shadows around by 4-6 stops doesn't buy you as much as you might think.

Pushing around exposure with more dynamic range than your screen is like tonemapping a 32-bit HDR image into 16-bit. You have twice the dynamic range (really, more than that, since 32-bit is floating point...so you have MANY times the dynamic range) as your output target. If you have a lot of shadow tones bunched up at the left end of your histogram, and have done much HDR, you would know how difficult it can be to tonemap all that extra dynamic range into the mere few shadow stops you have in a 16-bit scalar image. It's DIFFICULT! The results usually end up having that "HDR Look", where you end up with funky noise issues, posterization, odd tonal gradients, poor color fidelity, and even tonal inversions in the shadows.



I totally agree, though. I am not saying Canon shouldn't continue to push the envelope with their sensors. I just think dynamic range gets FAR more importance put on it by photographers than it really deserves. I think other in-camera factors are more important for most forms of photography, particularly any form of action photography (hence my earlier arguments about my observations of Canon's distribution of customers...honestly, observe a little yourself, and I think you'll come to the same conclusion...sports/action shooters are Canon Photography's bread and butter! :P) Autofocus is probably the most important factor for action shooters. After AF, I would say frame rate. If you don't capture the right frame in focus, it really doesn't matter a wit how much dynamic range you have...you either have the wrong frame where that basketball  players arm is half out of the frame, or the image is miss-focused. You might have 14 stops of missfocused arm-clipping wonder...but it ain't usable!

Finally...I do believe Canon will deliver more DR. How and when, and at what cost, I honestly don't know. However having been a Canon customer for almost five years now, I've only grown more fond of them. They make a good product, they put in the effort to ensure their product, whatever it is (even if it is perceptually inferior to the competitions) is solid, reliable, and always backed by the best customer support in the industry (and I speak from experience on a few occasions where I needed to send my lenses in for repair.) Canon, in my opinion, very carefully listened to the most important and broadly held demands of their customers with the designs of both the 1D X and 5D III. I mean, people here on CR and on many other forums like DPR seemed surprised that Canon chose to release the 1D X with "only" 18 megapixels, and were surprised that the 5D III "only" got a 1mp boost. Personally, I felt that Canon delivered exactly what their customers were asking for...as I'd heard, and even called for myself, the following on photo.stackexchange.com (that site I moderate...which actually has a very broad worldwide participation, so I think it is a decent source of information like this):

"I'm so sick and tired of the megapixel wars! I want better pixels, not more pixels!"
"I don't care about pixel count. I just wish Canon would make less noise at high ISO."
"Fewer megapixels! All I need is 10mp. I can print very large with just 10mp. Give me low noise ISO 204,800!! I want to photograph the aurora as I see them with my own eyes."
"The 1D III AF system really needs to be replaced with something much better. Something like Nikon's D3 AF system." (The guy was talking about a denser, reticular AF design, which Nikon actually had first.)
"The 5D II 9pt AF is horrible. I just don't have the points I need to focus where I need. I wish they would make the hidden AF assist points selectable." (That guy got a lot more than he asked for!)

I heard a lot of this kind of stuff from 2010 (and maybe late 2009) through 2012. It was only AFTER the release of the D800 and the 5D III that the things people complained about regarding Canon's equipment changed. Once they actually had fewer megapixels and better high ISO, they stopped complaining about it. Now, they complain about not having more DR (although I know for a fact that many of them are action shooters, and rarely use ISO settings below 800...so more DR wouldn't do them a bit of good. Fewer megapixls, bigger pixels, and less noise at high ISO, however, would do them a LOT of good! :P)

Personally, I was one of those hoping the 5D III would get more DR and more megapixels. I wanted a landscape camera. I still do. Ironically, the 5D III will still fill a role as a bird and wildlife photography camera given its feature set, and it is one of the top two things on my list of camera gear to buy this year. So I still find the 5D III to be a great camera. Regardless, I still hope Canon releases a landscape camera with lots of megapixels and lots of DR at some point in the future, because that is the one area of my work where more megapixels and more DR are literally the most useful things (AF and frame rate don't matter for jack when it comes to landscapes...but I also admit that as a landscape photographer, I become part of a much smaller minority of Canon customers, so I don't expect them to release anything with tons of MP and more dynamic range in the price range I want...I suspect the camera will be a 1Ds X at $5000. :\)

1460
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 11, 2014, 02:40:06 PM »
For ISO 160, Canon does a downstream third-stop "pull", which again costs you a third of a stop dynamic range (crushing blacks). Since it is a post-read push, it reduces read noise by a third of a stop (hence, the notion that ISO 160 is "cleaner" than ISO 100...it is, by a minuscule amount.)

Bill Claff's data show the 'jagged' relationship of read noise vs. ISO, where the 'valleys' of noise are 160-320-640-.  But, his data show the same jagged plot for DR vs. ISO, except the 'peaks' are the inverse of noise, with the greatest DR at 160-320-640.  Your statement of a loss of 1/3-stop ISO at 'pulled' settings doesn't jive with his data that ISO 320 has greater DR than ISO 400 as well as less noise.

You express certainty that ISO 100 is the true base, but I think a model where the real base ISO was somewhat lower than 100, such that multiples of 100 are a light push, multiples of 125 are a harder push, and multiples of 160 are a light pull, would explain Claff's data.

Any idea what's going on there?

I am not exactly sure how Bill Claff derives his data. Logically, I am unsure how one could gain dynamic range by shifting the signal after the pixels are read. Think of what Canon does as a digital post-process exposure boost. If you utilize 100% of the sensor's available dynamic range at ISO 400, you have information at every level from 0 through 65535. If you "push", you shift the whole histogram to the right...and clip the highlights. If you "pull", you shift the whole histogram to the left...and clip the shadows. Now, in post, your probably working with a tool like Lightroom, in which case your actual RAW data is never changed, and you can shift the digital signal around to your hearts content without loss of precision or data. However with third-stop ISO settings in a Canon camera, the results after the push or pull are baked in. If you push, anything that clipped past pure white is permanently pure white. You cannot recover it. Similarly, if you pull, anything clipped past pure black is permanently pure black.

I'm wondering if the apparent reduction in the read noise floor is what gives Claff's results this third-stop "pull" boost to DR. If he is ignoring the fact that the highlights shift down along with the rest of the exposure, and simply uses a the 5D III FWC as a constant for the upper limit for the signal, then I can see how dynamic range would theoretically increase. The read noise floor is indeed lower, however because of the pull, you actually destroyed data in the shadows, and the camera itself is not actually utilizing the headroom gained by shifting highlights down (since this all occurs AFTER the sensor has been read)...hence the "loss" of a third of a stop dynamic range.

1461
EOS Bodies / Re: Will Canon ditch the AA Filter?
« on: January 11, 2014, 03:49:09 AM »
Yes, you can get slightly more detail without an AA filter, but for many forms of photography the minor advantage is nothing compared to the major disadvantage. Also, some here have argued that if the number of MP are increased to the point where the sensor out resolves most lenses, you can get away without it. My question is, in that case, why do you need no AA filter? If the output is already optically blurred at a pixel level by the lens, no lack of AA filter will sharpen it back up. And then what happens if you buy a new yet-to-be-released super sharp lens in the future and it does out resolve the sensor at certain apertures? Moire, and you being forced to defocus some shots. So why not just keep the AA filter?

If we had 32MP APS-C sensors (83MP FF), I would just be jumping for joy if a lens came out that out-resolved it. Really, I doubt any lens will ever come out that significantly out-resolves current lenses at their best (usually macro lenses at f5.6). If we pick something good enough for those it shouldn't be a problem.

Resolving power increases with aperture. At f/5.6, MTF 50% (standard measure for photography), with a diffraction limited lens you have 123lp/mm of resolving power. We already have sensors that resolve that much, and it is certainly no stretch to say that many lenses are diffraction limited at f/5.6. At f/4, diffraction limited MTF50 resolving power increases to 173lp/mm. At f/2.8, diffraction limited MTF50 resolving power increases to 247lp/mm. Now we are really pushing resolution, however making a truly diffraction limited f/2.8 lens is a much more difficult ordeal than making a diffraction limited f/4 or f/5.6 lens. Zeiss once had a lens they had designed explicitly to test high resolution films, which was capable of resolving about 400lp/mm (which would have been diffraction limited around f/1.7-1.8)...however I'd be doubtful if many major brand name lenses, including any from Ziess these days, was actually diffraction limited at apertures above f/4. If we assume future lenses get better, then the only way they could resolve more would be to increase resolving power at apertures wider than f/4...maybe up to 200lp/mm at f/2.8 (not diffraction limited, but still better than anything we have today.)

Now, don't forget that the final resolving power of an entire camera system is a convolution of its component parts. We can't really know the exact PSF of any lens or camera sensor (some manufacturers probably do, but they don't actually publish the information), but we can use a simple formula to approximate: SQRT( lensBlur^2 + sensorBlur^2) allows us to determine total system blur, and from there we can extend the formula to tell us a whole lot of things. Because of the nature of system resolving power as demonstrated by this formula, you can never actually reach the maximum resolving power of your least powerful component...you can only approach it. That means, no matter how high your sensor resolution is, you could never actually "out"-resolve a diffraction limited f/4 lens at f/4...at best, you would be able to resolve 172.99999 with the whole system. The notion of either a sensor or a lens "outresolving" the other is a bit of a misnomer.

To demonstrate how ironic and even a little ridiculous this little relationship is...you would need an APS-C (1.6x) sensor with 1115000x745000 pixels (22.3mmx14.9mm) to render 172lp/mm spatial resolution in an output image with a diffraction limited f/4 lens...that is an 830.7 gigapixel sensor! :P Obviously that's impossible...the pixels would be 10 nanometers in size, and pixels that small would be too small to allow light through, so the sensor simply wouldn't function (not, at least, with visible light...it might function with gamma rays. :P)

The only other way to increase total system resolution is to use a lens with higher resolving power, which can only be achieved at wider apertures. If we assume we have our 200lp/mm f/2.8 lens (which gives us plenty of headroom to work with), to resolve 173lp/mm we would need a 158mp APS-C sensor (15380x10277 pixels @ 22.3mmx14.9mm). That comes out to 411mp FF (24830x16553 pixels @ 36mmx24mm). At any aperture below f/2.8, your total system resolving power would again become diffraction limited, and would be less than 173lp/mm...yet still higher than if you used a sensor with fewer megapixels.

Sorry if that comes off as too complicated. However it is the only way to be clear about "resolution". Sensors do not outresolve lenses. Lenses do not outresolve sensors. The two work together to create a final outcome, and that final outcome will continue benefit by increasing the resolution of either lens or sensor for a long time to come. At the moment, the best APS-C sensors are resolving around 124lp/mm. The best FF sensors resolve far less, and further still the actual output resolution of our actual photographs is even less (and if we apply noise reduction, EVEN LESS!). We have a LOT of headroom before we run into the limitations offered by current f/4 and faster lenses (many of which, while not diffraction limited, still offer resolving power that can be well above 124lp/mm) and experience diminishing returns. An 32mp APS-C/83mp FF sensor doesn't even scratch the surface of how far we could take sensor resolution before we started experiencing diminishing returns the the degree where there was no point in investing in shrinking pixels any further.

I'd actually bet that we could take pixels right down to the wavelength size of light (the point at which pixels become too small to allow visible wavelengths of light through), and still have a useful gain. As a matter of fact, with the new generation of small form factor CMOS image sensors (CIS) that will become mainstream through 2014-2015...the tiny sensors used in smartphones and the like...reaching 0.95µm, or 950nm, we are already well into the realm of wavelength-size or smaller for about half the range of infrared light. Within the next generation or two of smartphone CIS designs, pixels will be as small as they possibly can be...about 750nm, and we won't be able to shrink them any further without filtering out red light!!

This convolution of lens and sensor resolution should also make it clear that there won't necessarily be any benefit from removing AA filters for a very long time to come as well. Everything I've discussed here is at MTF50, or a 50% contrast level. As pixel size shrinks, sensors will be able to resolve detail at lower and lower levels of contrast as well, so the spatial resolutions scale up (more lp/mm) as pixel size drops. I wouldn't go so far as to say a sensor could produce any real meaningful result at MTF10 (10% contrast), but it is very likely that small pixels will be able to resolve detail at 20%, 15% contrast...and lens resolving power at those levels is considerably higher than it is at 50% contrast (i.e. at f/4 MTF15, resolution is closer to 350lp/mm!) We will always need AA filters...there really isn't any good reason to get rid of them.

The only use case where not having an AA filter might result in sharper detail would be if you only ever photograph scenes without ANY repeating patterns of any kind whatsoever. Landscape photography is the most likely scenario where you would encounter only purely random detail at nyquist, but even in the case of landscapes...personally, I don't like them being too sharp. I find that the best landscape photography tends to have that certain "softish" quality to them, clean edges, rather than finger-dicing sharp edges, kind of like "bloom" in modern games makes everything "soft", but in general just much better.

1462
EOS Bodies / Re: Will Canon ditch the AA Filter?
« on: January 11, 2014, 02:23:58 AM »
I don't buy that. Many things can be corrected in post, and moire isn't one of them. pcpro.co.uk aren't renowned for their photographic expertise. In this case I get the impression they just read a couple of marketing gimmicks and wrote it as fact.

So you would be saying that Adobe's moire correction brush cannot work, right?

Moire correction in post currently only takes care of color moire. In that, it only eliminates the color, but not really the moire. There are some moire patterns that Adobe's moire brush can deal with better than others, and if all you have is very light moire (like the kind you might bet with a just barely too weak AA filter), you might be able to clean it up entirely...albeit with consequences, you will lose detail in one way or another.

However, if you completely remove the AA filter from the sensor, there is absolutely no way that Adobe's moire brush is going to clean it up. It will remove color from it, but the actual moire pattern is baked into the image. Moire is a complex convolution, and without knowing the parameters of the interference that created it, you can't do anything to reverse it once it is baked in.

That's WHY we have optical low pass filters...they are really the only way to prevent aliasing and moire.

1463
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 11, 2014, 01:56:50 AM »

I levy the question again...how many of your shots are at ISO 100, and more importantly, of those ISO 100 shots, in how many did you desperately NEED to push shadows more than two or three stops?

You keep falling into the logical fallacy trap. The logic you are using can be applied to every argument ever made against improving imaging technology.

An improvement in technology is an improvement, regardless of how many use the improvement.

You've posted large blocks of text concerning how many photographers use Canon (one might even say heavy cheer-leading), but again that's a red herring when applied to the context of technological improvements. Again, an improvement is an improvement, regardless of usage or popularity.

Sorry, but you are still misunderstanding. I am not arguing against improving technology. I am trying to make the point that more DR is not as important as a great number of photographers these days think it is, the same great number of photographers who regularly bitch about Canon not having a mere two stops of additional DR (DR they probably wouldn't use most of the time.) I've never once said Canon shouldn't improve DR...your reading into something that simply isn't there (I actually have stated I have great confidence that Canon WILL improve DR.)

If someone has a critical need for more DR RIGHT THIS VERY MINUTE, they could have more DR. Just buy a D800, or a D600. I mean, that's the only option right now. I'm not arguing against DR...if you really need it, you really need it.

It just isn't nearly as common to need that much dynamic range, and it isn't so critical to image quality in general that it is the sole thing that would improve most photographer's work. For anyone who shoots ISO 400 and higher most of the time, I would actually offer that Canon currently offers the best gear which offers the best overall image quality for the greatest number of situations: Better AF, better high ISO, faster maximum frame rates, deeper frame buffers, more consistent and continuous frame rates after the frame buffer is full, and still good DR (even though it isn't "the best") that serves the majority of photographers needs most of the time.

My posts so far are making the point that for the majority of photographers who bitch and moan and complain about Canon's low ISO DR, even if they had it, they wouldn't actually need it most of the time. I've never made the point that we shouldn't try to progress technologically on all fronts (of course we should.)

The same general argument I'm trying to make could be applied to cameras that have no AA filter on the sensor. That seems to be as much a fad right now as more DR. I am always surprised by how many photographers blather on about how they want Canon to remove the AA filter from the next camera they want, "just like Nikon did." There are SOME cases where not having an AA filter can be useful, but it is far from a particularly desirable thing. Photographers just want it because its the new thing, and its "that thing the other guy has that I want." The lack of an AA filter, unlike DR, can actually result in a detrimental impact on IQ in a lot of circumstances. AA filters are actually useful and necessary most of the time, to avoid overly sharp and "nonsense" detail that can actually detract from overall IQ. I would really like to know how many photographers that want the AA filter to be removed would actually truly benefit from that...vs. how many would actually suffer from it. I think landscape photographers are really the key group who might benefit from no AA filter...but as others have argued, landscape photographers are a rather small segment of photographers at large. 

1464
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 11, 2014, 01:24:09 AM »
I'm quite content with the DR of my 5D3 ;D




Lovely shot. But is it only me who feels that a bit more detail in the burnt out sun area would be nicer? A grad filter or change in lighting. Just wondering... I know the hot spot is interesting but JUST A BIT MORE DETAIL perhaps?


Personally, I like it how it is. I might actually increase the glare just a bit. Not every region of a photo needs more detail, sometimes lower detail and less contrast is exactly what you want.

1465
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 11, 2014, 12:06:19 AM »
I've always loved your cars! Great stuff!

Thanks. I've learned a ton from your technical ramblings, and find it very informative. I haven't posted the "EOS Bodies" forum in quite some time, but I see everyone's still whining about the same thing :)

Oh, the DR debate has actually died down a lot as of late. It's actually kind of surprising...this is the first time since he-who-shall-not-be-named was everbanned that it became a major issue again.

I am still interested in an answer to my question...I really wonder how often people actually need to lift low ISO shadows more than a couple stops. I am entirely willing to admit I'm wrong if something like a hundred people said they always need to lift four or five stops...it would really blow my mind...but I'd still happily admit I was wrong. (I don't suspect I'll have to admit anything, though...if people really need to lift that much that frequently, they should probably head back to Photography Basics 101...)

1466
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 11, 2014, 12:03:22 AM »

Of those low ISO images, for how many did you have to blow highlights or block shadows to preserve the other end AND you had a difference of not more than 2-stops such that the greater low ISO DR of the Sony/Nikon sensor would have solved the problem AND of that subset, how many of those shots were rendered unusable by the lost detail in the shadows or highlights? 


You could apply this logic to every facet of image taking when arguing against technological improvements:


"Of all your high ISO shots, how many were so truly noisy that you simply couldn't use the image?"

"Of all those shots of grizzly bears, how many were truly ruined by using only a 2-stop IS system?"
 
etc....

The point is that Canon addressed those issues. We no longer rely on 2-stop IS systems, we have 4- to 5-stop IS systems. We no longer have to worry about noise at ISO 3200 or even ISO 6400, with the 1D X, 5D III, and 6D, they are amazingly clean.

Canon addressed the most vocal demands of their customers. Solving the problems you listed above were at the top of the customer demand list. Does no one remember what all the pros were literally demanding from Canon before the D800 hit the streets? Fewer megapixels! Better high ISO! An AF system that doesn't suck like the 1D III's did! Canon delivered what their customers asked for...so no, we no longer have to deal with the issues you listed.

I also believe Canon will deliver on the DR front. Why? Because its what the largest and most vocal group of Canon users are screaming for now. Canon users weren't calling for more dynamic range before the D800...they were all largely satisfied with what they had. It's only SINCE the D800 that the customer demand has changed...which indicates it is more a result of "Hey, that other guy over there has more DR than I do! I want more DR, too!" syndrome (all while concurrently ignoring that they already have better AF, faster frame rate, better frame buffer handling, better glass options, better...), than the all-encompassing, singularly important, most absolutely critical factor for IQ that photographers thing it is. I mean...no one complained about it when 11-12 stops was "all" ANY camera offered, including $60,000 MFD systems (which, ironically, is what they are STILL limited to...an yet no one complains!)...

Anyway...this is the same old thing that always crops up. Yeah, more DR == good. DR != Single Most Important IQ Factor (SMIIQF...pronounced like a "squeaky chick fart"). I'm out! Peace out!  8)

1467
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 10, 2014, 11:52:04 PM »
I'm quite content with the DR of my 5D3 ;D




I've always loved your cars! Great stuff!

1468
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 10, 2014, 11:42:06 PM »

Statistically, higher ISO settings are used more frequently these days than lower ISO settings...so it really baffles me that this is such a broad and ubiquitous issue. I am not saying that better low ISO DR is a bad thing, of course it's good...but it is still only one IQ factor out of many. Given how well the D800 has sold, I wouldn't go so far as to say Canon is now playing "catchup" in the low ISO DR arena yet.

Which statistic?  Can you point to the result of a scientific study?

And what exactly do you mean by “higher ISO settings”? Higher than 100, 400, 800, 1600…?

I buy into a camera system and the sensor is just one part of the system, but I would love to have the low iso DR of the D800 in my 5DIII.
It’s the one area Canon is really outperformed by just about all other manufacturers. I think they should address that for the next generation and they probably will.
All manufacturers are playing catchup in some way, because none of them are the best at everything.

High ISO...I'd call that ISO 800 or above. ISO 100, 200, and 400 I consider low ISO, although ISO 400 is kind of in the middle there, and others might have a different opinion.

I don't have a specific study. It's a simple observation, however on that I have been making over the last four years or so. (FYI, I've moderated photo.stackexchange.com since 2010, and have encountered and chatted with quite a number of photographers over the last four years from a wide range of photographic endeavors.) How many white Canon lenses do you see at pretty much every sporting event around the globe? Hundreds to thousands at each and every event. Canon dominates sports, hands down, no question. They really dominate action, not just sports. I spend a lot of time out in nature, and meet a fair number of nature photographers. The very vast majority of the people I've met out in the wilderness, including both wildlife and bird photographers as well as landscape photographers, overwhelmingly have Canon equipment. Canon 1D IV, Canon 5D III, and Canon 1D X are becoming almost ubiquitous in the wildlife and bird world. Canon great white lenses, 300s, 500s, and 600s, are extremely common (particularly the 500 Mark Is...lot of wildlifers and birders use that lens, guess it's at a sweet spot of weight and cost). I've met a few who have Nikon equipment, two of whom use D800's for bird photography. I know of one (now a good friend) who uses Pentax and Nikon. I also know and have encountered/chatted with a decent number of wedding & portrait photographers. Most use the Canon 5D II. A few still use the 5DC (they don't seem to care about resolution). Some use the 5D III (and all of the 5D line wedding photographers had one consistent complaint before the 5D III: Sucky AF.) I know of several wedding photographers who use Nikon and other brands (some have gone to mirrorless as of late, with a variety of brands.) I know two wedding and portrait photographers who use Nikon exclusively. One uses a D800 and D3, the other uses a D7000, with a D800 planned for very soon. I would say that Nikon seems to have a growing following in the strait portraiture arena...not so much for DR, but for the sheer amount of detail the D800 or D600 bring to the table...seems that ridiculous, razor-sharp detail that brings out every single pore is really "in" right now, and there is no question that the D800 offers that in spades.

So, sorry, I don't have an official study for you, but it really isn't a difficult observation to make. Just look around.  It's a very well-educated guess. The number of cameras and lenses that you can spot in the world that say "Canon" on them vastly outnumber  any other brand. Of those, the biggest group that uses the most cohesive set of camera features are the action shooters. Sports/Olympics, Wildlife, Birds...and you can throw in car racing, air shows (know a few guys who do this, damn good at it too), kayaking, boat racing, pretty much anything you could remotely call a sport, or has moving subjects...the camera is going to be at a higher ISO setting, and is probably a 5D III or a 1D X. The next two biggest groups would be Wedding & Portrait, and Landscapes. Not sure which is bigger...seems pretty evenly split here in Colorado, but if you hit larger metropolitan areas, I would make the educated guess that Wedding and Portrait photographers would end up significantly out-pacing the Landscape photographers (and I mean real landscape photographers...I know more people than I can count who use entry level cameras, Canon, Nikon, Sony, Pentax, everything...and call themselves landscape photographers, but their work wouldn't land on anyones walls (no offense to anyone like this, but)...blown clouds, random people in the frame, lack of interesting composition, effectively point-and-shoot mountain peaks and a few scattered rocks or stubby evergreen trees here and there, never any post processing, thrown up on Imgur, PhotoBucket, or Facebook.)

I honestly don't have all that much knowledge about studio photographers. I can't really say how big a customer segment studio photographers might be for Canon...but I guess big enough for them to create the 1Ds line in the past. What I DO know about studio photography, it seems to lean medium format (or maybe Leica S-system) a lot more than it leans Canon, Nikon or Sony. Phase One also seems to be the brand I hear about most from the studio photogs I do know or have crossed paths with.

1469
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 10, 2014, 11:17:07 PM »
This is interesting. I've never tested it in pictures, but I know in video that ISO noise is non-linear. So technically 160 is better than 100, and 320 is better than 200 or 250, so on and so forth. Wouldn't it be the same in terms of pictures?

This is a Canon quirk. It's actually one of the very few things I hate about my Canon cameras. Canon doesn't have native third-stop ISO settings. Unlike Nikon and Sony (and probably others) who use electronic gain to boost the signal on-sensor for all ISO settings (up to a certain point, like ISO 1600 or 3200, after which more complex means are usually employed to boost the signal), Canon uses a downstream secondary amplifier to additionally adjust for third-stop ISO settings.

This is also where the "ISO 160 base ISO" MYTH for Canon cameras comes from. Canon's native base ISO is literally ISO 100, no question. Canon employs a full stop gain for ISO 200. For ISO 160, Canon does a downstream third-stop "push" for ISO 125, which actually COSTS you a third of a stop of dynamic range (clipping highlights). Since it is a post-read push, it also amplifies read noise by a third of a stop (not much, but if you do end up having to push shadows around a LOT, you notice it.) For ISO 160, Canon does a downstream third-stop "pull", which again costs you a third of a stop dynamic range (crushing blacks). Since it is a post-read push, it reduces read noise by a third of a stop (hence, the notion that ISO 160 is "cleaner" than ISO 100...it is, by a minuscule amount.)

The push pattern is used for all third-stop settings just above full stops (125, 250, 500, etc.) The pull pattern is used for all third-stop settings just below full stops (160, 320, 640, etc.) Depending on the camera, this pattern is abandoned at higher ISO. It used to be that all Canon cameras employed this push/pull third-stop pattern up through ISO 1600, after which a different and more complicated approach was used. With the 5D III and 6D, I believe the pattern is employed up through ISO 6400, beyond which their more complicated approach is used. The 1D X does not seem to exhibit the same differences in noise and DR for third stops as all the rest of Canon's cameras. I am not sure why, but for whatever reason, the 1D X third stops are much better and more linear overall, and therefor much more usable (with no loss of DR.)

1470
EOS Bodies / Re: Will Canon Answer the D4s? [CR2]
« on: January 10, 2014, 11:08:12 PM »
When approaching the conversation with intellectual honesty, one cannot dispute that low ISO IQ is an important aspect of photography....especially for nature photographers.

AN important aspect of photography. Not THE important aspect of photography. You made my point for me.

Yes, it is important, I said as much. It is no where near as important as a growing number of photographers seem to think. It is AN important factor of IQ, but by no means the singular most important one. You are also missing part of my point. Canon has dynamic range, quite a lot of it in the grand scheme of things. The argument I was debating was that Canon so utterly and desperately needed more low ISO DR.

I levy the question again...how many of your shots are at ISO 100, and more importantly, of those ISO 100 shots, in how many did you desperately NEED to push shadows more than two or three stops?

Privatebydesign offered part of the answer:

That is an interesting question that illustrates why there are so many diverse opinions about th same piece of equipment, we all use them differently.

As for me I took a look, of my last 19,500 images, 9,000 were at 100iso, 7,500 at 200iso, 2,000 at 400 iso and 1,500 at 800 and other random intermediate iso stops.

I'd like higher low iso image quality. But I am not going to spit my dummy out waiting for it.

We know how many he takes at low ISO. He takes a lot, but that doesn't address my actual question, and the question that actually pertains to having more than 12 stops of DR at ISO 100: How many of those 9000 ISO 100 images needed to be pushed by four, five, or six stops? I would guess VERY FEW. Practically none, unless PBD shoots exceptionally difficult scenes with massive dynamic range on a regular basis/for a living. If that is the case, then hell, I highly recommend a D800 for him. In the grand scheme of things, though, I doubt most photographers even think about pushing shadows that much (or could even find a legitimate reason to.)

Personally, I've taken about 55,000 photos at ISO 800 - 3200. I've taken about 15,000 at ISO 400, and less than 10,000 at ISO 100 and 200. Of the ISO 100 photos, I have needed more dynamic range than my 7D offers in about 2000 shots, however I am usually short by maybe one stop (and that is more because of the 7D pixel size...if I had a 5D III, I would have what I need for pretty much everything I've shot before.) In the cases where slight vertical banding noise did show up in the shadows (maybe a couple hundred at most)...I used Topaz DeNoise 5, and was not only able to remove the banding, but I also gained more dynamic range (that's what happens when you reduce noise anyway...you gain DR, but Topaz has a feature that attempts to further recover DR that was lost to shadow noise due to a loss of tonal fidelity, which gains me even more.)

I use GND filtration for my landscape photography, so dynamic range is actually something I have a lot of control over in the field. I would actually greatly appreciate more native sensor DR, as it would reduce my need to use GND filters. It would also help me avoid that unsightly GND artifact where mountaintops end up dark or even black when you need to use more than two to three stops of filtration. That is the single situation where I think having more dynamic range would actually be the most important factor for IQ...ONE situation. I also suspect that tonemapping 14 stops into 8-10 stops without ending up with quirky shifts in contrast and color fidelity would still be very challenging, and I highly doubt I would stop using GND filters even if I had a D800. I still doubt I would push shadows around more than 2-3 stops....but it would be 2-3 stops along with fewer GND filters, which still makes the job easier in the end.

Again, I am not saying more DR is bad. Certainly not. I would just like to know, given how many people have started complaining about it since the release of the D800, how many of them actually have a real-world USE for more DR. Of PBD's 9000 ISO 100 shots...in how many did he actually push shadows around more than 2 stops? That's the real indicator of how much dynamic range we NEED, vs. how much more dynamic range we just WANT because, well, you know...the other guy has it.

Personally, I know the use cases I would love to have more DR for. It's a small fraction of my work. I also know that the kind of photography I do most is in line with the majority of DSLR users...action. Action photographers are what make the Canon Photography world go round. There are far more action shooters than any other kind of shooter, when you factor in sports, air shows, car races, bike races, watersports, wildlive, birds, and all those little children running around dimly lit houses. I want more DR, but I also know it isn't the singular most important IQ factor that so many forum talkers seem to think it is. If more people would honestly answer the question: "How often do you NEED to lift ISO 100 shadows more than 2-3 stops?" I think people might get a clearer picture of how important more dynamic range actually is to their work.

Pages: 1 ... 96 97 [98] 99 100 ... 263