fredmiranda.com Login
This topic is in the 2015 archive. You can read it and like its posts here. It is read only.

Archive 2015No DR improvement = no 5Ds for me

  
 
matthewsaville
Offline
Details
p.8 #1No DR improvement = no 5Ds for me
MintMar wrote:
But it will drive down 5D3 prices, so I will sooner or later (I expect later) open my wallet for 5D3. At least I will still get latest sensor tech.


You might be in for a long wait, I think for the time being this could actually drive the used price of a 5D3 UP. It's that big of an un-attractive camera to 90% of the market, IMO.

=Matt=
Feb 07, 2015 at 04:23 PM
RustyBug
Offline
Details
p.8 #2No DR improvement = no 5Ds for me
takowasa wrote:
Five pixels on FF would cover the same portion of the photo as two pixels on crop. So, per proportion of the photo, FF would have 1.6x more electronic noise and absorb 2.56x more light for the same exposure.


Still not getting it ... same pixels density = proportion is already calculated for, no additional adjustment is required.

Can you draw me a picture or refer a link ... still feel like I'm missing something here.
Feb 07, 2015 at 05:33 PM
kezeka
Offline
Details
p.8 #3No DR improvement = no 5Ds for me
RustyBug wrote:
kezeka wroteBeing satisfied with outdated technology may lead to a happier lifestyle but it has never advanced society in any meaningful way except to provide the old guard with a sense of security.


Then, you probably should go buy a new car every time they release a new feature each year.

BTW ... tell them to leave out the radio ... that's old technology.


Off topic response but I haven't used a radio tuner in a car in over 8 years. I just plug my phone into either USB or AUX port and use the amplifier to play music that I enjoy without having to listen to commercials or some other persons favorite songs.
Feb 07, 2015 at 05:39 PM
matthewsaville
Offline
Details
p.8 #4No DR improvement = no 5Ds for me
kezeka wrote:
RustyBug wrote
kezeka wroteBeing satisfied with outdated technology may lead to a happier lifestyle but it has never advanced society in any meaningful way except to provide the old guard with a sense of security.


Then, you probably should go buy a new car every time they release a new feature each year.

BTW ... tell them to leave out the radio ... that's old technology.
Off topic response but I haven't used a radio tuner in a car in over 8 years. I just plug my phone into either USB or AUX port and use the amplifier to play music that I enjoy without having to listen to commercials or some other persons favorite songs.


...And I suppose Facebook and Twitter are fine replacements for news sources too. ;-)
Feb 07, 2015 at 05:44 PM
Light_pilgrim
Offline
Details
p.8 #5No DR improvement = no 5Ds for me
I really do not think that it is a good camera body for ladscape photographers. Landscape is all about the DR and detail. Yes, it will give the resolution but not the DR. I recently talked to my friends from Iceland that are leading photo tours and it is not a camera for them. Imagine that you are out for 2 weeks in wilderness, you pack light. How many memory cards you need to have? What if you want to download and preview some images, what king of laptop you need to have with you? There is a practical aspect of being a landscape photographer too.

It is not a camera for landscape photographers - D810 is.

This is a camera for studio where you can control the DR and need to print big for billboards and magazines. It is to compete with MF.


garydavidjones wrote:
Chuck Westfall in the interview clearly is trying not to undercut the value of the 5D3.
He notes 5D3 has superior video and ISO range than the new 5Ds.

His statements plainly describe the new 5Ds as carefully designed for certain fields,
like landscape and architecture. Canon is continuing its niche design development
style.


Feb 08, 2015 at 12:20 AM
Shutterbug2006
Offline
Details
p.8 #6No DR improvement = no 5Ds for me
Perhaps Canon will come out with a $7000 1DXs and 1DXsR that has all the high megapixels of the new 5Ds, 4k video, and an extra 3 stops of dynamic range. Or maybe not.

The extra megapixels piques my interest, I like to print large at home. Whether or not my lenses are up to the task remains to be seen.

I'm surprised at the number of photographers who don't print. Yes, computers are the best thing since sliced cake - but I like to view photographs printed and hung on a wall - not on a computer screen.

My printer has a dynamic range significantly less than any of my cameras. But I get that if the camera had a higher dynamic range that I wouldn't have to work so hard to make certain photographs. Since it's been hard enough to squeeze a couple of extra stops of dynamic range out of the existing sensors - I can't imagine Canon will be releasing a camera with a lot more resolution AND more dynamic range any time soon.

But hey, I've been wrong once or twice before.
Feb 08, 2015 at 02:01 AM
takowasa
Offline
Details
p.8 #7No DR improvement = no 5Ds for me
takowasa wrote:
Five pixels on FF would cover the same portion of the photo as two pixels on crop. So, per proportion of the photo, FF would have 1.6x more electronic noise and absorb 2.56x more light for the same exposure.


RustyBug wrote:
Still not getting it ... same pixels density = proportion is already calculated for, no additional adjustment is required.

Can you draw me a picture or refer a link ... still feel like I'm missing something here.


I apologize for explaining poorly. Let me flesh things out a bit. The dynamic range is the range of light levels where detail is recorded over a particular portion of the photo. If we are computing the per-pixel DR, then, of course, a FF sensor and a crop sensor that use the same pixels will have the same DR. However, one pixel for the 50 MP 5Ds sensors covers a different area of the scene than one pixel from the 20 MP 7D2 sensor. So, to compare DR for the same portion of the scene, we would compare not per-pixel DR, but the DR over two pixels for the 7D2 and over 5 pixels for the 5Ds (or any other equal proportion).

If we are using the electronic noise of the sensor for the noise floor (that is, disregarding photon noise from the signal), then 5 pixels will have 1.6x more noise than 2 pixels (noise adds in quadrature), but they will collect 2.5x as much light for the same exposure (signal adds linearly). Thus, the DR will be 2.5/1.6 = 1.6x greater, which is 2/3 of a stop higher.

I hope I've explained this better (I'm afraid I don't have any pictures -- well, I do have pictures, just not those kind ).
Feb 08, 2015 at 02:25 AM
alundeb
Offline
Details
p.8 #8No DR improvement = no 5Ds for me
Light_pilgrim wrote:
I really do not think that it is a good camera body for ladscape photographers. Landscape is all about the DR and detail. Yes, it will give the resolution but not the DR. I recently talked to my friends from Iceland that are leading photo tours and it is not a camera for them. Imagine that you are out for 2 weeks in wilderness, you pack light. How many memory cards you need to have? What if you want to download and preview some images, what king of laptop you need to have with you? There is a practical aspect
...Show more →
garydavidjones wroteChuck Westfall in the interview clearly is trying not to undercut the value of the 5D3.
He notes 5D3 has superior video and ISO range than the new 5Ds.

His statements plainly describe the new 5Ds as carefully designed for certain fields,
like landscape and architecture. Canon is continuing its niche design development
style.


You arguments are not very consistent.

- If you need to pack light, the D810 has zero advantage over the 5Ds for landscape.

- Weight of memory cards is not a concern for landscape. The 5Ds takes SD cards, and a pile of cheap 128 GB cards weighs nothing.

- Dynamic range is not always more improtant than resolution for landscape, it depends.

- File size for preview is more a concern in studio and tethered shooting than for landscape. So in that respect the 5Ds is not better than the D810 for studio work.

For me the best option right now is to own both the A7r and 5Ds, for different situations. A D810 is an excellent camera too, but doesn't help with weight and doesn't complement the A7r as well.
Feb 08, 2015 at 04:45 AM
KKFung
Offline
Details
p.8 #9No DR improvement = no 5Ds for me
RustyBug wrote:
shirozina wroteI'm using every bit of the 14 stops of DR that I can get out of the A7r and there is no way I'll trade 2 stops for a few more pixels and I'm sure I'm not alone.
So, you've just made the exact point @ "fit for purpose" ... don't buy a Canon if that is what your needs are. Others have needs that their "fit for purpose" requirements rule out the A7R ... so they don't buy the A7R.

Unfortunately, we do not have any single camera that affords all things to all people ... such is life. We simply find those things that best fit our needs / goals / objectives / desires. Sometimes it isn't a matter of what we want that drives our direction, but a matter of what we DON'T want.

In the case
...Show more →

Agree, this is also exactly why I sold my A7r recently because it still have DR not enough for my landscape work, highlight still blow out or no details in shadow, it have 2 stops more DR than my canon but how I know the scene is just over 12 stop but under 14 stops? I still need to do bracket to make safe but I loss color fidelity which is more difficult for me to resolve in post when compare to my canon file.

Different kind of machine will fit for different kind of purpose. Just go for the one that best fit to you instead of yelling the manufacturer / designer which is no help at all.
Feb 08, 2015 at 04:50 AM
shirozina
Offline
Details
p.8 #10No DR improvement = no 5Ds for me
KKFung wrote:
RustyBug wrote
shirozina wroteI'm using every bit of the 14 stops of DR that I can get out of the A7r and there is no way I'll trade 2 stops for a few more pixels and I'm sure I'm not alone.
So, you've just made the exact point @ "fit for purpose" ... don't buy a Canon if that is what your needs are. Others have needs that their "fit for purpose" requirements rule out the A7R ... so they don't buy the A7R.

Unfortunately, we do not have any single camera that affords all things to all people ... such is life. We simply find those things that best fit our needs / goals / objectives / desires. Sometimes it isn't a matter of what we want that drives our direction, but a matter of what we DON'T want.

In the case of the A7R ... I don't want the ergonomics it provides with certain glass. I don't want to use adapters to use tilt shift glass. I don't want to reduce AF performance via adapters. Those are all "fit for purpose" issues. There are some aspects of the A7R or D810 that would be to my liking, but that there are other things that are not.

Your point that a few extra MP's are of little value to you ... I get that. Just because the extra MP's are available doesn't mean that you are compelled to adopt the advanced MP ... sounds like you've just exacted the very point @ contentedness (just in regard to a different criteria). Likewise, there are those who are not compelled to adopt the advanced DR that is available .... usually for their own "fit for purpose" and contentedness reasons that include criteria other than DR.

Some folks find DR to be their "Holy Grail" ... others, not so much.
Some folks find resolution to be their "Holy Grail" ... others, not so much.
Some folks find XYZ to be their "Holy Grail" ... others, not so much.

Point of constraints, utility, fit for purpose, contentedness, insatiable, fickle ... your call @ where you land with what works for you horses for courses.

Agree, this is also exactly why I sold my A7r recently because it still have DR not enough for my landscape work, highlight still blow out or no details in shadow, it have 2 stops more DR than my canon but how I know the scene is just over 12 stop but under 14 stops? I still need to do bracket to make safe but I loss color fidelity which is more difficult for me to resolve in post when compare to my canon file.

Different kind of machine will fit for different kind of purpose. Just go for the one
...Show more →
14 stops is indeed sometimes not enough. I always bracket in order to find the optimal exposure ( though not to combine images in HRDI) as the histograms and clipping indicators are only derived from jpeg previews and therefore are not indicative of the DR of the RAW file. In tricky light I set the bracketing ( single) to 0,+1,+2 and start the sequence where the highlights are just clipping in the preview. I'll then choose which exposure to process in LR, C1, DXO etc. It's rare that I have to combine more than 1 other exposure. Not sure what you are saying regarding color fidelity - I have found the colors on the A7r better than the Canon. Not only in their accuracy ( reds in particular) but in the subtle differences and hue seperations available.
Feb 08, 2015 at 05:45 AM
RustyBug
Offline
Details
p.8 #11No DR improvement = no 5Ds for me
takowasa wrote:

they will collect 2.5x as much light for the same exposure (signal adds linearly). Thus, the DR will be 2.5/1.6 = 1.6x greater, which is 2/3 of a stop higher.


Thanks for hanging in there with me.

Still struggling a bit the the premise of 2.5X more light @ = portion of image. Bear with me, as I think out loud.

If the composition / framing of the scene has been compensated for by using a proportionate focal length to the format to equate the two images, the FF counterpart will be using a longer glass than the crop counterpart. The physical / optical aperture for a given exposure (say f/2) of such longer glass will result in a larger AMOUNT of light in the larger format, but it will likewise be DISPERSED over the corresponding offset larger area (iaw with ISL), thus maintaining the same ratio of light being received on a per pixel basis (as previously concurred).

Meanwhile, (in a focal length equalized compensation) the 5 pixels will be covering a correspondingly smaller portion of the image as a byproduct of the focal length amplification to the projected image projected over the larger area of the FF format. As such, it would be an offset.

The premise that the 5 pixels in FF corresponds to the 2 pixels in 1.6x crop (@ portion of image) ... I think is a flawed premise, predicated upon an omission of the necessitated change in focal length for equalized overall image composition / framing and its corresponding distribution across the larger frame.

I'm certainly open to further explanation. My understanding of noise is spartan compared to other aspects of light, but for a moment there I thought we almost had a "free lunch" ... to which there never is such a thing.

Thoughts
Feb 08, 2015 at 07:01 AM
KKFung
Offline
Details
p.8 #12No DR improvement = no 5Ds for me
shirozina wrote:
KKFung wrote
RustyBug wrote
shirozina wroteI'm using every bit of the 14 stops of DR that I can get out of the A7r and there is no way I'll trade 2 stops for a few more pixels and I'm sure I'm not alone.
So, you've just made the exact point @ "fit for purpose" ... don't buy a Canon if that is what your needs are. Others have needs that their "fit for purpose" requirements rule out the A7R ... so they don't buy the A7R.

Unfortunately, we do not have any single camera that affords all things to all people ... such is life. We simply find those things that best fit our needs / goals / objectives / desires. Sometimes it isn't a matter of what we want that drives our direction, but a matter of what we DON'T want.

In the case of the A7R ... I don't want the ergonomics it provides with certain glass. I don't want to use adapters to use tilt shift glass. I don't want to reduce AF performance via adapters. Those are all "fit for purpose" issues. There are some aspects of the A7R or D810 that would be to my liking, but that there are other things that are not.

Your point that a few extra MP's are of little value to you ... I get that. Just because the extra MP's are available doesn't mean that you are compelled to adopt the advanced MP ... sounds like you've just exacted the very point @ contentedness (just in regard to a different criteria). Likewise, there are those who are not compelled to adopt the advanced DR that is available .... usually for their own "fit for purpose" and contentedness reasons that include criteria other than DR.

Some folks find DR to be their "Holy Grail" ... others, not so much.
Some folks find resolution to be their "Holy Grail" ... others, not so much.
Some folks find XYZ to be their "Holy Grail" ... others, not so much.

Point of constraints, utility, fit for purpose, contentedness, insatiable, fickle ... your call @ where you land with what works for you horses for courses.

Agree, this is also exactly why I sold my A7r recently because it still have DR not enough for my landscape work, highlight still blow out or no details in shadow, it have 2 stops more DR than my canon but how I know the scene is just over 12 stop but under 14 stops? I still need to do bracket to make safe but I loss color fidelity which is more difficult for me to resolve in post when compare to my canon file.

Different kind of machine will fit for different kind of purpose. Just go for the one that best fit to you instead of yelling the manufacturer / designer which is no help at all.
14 stops is indeed sometimes not enough. I always bracket in order to find the optimal exposure ( though not to combine images in HRDI) as the histograms and clipping indicators are only derived from jpeg previews and therefore are not indicative of the DR of the RAW file. In tricky light I set the bracketing ( single) to 0,+1,+2 and start the sequence where the highlights are just clipping in the preview. I'll then choose which exposure to process in LR, C1, DXO etc. It's rare that I have to combine more than 1 other exposure. Not sure what
...Show more →

Regarding the color fidelity, my experience is from my shooting during the sun rises / dawn in summer time in my country (Hong Kong), there is always have a light pink color in the sky but only my 1Dx can capture it in the picture but A7r (even I tried to enhance it from the post. It is not exist). I know this may not be an issue to other people, just a share here and that's why I agree Rustybug's statement.

Feb 08, 2015 at 07:59 AM
gdanmitchell
Offline
Details
p.8 #13No DR improvement = no 5Ds for me
Light_pilgrim wrote:
I really do not think that it is a good camera body for ladscape photographers.... Imagine that you are out for 2 weeks in wilderness, you pack light. How many memory cards you need to have? What if you want to download and preview some images, what king of laptop you need to have with you? There is a practical aspect of being a landscape photographer too.

It is not a camera for landscape photographers - D810 is.


I don't have to imagine, since I do these things. I do photograph in the backcountry for periods of up to 10-14 days at a time. (Do you?)

Weight and bulk are issues. When traveling on foot and without pack animal support, I limit myself to one full frame body (5DII at the moment), 2-3 L zooms, a slightly lighter tripod and head than I would otherwise use, and so forth. I do not carry a laptop and I have no interest in previewing images other than on the camera's LCD display. My total photography gear weight is likely between about 12 and 15 pounds. (Friends who shoot MF film in the backcountry and go out for many days at a time have been known to carry more than 35 pounds of camera gear in addition to their backpacking equipment.

I'm baffled by your concern about the number of memory cards. They are tiny and very light and capacious. I rarely fill more than one, but I carry perhaps three.

As to laptops, once per year I head out for 10-12 days with a group of Sierra landscape photographers on a packer-supported backcountry trip. Typically we hike, carrying camera gear plus water/snacks, and we meet the wranglers at our campsite where we set up a basecamp and then work outwards from their. On these pack animal supported trips we do bring more gear than on self-supported backpacking trips, and most do bring a laptop and even an external USB drive... along with solar charger gear to supply power.

Dan

(I have been backpacking and photographing for decades in the Sierra Nevada. If you are interested in approaches to photographing in the backcountry, I've written a bit about my approach here: http://www.gdanmitchell.com/2007/08/26/thoughts-on-my-summer-2007-backpacking-photography-kit)
Feb 08, 2015 at 10:44 AM
Snapgrapher
Offline
Details
p.8 #14No DR improvement = no 5Ds for me
Great! This just means the lineup to get one of these beauties will be 1 person shorter.
Feb 08, 2015 at 11:02 AM
takowasa
Offline
Details
p.8 #15No DR improvement = no 5Ds for me
takowasa wrote:

they will collect 2.5x as much light for the same exposure (signal adds linearly). Thus, the DR will be 2.5/1.6 = 1.6x greater, which is 2/3 of a stop higher.


RustyBug wrote:
Thanks for hanging in there with me.


Not at all -- I'm well paid.

Still struggling a bit the the premise of 2.5X more light @ = portion of image. Bear with me, as I think out loud.

If the composition / framing of the scene has been compensated for by using a proportionate focal length to the format to equate the two images, the FF counterpart will be using a longer glass than the crop counterpart. The physical / optical aperture for a given exposure (say f/2) of such longer glass will result in a larger AMOUNT of light in the larger format, but it will likewise be DISPERSED over the corresponding offset larger area
...Show more →

The ISL (Inverse Square Law) does not apply, nor does the area over which the light is spread matter. What matters is the total amount of light landing on the pixels that make up the portion of the photo that we are measuring the DR over and the electronic noise added by the sensor over the same area.

Meanwhile, (in a focal length equalized compensation) the 5 pixels will be covering a correspondingly smaller portion of the image as a byproduct of the focal length amplification to the projected image projected over the larger area of the FF format. As such, it would be an offset.

The premise that the 5 pixels in FF corresponds to the 2 pixels in 1.6x crop (@ portion of image) ... I think is a flawed premise, predicated upon an omission of the necessitated change in focal length for equalized overall image composition / framing and its corresponding distribution across the larger frame.


If we used, for example, 100mm on a 7D2 and 160mm on a 5Ds, and took a photo of the same scene from the same position, then two of the 7D2 pixels would span the same portion of the scene as five of the 5Ds pixels.

I'm certainly open to further explanation. My understanding of noise is spartan compared to other aspects of light, but for a moment there I thought we almost had a "free lunch" ... to which there never is such a thing.

Thoughts


You know that the reason FF has lower noise than crop is because more light falls on the sensor for a given exposure, right? The reason newer crop sensors have the same noise performance of older FF sensors is because the improved designs are able to collect more of that light falling on their sensors and add less corresponding noise.

Since dynamic range is generally expressed as the number of stops from the noise floor to the saturation limit over the area where we are measuring the DR, the dynamic range is simply another manifestation of the role that noise plays in photography.

I hope this clears things up a bit, if not completely.
Feb 08, 2015 at 11:26 AM
gdanmitchell
Offline
Details
p.8 #16No DR improvement = no 5Ds for me
takowasa wrote:
You know that the reason FF has lower noise than crop is because more light falls on the sensor for a given exposure, right? The reason newer crop sensors have the same noise performance of older FF sensors is because the improved designs are able to collect more of that light falling on their sensors and add less corresponding noise.


It is a bit more complex than that. If you compare FF and cropped sensors with the same number of photo sites, the FF sensor will generally have better noise performance because the individual photo sites are larger.

If you scale up a cropped sensor (such as the 7DII) to the full frame size (as seems to be the case with the 5DS), your individual photo sites are subject to exactly the same amount of noise on either sensor, all else being equal. However, in a photograph made from the higher MP sensor the noise will be less visible at a given print size since it has a smaller "grain."

It is a common mistake to attribute various things to "more light falling on a larger sensor," but that's not how it actually works.

Dan
Feb 08, 2015 at 11:30 AM
woos
Offline
Details
p.8 #17No DR improvement = no 5Ds for me
gdanmitchell wrote:
takowasa wroteYou know that the reason FF has lower noise than crop is because more light falls on the sensor for a given exposure, right? The reason newer crop sensors have the same noise performance of older FF sensors is because the improved designs are able to collect more of that light falling on their sensors and add less corresponding noise.
It is a bit more complex than that. If you compare FF and cropped sensors with the same number of photo sites, the FF sensor will generally have better noise performance because the individual photo sites are larger.


Nah it's because the full frame sensor is bigger. Pixel size doesn't really have anything to do with it. Well, it does, but generally smaller ones are better. The 6d sensor isn't better than the 5d3 because it has 2mp less, it's better because Canon did a better job with it.

If, in theory, you could make a full frame sensor that had pixels from say, the iphone 6s, it would be extremely good. Sure it might look bad at 100% but that doesn't mean anything because you'd be comparing different enlargement sizes.

Sure there are exceptions but they are because of large differences in tech (old ccd vs new cmos, etc). Full frame also has an inherent advantage in that it doesn't need to be enlarged as much, percentage wise, to get to a certain print size. This helps quite a bit.

There also may be exceptions that don't have anything to do with the pixels themselves. Here's an example: A 12mp sensor like the one in the a7s might have an easier time making really good (possibly even 2 element!) micro lenses. A high mp sensor might have more problems with getting really great micro lenses constructed for it. Just an example.
Feb 08, 2015 at 11:50 AM
takowasa
Offline
Details
p.8 #18No DR improvement = no 5Ds for me
takowasa wrote:
You know that the reason FF has lower noise than crop is because more light falls on the sensor for a given exposure, right? The reason newer crop sensors have the same noise performance of older FF sensors is because the improved designs are able to collect more of that light falling on their sensors and add less corresponding noise.


gdanmitchell wrote:
It is a bit more complex than that. If you compare FF and cropped sensors with the same number of photo sites, the FF sensor will generally have better noise performance because the individual photo sites are larger.


That's a common myth. It's not because the photosites are larger, but because they gather more light for a given area of the photo and exposure. The only reason that the pixel size matters is that a greater number of smaller pixels generally results in more additional electronic noise than a smaller number of larger pixels for a given tech level. However, the electronic noise is only significant for portions of the photo made from very little light, such as deep shadows or high ISO.

If you scale up a cropped sensor (such as the 7DII) to the full frame size (as seems to be the case with the 5DS), your individual photo sites are subject to exactly the same amount of noise on either sensor, all else being equal. However, in a photograph made from the higher MP sensor the noise will be less visible at a given print size since it has a smaller "grain."

The per-pixel noise will be the same, as you say. The per-image noise will be less, however, because more light made up the photo.

With regards to the "grain", yes, there is the issue that the frequency of the FF sensor noise would be higher since it would be made with more pixels, so we are in a position of comparing noise at different frequencies. Here's a good visual description of what that means:

http://www.dpreview.com/forums/post/37549285

It is a common mistake to attribute various things to "more light falling on a larger sensor," but that's not how it actually works.

Dan


I'm afraid it's not a mistake at all. If two sensors have around the same quantum efficiency and electronic noise, then the differences in the quantity of noise in the photos they produce when displayed at the same size will be entirely due to the amount of light that fell on the sensors during the exposure, regardless of how many pixels that light fell into (keeping in mind that more smaller pixels generally have more electronic noise than fewer larger pixels, which presents itself in the portions of the photo made from very little light, such as heavy shadow pushing at base ISO or high ISOs).

Another way to understand it is take a photo of a scene at, say, 1/100 ISO 100 and 1/1600 ISO 1600 (or, if you prefer, 1/1600 ISO 100 and push the photo four stops) with the same camera. Why is the ISO 1600 photo more noisy? Simple: it was made with 1/16 the amount of light as the ISO 100 photo.
Feb 08, 2015 at 11:51 AM
jcolwell
Offline
Details
p.8 #19No DR improvement = no 5Ds for me
takowasa, I really think you're on the wrong track.
Feb 08, 2015 at 12:14 PM
RustyBug
Offline
Details
p.8 #20No DR improvement = no 5Ds for me
Okay, so were mixing apples & refrigerators a bit now @ chicken & egg.

+1 @ a given MP count, the FF will have larger sites than the crop.

+1 @ the ISO 1600 vs. 100 @ SAME CAMERA = same pixel density & less light = more noise. Most folks can go along with that just fine, I'd think.

Repeat that with a crop camera of equal pixels size/density and you again get the diff @ 1600 vs. 100 ... again, no surprise here either. But, while that correlation holds true, it isn't really the question.

What will (@ 100%, i.e. we aren't @ enlargement % to print size yet) the noise of crop @ 100 vs. the noise @ 100 of the FF will be for same pixel density (bearing in mind these are the same pixel size / density, not same quantity of MP)?

I agree that if the pixel count were the same, that the 5 pixels of the FF would indeed be larger. But, that isn't the case with regard to a same density comparison.

Conventional logic says similar density = similar noise (ceteris paribus). I can't find any theory that I'm aware of that fully refutes this.


Edited on Feb 08, 2015 at 12:20 PM · View previous versions
Feb 08, 2015 at 12:14 PM
FM ForumsCanon ForumJoin Upload & Sell
  
 

You are not logged inLog inRegister