Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
03-17-2016, 12:49 PM   #226
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 40,442
QuoteOriginally posted by gmans Quote
Isn't it ironic spend years arguing for FF and now it's becoming reality people start arguing that it's not required. I think half you mob on here would argue black and blue. But carry on anyhows.
Hey, I personally haven't changed a bit. What has changed is, I can buy an FF as easily as would a new lens. But I would never argue it's required. It's not.

But if I'm not wrong, where derekite lives light is way more of an issue than it is around where I live, I can count on having some direct sun, even on some rainy days. Where as he has to have a system that deals with the light he has, I just wait for an opportunity with good light. It's a different level of need.

03-17-2016, 12:53 PM   #227
Veteran Member




Join Date: Aug 2011
Location: Austin, Texas
Photos: Albums
Posts: 328
It is just another format with its own advantages and problems. Personally, I have always taken my best photos on tripod with 4x5 and 8x10 cameras. Slowing down does quite a lot to help you see better. Ansel Adams, were he alive today, would be all over digital as would HCB, Capa, DDD, etc.
03-17-2016, 01:06 PM   #228
Pentaxian




Join Date: Sep 2011
Location: Nelson B.C.
Posts: 3,782
QuoteOriginally posted by normhead Quote
Hey, I personally haven't changed a bit. What has changed is, I can buy an FF as easily as would a new lens. But I would never argue it's required. It's not.

But if I'm not wrong, where derekite lives light is way more of an issue than it is around where I live, I can count on having some direct sun, even on some rainy days. Where as he has to have a system that deals with the light he has, I just wait for an opportunity with good light. It's a different level of need.
Absolutely. On my way to work this AM there was an eagle on a post eating a fish. 1/320, f4.5 handheld, ISO somewhere north of 1600. We will see if I got anything tonight, but what I would do for a stop of light.

---------- Post added 03-17-16 at 01:10 PM ----------

QuoteOriginally posted by jeverettfine Quote
It is just another format with its own advantages and problems. Personally, I have always taken my best photos on tripod with 4x5 and 8x10 cameras. Slowing down does quite a lot to help you see better. Ansel Adams, were he alive today, would be all over digital as would HCB, Capa, DDD, etc.
What equipment we have determines in many ways the shots we take. My experience with apsc and landscape is disappointment with the loss of detail to diffraction. Forget hair breadth dof, I want f32.
03-17-2016, 01:15 PM   #229
osv
Veteran Member




Join Date: Mar 2007
Location: So Cal
Photos: Gallery
Posts: 2,080
QuoteOriginally posted by derekkite Quote
OK, so where do the curves cross? At 100% the d810 is almost a stop better. But resolution matters, so the perceived noise of a higher resolution yet noisier sensor is better normalized, at some point away from 100%. Where?
the curves never cross, when viewing with the print tab, which is the only way you can compare different sensor sizes: Pentax K-3 II vs Nikon D800

the d800 is superior all the way across the range.

QuoteOriginally posted by derekkite Quote
I'm not talking about some theoretical possibility, this is where I live in my photography. If an apsc body a stop better noise was announced I'd preorder two... The cropped sensor is not satisfactory.
the pq differences between aps-c/m4-3 and ff are rather large, they will never be comparable.

but there are also differences between ff sensors, here it is in a real-world example, a7r vs. canon: Sony A7R teams up with Canon glass

03-17-2016, 01:34 PM   #230
Veteran Member
gmans's Avatar

Join Date: Dec 2012
Location: Hunter Valley,NSW, Australia
Photos: Gallery | Albums
Posts: 2,466
QuoteOriginally posted by normhead Quote
Hey, I personally haven't changed a bit. What has changed is, I can buy an FF as easily as would a new lens. But I would never argue it's required. It's not.

But if I'm not wrong, where derekite lives light is way more of an issue than it is around where I live, I can count on having some direct sun, even on some rainy days. Where as he has to have a system that deals with the light he has, I just wait for an opportunity with good light. It's a different level of need.
Was not mentioning any names, just thought it's ironic. Beg to differ it was needed though by a large audience and Ricoh.
03-17-2016, 02:39 PM   #231
Veteran Member




Join Date: Aug 2011
Location: Austin, Texas
Photos: Albums
Posts: 328
QuoteOriginally posted by derekkite Quote
Absolutely. On my way to work this AM there was an eagle on a post eating a fish. 1/320, f4.5 handheld, ISO somewhere north of 1600. We will see if I got anything tonight, but what I would do for a stop of light.

---------- Post added 03-17-16 at 01:10 PM ----------



What equipment we have determines in many ways the shots we take. My experience with apsc and landscape is disappointment with the loss of detail to diffraction. Forget hair breadth dof, I want f32.
Yes, quite true. It is what makes me think hard about eventually getting a K-1 or other FF digital. I have a friend who keeps raving about his Sony equipment. I have to admit his images look pretty nice.
03-17-2016, 02:40 PM   #232
Veteran Member




Join Date: Dec 2007
Photos: Gallery
Posts: 8,237
QuoteOriginally posted by jeffshaddix Quote
Hmm, regarding DOF differences, the main advantage I see could be stated differently: I can stop down my lens another stop with the same DOF. There's often a big difference in sharpness and aberrations between f/1.4 and f/2, or f/2.8 and f/4, especially on normal primes. What I could only do with the lens wide open on APS-C, I will now be able to do on FF with the lens performing better stopped down.

I look forward to seeing how my A50/1.2 performs at f/1.8
Exactly, and I've written about that ^ plenty. When you match the DOF, you're often seeing a bit more sharpness and contrast because lens A at f/4 may have more sharpness than lens B at f/2.5 (but certainly not always)

03-18-2016, 01:09 PM   #233
Veteran Member




Join Date: Dec 2007
Photos: Gallery
Posts: 8,237
QuoteQuote:
@Dosdan actually does not understand it,
Dosdan does understand it, fully. Since you think both Falk Lumo and Dosdan are frauds, it doesn't give me much hope that you'll ever listen to me.



QuoteOriginally posted by clackers:
I suggest for a second (and final) time that you go to the DXO Mark site and see for yourself - otherwise you can't explain why the Pentax K-50 does better in the Normalization step than the Sony A7S, and the APS-C Pentax K-3 does equally well as the FF Nikon D750.
The screen tab (100% view) shows that - the print tab shows something quite different.

Just want it to be noted once again for the record that your 'screen tab depicts real life' logic is leading you to believe that the K-50 beats the FF A7S, and that the K3 equals the Nikon D750. You wrote it right there.

Wow.

How many people are on board with that? Please, is there a slient contingent out there I'm not aware of that is thinking this way?

.

Last edited by jsherman999; 03-19-2016 at 10:20 PM.
03-18-2016, 06:31 PM   #234
Veteran Member




Join Date: Jan 2012
Photos: Gallery
Posts: 604
QuoteOriginally posted by jsherman999 Quote
Yes, yes it is. The images are all shot at the same exposure, which results in larger sensors gathering more total light for that same exposure. And the difference in total light in the image gives it it's noise advantage, which is reflected in the print tab.

I suspect you think the print tab normalizes for MP differences only - and not for total light - but it does. Same exposure = more total light for the larger sensor. (Why don't you ask a DXO rep directly if that's true? )





Dosdan does understand it, fully. Since you think both Falk Lumo and Dosdan are frauds, it doesn't give me much hope that you'll ever listen to me.

Let me ask, is there anyone you would listen to, at the risk of 'appealing to authority'?

Eric Fossum asked permission to use this article (Link to DXO article) as reading material for his students in a class he teaches - in the article, it mentions about how sensor size & total light is the determining factor in why larger sensors get their advantage over smaller of the same gen, not pixel size. (It also quotes material and a table from Falk's article... ooops, fraud again )

I also just asked in the DXOMark forums - really on your behalf, no names though - if the 'print' tab normalizes for sensor size (as well as resolution) because the exposure is the same and thus total light becomes a factor. Go follow that discussion if you want, let me know if you need a link.



The screen tab (100% view) shows that - the print tab shows something quite different.

Just want it to be noted once again for the record that your 'screen tab depicts real life' logic is leading you to believe that the K-50 beats the FF A7S, and that the K3 equals the Nikon D750. You wrote it right there.

Wow.

How many people are on board with that? Please, is there a slient contingent out there I'm not aware of that is thinking this way?

.

If you compare the dxomark snr measurements to film then we have:

SCREEN SNR = film grain (dependent on rated ISO)
PRINT SNR = film enlargement (interpolated from SCREEN snr)
NATIVE ISO = film speed as indicated in the canister
ISO = how much you push in the lab

So obviously in the case of K5 vs D800 and E-M5 vs 5DS the SCREEN snr is telling us that these sensors are essentially the same film emulsion. Meaning, they have exactly the same performance if exposed in the same way; same shutter speed, same f-stop. Size does not matter. A 135 film has exactly the same grain (noise) as the largest format of the same emulsion.

When you enlarge a film during printing the results will vary depending on enlargement size. An 8x10 sheet printed at 8x10 will obviously look better than an enlarged 135 film. Does this indicate film performance? It doesnt! You can enlarge 135 film to billboard size and you will never notice any quality degradation when you view it from 100 meters away. PRINT snr is relative to print size AND viewing distance and is NEVER an indication of sensor performance. FACT. MATH. SCIENCE.

Further to that, PRINT snr has NO, ZERO, NADA, ZILCH, NAUGHT, relationship to sensor size. PRINT snr is only dependent on
1) SCREEN snr
2) number of megapixels

An 50Mp E-M5, if such a beast exists, will match the performance of a full frame 5DS at any print size. Sensor size does not matter.

Therefore, equivalence is wrong.
03-18-2016, 07:45 PM - 3 Likes   #235
Veteran Member




Join Date: Jan 2016
Location: Roi-et, Thailand
Posts: 773
36mp FF will make little difference to those who only view on screen (even the best monitors like the iMac 5k can only resolve 15mp of detail) and likely not that much difference to those who print (very big prints is where FF 36mp works). It'll be frustrating for those coming off APS-C regards depth of field, it becomes razor thin to the point peoples eyes might be in focus but their noses are a blur of bokeh (this still happens to me shooting the Sony Zeiss 35/1.4 ZA wide open and stopped to f/2 even, on the A7r and it is frustrating). Focus may well be slower than APS-C (have to wait and see) and certainly frames per second as we already know.

If it's the same sensor as in the A7r and Pentax haven't managed to work their magic, expect colour noise creeping in by ISO1600 and unworkable by 6400 (like seriously purple and red blotchy). Not a good sensor for low light in my experience. IBIS should help here however allowing you to avoid such high ISO's.

You'll notice the size and weight difference, even though on paper it doesn't look that much. Apart from the lovely FA Ltds, that new umpteen element zoom glass is going to be big and heavy and that becomes a burden, takes the joy out of shooting for pleasure. It's a shame there's not f/4 zoom glass like Sony/Nikon/Canon, because in my experience the FE 4/70-200 G OSS is just the right size and weight (and it's smokin' hot on the A6000). I also say 70-200 is a wonderful zoom range regardless if on FF or crop. I wouldn't be without it now.

Wide lenses become 1.5x wider and long lenses 1.5x shorter, keep hold of your APS-C bodies if you shoot long. The 4/70-200 G OSS never gets mounted to the A7r, only the A6000. Long glass and APS-C makes sense, economically and in pretty much every other respect unless you're a pro sports shooter/extremely serious birdy snapper. IMO.

Lenses become much more expensive (full frame glass not cheap), unless you're using old manual primes, which is recommended (good read and I agree when I compare the PK FA's to the Sony Zeiss ZA's) . . .

The Problem with Modern Optics ? YANNICK KHONG

Just my ramblings having had Pentax crop and Sony full frame/crop. I'll confess I wish I hadn't bothered going down the Sony FF (A7r) route, painfully expensive and ghastly experience, but the A6000 has been a wee marvel on that zoom. Love the tracking AF-C, AF points coverage, eye detect AF and high burst rate. It's great fun, which is what shooting for pleasure is all about.

I wouldn't write off APS-C as a suitable format for the majority of enthusiast shooters.
03-18-2016, 09:13 PM   #236
Veteran Member
cali92rs's Avatar

Join Date: May 2011
Location: Long Beach, CA
Posts: 3,354
QuoteOriginally posted by dtmateojr Quote
If you compare the dxomark snr measurements to film then we have:

SCREEN SNR = film grain (dependent on rated ISO)
PRINT SNR = film enlargement (interpolated from SCREEN snr)
NATIVE ISO = film speed as indicated in the canister
ISO = how much you push in the lab

So obviously in the case of K5 vs D800 and E-M5 vs 5DS the SCREEN snr is telling us that these sensors are essentially the same film emulsion. Meaning, they have exactly the same performance if exposed in the same way; same shutter speed, same f-stop. Size does not matter. A 135 film has exactly the same grain (noise) as the largest format of the same emulsion.

When you enlarge a film during printing the results will vary depending on enlargement size. An 8x10 sheet printed at 8x10 will obviously look better than an enlarged 135 film. Does this indicate film performance? It doesnt! You can enlarge 135 film to billboard size and you will never notice any quality degradation when you view it from 100 meters away. PRINT snr is relative to print size AND viewing distance and is NEVER an indication of sensor performance. FACT. MATH. SCIENCE.

Further to that, PRINT snr has NO, ZERO, NADA, ZILCH, NAUGHT, relationship to sensor size. PRINT snr is only dependent on
1) SCREEN snr
2) number of megapixels

An 50Mp E-M5, if such a beast exists, will match the performance of a full frame 5DS at any print size. Sensor size does not matter.

Therefore, equivalence is wrong.
Thanks for this info!
I hope you also post on the Medium Format board and inform those folks how it is a complete waste of money to buy a 645Z. They will appreciate your knowledge.
03-18-2016, 09:34 PM - 1 Like   #237
Forum Member




Join Date: Jul 2012
Posts: 60
QuoteOriginally posted by dtmateojr Quote
If you compare the dxomark snr measurements to film then we have:

SCREEN SNR = film grain (dependent on rated ISO)
PRINT SNR = film enlargement (interpolated from SCREEN snr)
NATIVE ISO = film speed as indicated in the canister
ISO = how much you push in the lab

So obviously in the case of K5 vs D800 and E-M5 vs 5DS the SCREEN snr is telling us that these sensors are essentially the same film emulsion. Meaning, they have exactly the same performance if exposed in the same way; same shutter speed, same f-stop. Size does not matter. A 135 film has exactly the same grain (noise) as the largest format of the same emulsion.

When you enlarge a film during printing the results will vary depending on enlargement size. An 8x10 sheet printed at 8x10 will obviously look better than an enlarged 135 film. Does this indicate film performance? It doesnt! You can enlarge 135 film to billboard size and you will never notice any quality degradation when you view it from 100 meters away. PRINT snr is relative to print size AND viewing distance and is NEVER an indication of sensor performance. FACT. MATH. SCIENCE.

Further to that, PRINT snr has NO, ZERO, NADA, ZILCH, NAUGHT, relationship to sensor size. PRINT snr is only dependent on
1) SCREEN snr
2) number of megapixels

An 50Mp E-M5, if such a beast exists, will match the performance of a full frame 5DS at any print size. Sensor size does not matter.

Therefore, equivalence is wrong.
I'm not sure where you got some of these ideas, and feel the need to assert them so strongly, when there doesn't seem to be any evidence for them. The technical explanation behind the normalization scaling performed by DxOMark to get the Print scores from the Screen scores quite clearly does not take viewing distance into account at all, it simply normalizes to a common output resolution (N_0 = 8 MP). Detailed computation of DxOMark Sensor normalization - DxOMark

Furthermore, while the normalization 'process' itself does not take into account sensor size directly in that step, that does NOT mean that the normalized or Print scores will not yield higher scores for larger sensors than smaller ones (of a common sensor tech/age). To see this, first consider the initial Screen scores which reflect the per-pixel noise levels. Per-pixel noise will be lower for larger pixels than for smaller ones of the same tech age. Now consider some examples of APS-C vs FF sensor combinations that we would like to compare.

1. 16 MP APS-C vs 16 MP FF
2. 24 MP APS-C vs 16 MP FF
3. 16 MP APS-C vs 36 MP FF

In case 1, the FF sensor will yield a higher Screen SNR score thanks to it's larger pixel pitch. Then the normalization step will treat both sensors equivalently, since they have the same number of pixels, and thus the normalized Print scores will show the same trend as seen with the Screen scores -- the FF will still be better.

In case 2, the FF will yield a relatively even higher Screen SNR score than the APS-C since it's pixels are even more relatively large compared to the 24 MP APS-C sensor. This time the normalization step will even the playing field a bit, however, as the larger number of pixels for the APS-C sensor will bring it's Print score more in line with the FF's Print score. (In fact the 24 MP APS-C Print score should be quite similar to the 16 MP APS-C score for the same sensor tech/age.)

In case 3, the FF will yield about the same Screen SNR score as the APS-C since they have about the same pixel size. However, this time the normalization step will reveal the big advantage of the larger sensor, as the much larger number of pixels will result in higher Print score for the FF than the APS-C.

So from these examples we can see why sensor size will lead to improved ie. reduced image noise when using an output image at the same size (the normalization step is required in order to do this comparison properly), as indicated by its higher Print score.

We can also see how (at least with current sensor tech -- it didn't used to strictly be the case) for a given sensor size, the number of pixels does not really matter for output image noise, as the effect of noiser per-pixel levels for a higher resolution sensor will be negated during the normalization step in which the higher resolution sensor has it's noise 'averaged out' and ends up with about the same Print score as the lower resolution sensor. So, the only down side to a higher resolution sensor is typically cost, and intermediate image file sizes, while the up side is higher resolution in some situations (at least at low ISO's).

These are general rules of thumb and of course if sensors have significantly different tech or are customized in significant ways relative to each other, performance may deviate somewhat from these guidelines.

I hope this is relatively clear and helpful.
03-18-2016, 09:49 PM   #238
Pentaxian




Join Date: May 2009
Location: Somewhere over the rainbow
Photos: Gallery
Posts: 1,528
QuoteOriginally posted by dtmateojr Quote
If you compare the dxomark snr measurements to film then we have:
SCREEN SNR = film grain (dependent on rated ISO)
PRINT SNR = film enlargement (interpolated from SCREEN snr)
NATIVE ISO = film speed as indicated in the canister

Please tell me the sizes of the following images 36mp FF , 36mp cropped, 16mp FF and 16mp cropped. Figure this out and you will know how much enlargement (interpolated) is need for them all to be viewed 6inch X 4inch print.

Or when viewed on my monitor 6inch X 4inch

Last edited by Ian Stuart Forsyth; 03-18-2016 at 11:20 PM.
03-18-2016, 11:07 PM   #239
Pentaxian




Join Date: Sep 2011
Location: Nelson B.C.
Posts: 3,782
QuoteOriginally posted by crussellsprout Quote
I'm not sure where you got some of these ideas, and feel the need to assert them so strongly, when there doesn't seem to be any evidence for them. The technical explanation behind the normalization scaling performed by DxOMark to get the Print scores from the Screen scores quite clearly does not take viewing distance into account at all, it simply normalizes to a common output resolution (N_0 = 8 MP). Detailed computation of DxOMark Sensor normalization - DxOMark

Furthermore, while the normalization 'process' itself does not take into account sensor size directly in that step, that does NOT mean that the normalized or Print scores will not yield higher scores for larger sensors than smaller ones (of a common sensor tech/age). To see this, first consider the initial Screen scores which reflect the per-pixel noise levels. Per-pixel noise will be lower for larger pixels than for smaller ones of the same tech age. Now consider some examples of APS-C vs FF sensor combinations that we would like to compare.

1. 16 MP APS-C vs 16 MP FF
2. 24 MP APS-C vs 16 MP FF
3. 16 MP APS-C vs 36 MP FF

In case 1, the FF sensor will yield a higher Screen SNR score thanks to it's larger pixel pitch. Then the normalization step will treat both sensors equivalently, since they have the same number of pixels, and thus the normalized Print scores will show the same trend as seen with the Screen scores -- the FF will still be better.

In case 2, the FF will yield a relatively even higher Screen SNR score than the APS-C since it's pixels are even more relatively large compared to the 24 MP APS-C sensor. This time the normalization step will even the playing field a bit, however, as the larger number of pixels for the APS-C sensor will bring it's Print score more in line with the FF's Print score. (In fact the 24 MP APS-C Print score should be quite similar to the 16 MP APS-C score for the same sensor tech/age.)

In case 3, the FF will yield about the same Screen SNR score as the APS-C since they have about the same pixel size. However, this time the normalization step will reveal the big advantage of the larger sensor, as the much larger number of pixels will result in higher Print score for the FF than the APS-C.

So from these examples we can see why sensor size will lead to improved ie. reduced image noise when using an output image at the same size (the normalization step is required in order to do this comparison properly), as indicated by its higher Print score.

We can also see how (at least with current sensor tech -- it didn't used to strictly be the case) for a given sensor size, the number of pixels does not really matter for output image noise, as the effect of noiser per-pixel levels for a higher resolution sensor will be negated during the normalization step in which the higher resolution sensor has it's noise 'averaged out' and ends up with about the same Print score as the lower resolution sensor. So, the only down side to a higher resolution sensor is typically cost, and intermediate image file sizes, while the up side is higher resolution in some situations (at least at low ISO's).

These are general rules of thumb and of course if sensors have significantly different tech or are customized in significant ways relative to each other, performance may deviate somewhat from these guidelines.

I hope this is relatively clear and helpful.
Yes, very well explained. And we have to remember that these numbers are acquired by taking a shot of an 18% grey scene. There is another chart that gives the numbers for other gradients of grey.

I think it is clear that this isn't simply an upgrade to what we have, but a different beast altogether. How we go about framing a shot depends on what we shoot. Any time I've acquired something I've had to learn where it opens possibilities, and closes others.

I suspect that the larger the sensor the more demanding in technique. Ultimately that may be the most revealing curve.
03-19-2016, 02:04 AM   #240
Veteran Member




Join Date: Jan 2012
Photos: Gallery
Posts: 604
QuoteOriginally posted by crussellsprout Quote
I'm not sure where you got some of these ideas, and feel the need to assert them so strongly, when there doesn't seem to be any evidence for them. The technical explanation behind the normalization scaling performed by DxOMark to get the Print scores from the Screen scores quite clearly does not take viewing distance into account at all, it simply normalizes to a common output resolution (N_0 = 8 MP). Detailed computation of DxOMark Sensor normalization - DxOMark

Furthermore, while the normalization 'process' itself does not take into account sensor size directly in that step,
We could have ended here. Let me repeat what you just said: DOES NOT TAKE INTO ACCOUNT SENSOR SIZE. You are correct. Print SNR has got nothing to do with sensor size or for that matter TOTAL LIGHT.

But let's continue...

QuoteQuote:
that does NOT mean that the normalized or Print scores will not yield higher scores for larger sensors than smaller ones (of a common sensor tech/age). To see this, first consider the initial Screen scores which reflect the per-pixel noise levels. Per-pixel noise will be lower for larger pixels than for smaller ones of the same tech age. Now consider some examples of APS-C vs FF sensor combinations that we would like to compare.

1. 16 MP APS-C vs 16 MP FF
2. 24 MP APS-C vs 16 MP FF
3. 16 MP APS-C vs 36 MP FF

In case 1, the FF sensor will yield a higher Screen SNR score thanks to it's larger pixel pitch.
Correct. The larger pixel results in better SNR. So obviously you are now comparing different sensors, yeah? Again you could have ended your argument here.

QuoteQuote:
Then the normalization step will treat both sensors equivalently, since they have the same number of pixels, and thus the normalized Print scores will show the same trend as seen with the Screen scores -- the FF will still be better.
Of course because if you read my previous post, print SNR is affected by SCREEN snr.

QuoteQuote:
In case 2, the FF will yield a relatively even higher Screen SNR score than the APS-C since it's pixels are even more relatively large compared to the 24 MP APS-C sensor. This time the normalization step will even the playing field a bit, however, as the larger number of pixels for the APS-C sensor will bring it's Print score more in line with the FF's Print score. (In fact the 24 MP APS-C Print score should be quite similar to the 16 MP APS-C score for the same sensor tech/age.)
So obviously, PRINT snr is not a measure of sensor size performance, right? Or light gathering. Otherwise there's no way a APS-C could match FF.

QuoteQuote:
In case 3, the FF will yield about the same Screen SNR score as the APS-C since they have about the same pixel size. However, this time the normalization step will reveal the big advantage of the larger sensor, as the much larger number of pixels will result in higher Print score for the FF than the APS-C.
You just repeated what I said. PRINT snr has got nothing to do with sensor size. It's all about image size. If the APS-C was 36Mp it will match the FF in print snr.

QuoteQuote:
So from these examples we can see why sensor size will lead to improved ie. reduced image noise when using an output image at the same size (the normalization step is required in order to do this comparison properly), as indicated by its higher Print score.
Correct but not because it is able to gather more light but simply because it gives a larger image in terms of megapixels.

QuoteQuote:
We can also see how (at least with current sensor tech -- it didn't used to strictly be the case) for a given sensor size, the number of pixels does not really matter for output image noise, as the effect of noiser per-pixel levels for a higher resolution sensor will be negated during the normalization step in which the higher resolution sensor has it's noise 'averaged out' and ends up with about the same Print score as the lower resolution sensor. So, the only down side to a higher resolution sensor is typically cost, and intermediate image file sizes, while the up side is higher resolution in some situations (at least at low ISO's).

These are general rules of thumb and of course if sensors have significantly different tech or are customized in significant ways relative to each other, performance may deviate somewhat from these guidelines.

I hope this is relatively clear and helpful.

A very long winded response that repeats my last post:

PRINT SNR is affected by
1) SCREEN SNR
2) image size

Sensor size is nowhere in the equation.

If a smaller sensor is more efficient than the larger sensor then the higher SCREEN snr may result in a better print snr. If the smaller sensor has the same SCREEN snr as the larger sensor but has more megapixels it will also result in a higher print snr.

It just so happened that current sensor technology does not allow this anymore because FF sensors are really just enlarged APS-C as in the K5 and the D800.

Back when digital imaging was new and APS-C was only 4Mp I bet that the E-M5 will run circles around them for the fact that the latter has 4x the megapixels and the SCREEN snr will probably be the same. Even if the APS-C is slightly better in SNR, the sheer size of the m43 image will kill it.

So bottomline, sensor size has got nothing to do with print snr.

---------- Post added 03-19-16 at 19:08 ----------

QuoteOriginally posted by Ian Stuart Forsyth Quote
Please tell me the sizes of the following images 36mp FF , 36mp cropped, 16mp FF and 16mp cropped. Figure this out and you will know how much enlargement (interpolated) is need for them all to be viewed 6inch X 4inch print.

Or when viewed on my monitor 6inch X 4inch

Are you insulting me? Which is heavier a kilo of nails or a kilo of cotton?
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
24x36mm, 300mm, advantages, app, aps-c, body, camera, cost, crop, dslr, f/2.8, ff, field, focus, frame, full-frame, k-1, k-3, k3, lens, ovf, pentax, pentax k-3, quality, sensor, sigma, view
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Crop Sensors vs Full Frame :: Crop Or Crap? i83N Photographic Industry and Professionals 44 07-30-2014 06:00 AM
No Pentax Full Frame DSLR, What to Do Now Then? RiceHigh Pentax DSLR Discussion 95 07-19-2014 10:57 AM
The Full Frame NEX is going to be announced in October ! jogiba Canon, Nikon, Sony, and Other Camera Brands 30 09-15-2013 03:23 PM
So.... Pentax Full Frame is a sure thing? LFLee Photographic Industry and Professionals 25 11-02-2012 12:55 PM
So It's True....a Full-Frame 'Professional' Samsung Dslr Coming !!!! Confused Pentax DSLR Discussion 9 05-02-2008 04:12 PM



All times are GMT -7. The time now is 11:18 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top