[jsherman999]

Doing the same thing on digital with switching cameras for the one and using FF in crop mode for the other there would be no difference in DoF as enlargerment is not a factor, the image would be the same size depending only on sensor pixel density.

Yes, but If I understand you, that's not what was being proposed by the OP - the OP was asking about using a 'crop lens' on FF, and the effect on noise. We're inferring from that question that the OP intends to shoot the combo in 'crop mode' and has no intention of viewing/printing those results any smaller than before.

If that's the case, there will be a difference in noise and in DOF in the resulting image. You can't crop (and enlarge that crop) without consequences to IQ.

We have a FF camera coming. I'm hoping that there are not Pentaxians out there who think that shooting their DA limiteds in crop mode will give them any FF advantage - it won't. aps-c lens shot in crop mode and displayed at the same size == an aps-c camera, basically.

.

.

---

Last edited by jsherman999; 03-17-2016 at 03:12 PM.

[amoringello]

Cherry picking words out of context just makes most of that white noise.
The rest is just arguing with clearly made statements. More white noise.
tl;dr

If there is a true desire to discuss where we may be communicating differently feel free to private message or start a new thread.
I'm sure you know what you're talking about. And I know what I'm talking about, but something is missing.

Of course if someone can prove how I am wrong, I would love to have a coherent and accurate reasoning why.

Meanwhile, this thread has dissolved into trolling and is not worth perpetuating.

---------- Post added 03-17-16 at 06:12 PM ----------

Yes, but If I understand you, that's not what was being proposed by the OP - the OP was asking about using a 'crop lens' on FF, and the effect on noise. We're inferring from that question that the OP intends to shoot the combo in 'crop mode' and has no intention of viewing/printing those results any smaller than before.

If that's the case, there will be a difference in noise and in DOF in the resulting image. You can't crop (and enlarge that crop) without consequences to IQ. I'd hope both common sense and diligence in reading/viewing the evidence would correlate in people's minds enough here to put this to bed

.

Lets make it clear and to the point as possible.... FF was never mentioned by the original post at all.
Stop making this into a FF issue.

[jsherman999]

Lets make it clear and to the point as possible.... FF was never mentioned by the original post at all.
Stop making this into a FF issue.

Here's the exact question from the OP, pasted:

Would the high ISO performance be better with a crop sensor lens on the K1 compared to the K5iis?

The K1 is a FF camera. This is posted in the FF subforum, and the K1 is a FF camera. How much more clear can this be?


.

[amoringello]

Here's the exact question, pasted:

The K1 is a FF camera. This is posted in the FF subforum, and the K1 is a FF camera. How much more clear can this be?
.

With the option to shoot in a cropped portion with DA lenses.
Comparing specifically to the K5 which has the same MP count as the K1 cropped area, the issue pertains only to the cropped area.

I guess it would be nice if the OP clarified that point.

But it seems perfectly clear to me, and some others.
Of course the opposite is apparently quite clear to you and some others. So I guess we're at an impasse.


Q: My question would then be... if it were about the cropped area only, to match the cropped lens and coming form an otherwise equal sensor size and MP count as the K5, how would I be incorrect?



Other than that and obviously directly relevant to the original post, regardless of the context, the K1 is going to give better high Iso quality than the K5.

[audiobomber]

Q: My question would then be... if it were about the cropped area only, to match the cropped lens and coming form an otherwise equal sensor size and MP count as the K5, how would I be incorrect?

A K-5 would have lower resolution due to its blur filter. The OP has a K-5 IIs.

Other than that and obviously directly relevant to the original post, regardless of the context, the K1 is going to give better high Iso quality than the K5.

Assuming we are talking about the K-5 IIs, the K-1 will give 1-1/4 to 1-1/3 stop better ISO performance, but only in FF mode, not in crop mode. K-1 crop mode vs. K-5 IIs will be indistinguishable to users.

[amoringello]

A K-5 would have lower resolution due to its blur filter. The OP has a K-5 IIs.

... true (true being lower sharpness, not actually lower resolution... well depending what definition of resolution you are using. hopefully not lower MP count)
Regardless, that is relevant to my being incorrect, how?

Assuming we are talking about the K-5 IIs, the K-1 will give 1-1/4 to 1-1/3 stop better ISO performance, but only in FF mode, not in crop mode. K-1 crop mode vs. K-5 IIs will be indistinguishable to users.

So what you are saying is that when using the full FF sensor, the crop area within it will be brighter than that same crop area when using only the crop area (i.e. not full FF mode).
That doesn't even make sense.
You would need to adjust shutter speed or aperture to make a correct exposure if the ISO performance was somehow magically different.
Or does the K1 ISO range get limited to less than the 204,000K when in crop mode? I did not hear that the camera was crippled in that mode.

[enoeske]

So what you are saying is that when using the full FF sensor, the crop area within it will be brighter than that same crop area when using only the crop area (i.e. not full FF mode).
That doesn't even make sense.
You would need to adjust shutter speed or aperture to make a correct exposure if the ISO performance was somehow magically different.
Or does the K1 ISO range get limited to less than the 204,000K when in crop mode? I did not hear that the camera was crippled in that mode.

That's not what he said. I don't get why this is so hard for you.

[jsherman999]

So what you are saying is that when using the full FF sensor, the crop area within it will be brighter than that same crop area when using only the crop area (i.e. not full FF mode).
That doesn't even make sense.

Amoringello,

That's not what he's saying. The crop area within the FF (uncropped) image doesn't physically change or differ after it's simply cropped, what changes is the visible noise when you enlarge that crop to the original dimensions.

The only way a crop does not have more noise is if you leave it proportionally smaller, as in the examples given - crop 5x7 out of an 8x10**, and then just leave the crop at 5x7 - same noise in the crop as before. Enlarge the crop to the original 8x10, and the noise increases proportionally - or to maybe phrase it better, the noise that is already there becomes more visible.

You can of course test this yourself on one format if you wish. Crop 30% away from an image, preferably low-light/high ISO to make things easier to see, display both cropped and uncropped next to each other in same-size viewing panes. Repeat by doing that again, this time cropping the crop by 30%, display all three side-side at the same size. You'll see the noise increase with each succeeding frame.

**By the way, 5x7 out of 8x10 is a good example crop, because that's almost exactly an 'aps-c to FF' ratio crop.

[amoringello]

Amoringello,

That's not what he's saying. The crop area within the FF (uncropped) image doesn't physically change or differ after it's simply cropped, what changes is the visible noise when you enlarge that crop to the original dimensions.

This is exactly the point I've been making all along! -- Make up your minds!
Is it the fact that the image is made on Full Frame, or is it that you're CHANGING the image after the fact!?

If one does not understand how to stay consistent with what they're talking about, how the heck des one ever expect to understand what is going on? Or perhaps more accurately for other people to understand that you do know what you're saying and what it is that you''re trying to say?

These differences are NOT some magical effect of the sensor, but instead due to the fact that the image is being MODIFIED after capture.

And without understanding that point, my paraphrasing of audiobomber above is exactly what was being inferred.

The only way a crop does not have more noise is if you leave it proportionally smaller, as in the examples given - crop 5x7 out of an 8x10**, and then just leave the crop at 5x7 - same noise in the crop as before.

Yes, this is exactly what I've been saying consistently this whole time, and is the example I have used.
I'm always quite clear on staying relevant ONLY to what is captured at the sensor with the light coming from the lens.
I am never changing variables by increasing viewing or print size.

As I have always said, *stop changing things*. Once you change things, things change. (sort of a no brainer -- I don't know what is so difficult about it).
You cannot talk about one variable (the sensor) and then go of and change several other variables and talk about the result as if you never touch them.
(You'll find I've mentioned that several times, and people keep doing exactly that.)


At least you understand what is going on.
Thank you for articulating your reasoning.

It is sad that too many others have been duped by this sort of BS and stop short of understanding that the image has to be modified in order for the claimed effects to be seen. They simply become FF zealots and spew charts and sats without understanding what is behind them.

I understand its purpose of "equivalence" (worst name ever), but without including the additional steps of how/where that is being achieved, the blanket statement that FF changes noise, DOF, etc... is a lie and it does not give people the proper information they need to understand what is being said.

[Fogel70]

Yes, this is exactly what I've been saying consistently this whole time, and is the example I have used.
I'm always staying relevant ONLY to what is captured at the sensor with the light coming from the lens.
I am never changing variables by increasing viewing or print size.

So do you only view or print images in size of 24x17mm from APS-C cameras, and 36x24mm from FF cameras?

[amoringello]

So do you only view or print images in size of 24x17mm from APS-C cameras, and 36x24mm from FF cameras?

I do not even understand the relevance of the question.
Any question that contains the word "only" is trolling.

I may crop in 4x6, 5x7, 8x10, or other odd sizes to whatever is necessary for the intent of the image captured.
I may use 100% of the image, or I may use small percentage of the frame to whatever is necessary from the image.
I may print at 4x6, 5x7, 30x40, 40x60, and other odd sizes to whatever is necessary for the customer.

Go on and explain. How does your strange little world of two sizes fit into the real world?

If you're going to Troll please just move along. We've had enough troll BS in this thread.
But if you have an actual relevant contribution to the subject please feel free to do so.
... But PLEASE DO SO.

[audiobomber]

I do not even understand the relevance of the question.

Go on and explain. How does your strange little world of two sizes fit into the real world?

It has been explained to you repeatedly in two threads. Your inflammatory tone will likely result in this thread being closed if it continues.

The real world is viewing images on a digital screen at the screen's normal viewing size and resolution, or printing images to the same size from each camera. Comparing a K-1 and K-5 IIs:

• Shooting the same scene from the same position, with the same settings and lens and printing to the same size, the K-1 image will have superior DR, colour depth and SNR by about 1-1/3 stops. DOF and FOV will differ.
  
• Shooting the same scene from the same position, with an equivalent* lens on the K-5 IIs will result in the images being indistinguishable for practical purposes when printed to the same size, assuming no interpolation is needed (i.e. 16mp is enough to cover the print). Similar SNR, DR, colour depth, resolution, perspective, DOF, FOV and diffraction.

*Equivalent lens and settings example:
K-1: FA 50mm @ f4, 1/100s, ISO 400
K-5 IIs: FA 35mm @ f2.7, 1/100s, ISO 160

[amoringello]

So do you only view or print images in size of 24x17mm from APS-C cameras, and 36x24mm from FF cameras?

Here, I'll make this easy and not troll back, as I believe I know where you're trying to go with this.
But I am not sure you actually understand the issue.


Anyway, lets say I have the following situation.
* First lets get suspension of disbelief out of the way and say we have two cameras with peculiar specs.
Nothing too wild. Just for the sake of argument and to keep changes in technology out of the equation...

1) I have the K1 it is purely full frame. but it cannot shoot in crop mode, only the full frame may be used at all times.
2) I have the K5. but it has the same sensor specs as the K1 as far as quality, noise, pixel density, etc...
That should be a reasonable baseline and not cause too much strife. Since the whole argument is how FF makes the image different and "better"?

Each body has an identical 300mm f/4 lens attached.
... for the sake of argument -- remember we're not talking about technology or lens physics.

And let me know if I'm not correct with things so far or if I'm losing you along the way.

OK,
So, I am at a local game.
I'm at my favorite spot which will not change. (I'm all about limiting variables for the sake of argument)

I have the choice,
1) I use my K5 and fill the frame with one of the players faces.
Here the lens has an effective field of view of a 450mm lens so filling the crop sensor frame is simple.

2) I use the K1 and use about half of the frame to capture the same player's face at the same location.
Here the lens remains with it's normal 35mm frame's field of view of 300mm so I simply cannot get the reach necessary to fill the frame.

I get home and can easily send off the K5 photo of the player's face for print or display.
I look at the photo form the K1 and realize I need to crop in to use only the center 16MP to get the image desired.

Remember: Same lens, same focal length, same aperture, subject has not moved.
The shot from the K5 uses the full 16MP frame.
The shot from the K1 requires that I crop into the center 16MP area to give the desired result.

Now, please go on to explain and educate me in what ways the Full Frame sensor gives a "better" image in this example.


I could be wrong. I am always open to learn.

[Gimbal]

Remember: Same lens, same focal length, same aperture, subject has not moved.
The shot from the K5 uses the full 16MP frame.
The shot from the K1 requires that I crop into the center 16MP area to give the desired result.

Now, please go on to explain and educate me in what ways the Full Frame sensor gives a "better" image in this example.

It doesn’t. (inset face palm smilie) That is what we have been saying all along.

[amoringello]

It doesn’t. (inset face palm smilie) That is what we have been saying all along.

No! It is what IVE been saying all along. (in my whining four year old voice)

I think if people would have actually read what I've written over and over again rather than taking shortcuts and trolling, I think that would have been clear long ago. Like I've said before, we're likely talking about the same issues.

Without the simple physical base upon which this is all built, the partial explanations get lost and become some form of magic sauce.
With a clear road from light from the lens, to capture at the sensor plane to display/print, it really isn't that difficult.

Sadly many people actually do not understand these basics and like to believe in the magic sauce and will fight (and have) for that magic, tooth and nail.

It has been explained to you repeatedly in two threads. Your inflammatory tone will likely result in this thread being closed.

The real world is viewing images on a digital screen at the screen's normal viewing size and resolution, or printing images to the same size from each camera. Comparing a K-1 and K-5 IIs:

• Shooting the same scene from the same position, with the same settings and lens and printing to the same size, the K-1 image will have superior DR, colour depth and SNR by about 1-1/3 stops.
  
• Shooting the same scene from the same position, with an equivalent* lens on the K-5 IIs will result in the images being indistinguishable for practical purposes when printed to the same size, assuming no interpolation is needed (i.e. 16mp is enough to cover the print).

*Equivalent lens and settings example:
K-1: FA 50mm @ f4, 1/100s, ISO 400
K-5 IIs: FA 35mm @ f2.7, 1/100s, ISO 160

I do not know what you consider an explanations, but only the one response from late last night has been complete as well as accurate.

You say "Shooting the same scene from the same position, with the same settings and lens and printing to the same size"
This is a clear indication that you do not quite understand what is happening. It is also ambiguous. What does "printing to he same mean".. Both crop and FF printed 8x10, or same pixel to print ratio? i.e. crop at 5x7 and ff at 8x10 (approx)? Lets keep terms consistent and defined. Without an understanding of some basic guidelines partial explanations become white noise.


You cannot possibly have the same image from a FF vs Crop if you do not change some combination of the position, focal length, and/or aperture. Otherwise FF is going to contain a wider angle of view which needs to be compensated for.

I don't know where your lack of comprehension comes in. You're still sound confused on the basics.
It sounds like you're almost there, (as the second bullet point may be correct in proper cotext) but it comes across that you are still missing the most basic physics on which all of this stands.

That should not offend you, but instead simply be an indication of where you need to learn more.


Honestly, if you have the understanding you are not expressing it at all.
Stop taking shortcuts. -- partial explanations are meaningless.
Stop changing things (print size, captured size) -- if you don't understand the basics, you cannot possibly understand nor properly explain what happens later.