[dadipentak]

I'm not fooled. Sure, it looks great but I suspect it's as much JohnBee's virtuosity as the camera.

[JohnBee]

Astounding.

My K20D ISO6400 shots look like Seurat.

The K-5, at 51200, is breathtaking. I know downscaling helps a lot, but still!

I've managed to push it up a tad:

K-5, ISO51200


It's a work in progress

[dgaies]

I've managed to push it up a tad:

K-5, ISO51200


It's a work in progress

Is that giraffe sitting on top of a K5 box

[pefoto01]

Bravo!

[mickey]

Those images are great.
Really impressed. Just shows what can be achieved with a little patience and knowhow.

[creampuff]

JohnBee, I think a lot of Pentax users are amazed by the way you post process to reduce noise. As a service to the Pentax community, why not publicly share how you go about doing it? I think a lot of people are asking but so far you've not yet responded. I personally think the benefits of sharing is a good way to help shooters of all abilities to enhance their photographic knowhow. For your kind consideration...

[kittykat46]

Am I missing something? Bearable????? Could you ever have imagined even getting a shot at 51,200? I think expectations are a bit high. Those pictures look awesome! I relate this to dxing radio stations. Some people will complain that they lose their stereo signal and have to listen in mono when the station is far away. At least you're getting the station! I'd imagine that after the noise reduction a pretty good size print could be made.

Haha... sorry, I wasn't referring to John Bee's work, which is excellent. I think those are due more to his skill and knowledge of the NR software.

Of course, having ISO up to 51200 is a tremendous advance with digital sensors.
I was looking at some other K-5 ISO 51,200 shots I've seen with just basic NR.
They are not half as good, and no doubt could be a lot better with more work on the NR sliders.

[Rondec]

Wow, John, that is amazing! I really am shocked that there is anything "usable" at all that comes out of that iso 51,200 setting. I thought the fact that Pentax included it was just an attempt to one up Nikon, not thinking that anyone (even with your skills) could save a photo from it.

As others have said, Bravo!

[JohnBee]

JohnBee, I think a lot of Pentax users are amazed by the way you post process to reduce noise. As a service to the Pentax community, why not publicly share how you go about doing it? I think a lot of people are asking but so far you've not yet responded. I personally think the benefits of sharing is a good way to help shooters of all abilities to enhance their photographic knowhow. For your kind consideration...

I understand, and I've seen quite a few posts asking similar questions(both now and in the past).

The best answer is... I can't count the number of times I've actually sat down and began writing and editing a tutorial for my NR workflow.
I think the biggest issue here is not so much in writing what I do, but in writing so that it ends-up being practical rather than a recipe for a single image or instance.

With this in mind, all of my tutorials began quite well... until they all took flight in what we could only call extremely long and tedious documents.
Even on video(which I've tried), the process ends-up being far to involving to remain practical...

Ex. which slider does what and why? what approach do we take with certain conditions or scenario's? how many scenario's are there? and so on and so forth...

And, though I really don't want to sound like I'm inflating NR into some lofty enigma, the bottom line is, that's its often much easier to do something, than it is to explain it.
However, the good news is... that I have hardly ever used advanced NR in most all of the high ISO samples to date(other than the 51K samples). Therefore... its as easy as grabbing a RAW, developing it and applying NR(no magic, no tricks).

I think alot of people are being mislead by the before and afters, in thinking the NR work is the key, but I assure you... that its the files that are the source of the output and not the editing.
To which I'd say... grab yourselves a few DNGs and see for yourself...

[secateurs]

The best answer is... I can't count the number of times I've actually sat down and began writing and editing a tutorial for my NR workflow. I think the biggest issue here is not so much in writing what I do, but in writing so that it ends-up being practical rather than a recipe for a single image or instance.

In other words, it's an art, and you have a natural eye for it!

On a different note, is it worth shooting above ISO 1600, if all higher ISO's are simply in-camera re-processing?

Can we get just as good an image with computer NR work from an ISO 1600 image as one taken at 51200? Wouldn't it be better to do a single "step" of NR in computer software (i know it's more than a single step in reality), than to take an image which has already been tampered with by the camera and tampering with it again on the computer?

[audiobomber]

On a different note, is it worth shooting above ISO 1600, if all higher ISO's are simply in-camera re-processing?

Can we get just as good an image with computer NR work from an ISO 1600 image as one taken at 51200? Wouldn't it be better to do a single "step" of NR in computer software (i know it's more than a single step in reality), than to take an image which has already been tampered with by the camera and tampering with it again on the computer?

The only reason one raises ISO is to provide higher shutter speeds.

[jolepp]

In other words, it's an art, and you have a natural eye for it!

On a different note, is it worth shooting above ISO 1600, if all higher ISO's are simply in-camera re-processing?

Can we get just as good an image with computer NR work from an ISO 1600 image as one taken at 51200? Wouldn't it be better to do a single "step" of NR in computer software (i know it's more than a single step in reality), than to take an image which has already been tampered with by the camera and tampering with it again on the computer?

Assuming that the highest native ISO is where the camera firmware simply stops turning up the gain of some analog amplifier(s), and, instead, starts to use arithmetics on the bits from an AD conversion (losing one significant bit for each stop up thereafter) it would seem that the same raw image should result regardless of where the artihmetics were carried out. I'm probably grossly oversimplifying the process that actually goes on inside the camera, though.

[secateurs]

jolepp, please see my thread on how high sensitivity works on the K-5. https://www.pentaxforums.com/forums/pentax-dslr-discussion/118892-how-iso-320...hnical%5D.html

[RBellavance]

Assuming that the highest native ISO is where the camera firmware simply stops turning up the gain of some analog amplifier(s), and, instead, starts to use arithmetics on the bits from an AD conversion (losing one significant bit for each stop up thereafter) it would seem that the same raw image should result regardless of where the artihmetics were carried out. I'm probably grossly oversimplifying the process that actually goes on inside the camera, though.

One difference is that there is apparently in-camera noise reduction to applied to RAW files above ISO 1600. You would not have that in an underexposed ISO 1600 file.

[creampuff]

I understand, and I've seen quite a few posts asking similar questions(both now and in the past).

The best answer is... I can't count the number of times I've actually sat down and began writing and editing a tutorial for my NR workflow.
I think the biggest issue here is not so much in writing what I do, but in writing so that it ends-up being practical rather than a recipe for a single image or instance.

With this in mind, all of my tutorials began quite well... until they all took flight in what we could only call extremely long and tedious documents.
Even on video(which I've tried), the process ends-up being far to involving to remain practical...

Ex. which slider does what and why? what approach do we take with certain conditions or scenario's? how many scenario's are there? and so on and so forth...

And, though I really don't want to sound like I'm inflating NR into some lofty enigma, the bottom line is, that's its often much easier to do something, than it is to explain it.
However, the good news is... that I have hardly ever used advanced NR in most all of the high ISO samples to date(other than the 51K samples). Therefore... its as easy as grabbing a RAW, developing it and applying NR(no magic, no tricks).

I think alot of people are being mislead by the before and afters, in thinking the NR work is the key, but I assure you... that its the files that are the source of the output and not the editing.
To which I'd say... grab yourselves a few DNGs and see for yourself...

So in a nutshell it sounds like a NO...