[Ex Finn.]

I don't see that, in all honesty. I can understand folks being skeptical - defensive, even - until we see side-by-side comparisons between the K-1 and K-1II. The early reputation of the camera - both on and off these forums - is at stake, after all

The K1 cranks out impressive images.
If the output from the Mk II is similar to the KP in high ISO, then some cooking is going on behind the scenes... just saying.
I see no need to defend the brand. It is doing just fine in my world.

[BigMackCam]

The K1 cranks out impressive images.
If the output from the Mk II is similar to the KP in high ISO, then some cooking is going on behind the scenes... just saying.
I see no need to defend the brand. It is doing just fine in my world.

Noted. I just don't think we need to call other members fan-boys

[Ex Finn.]

Noted. I just don't think we need to call other members fan-boys

I should maybe apologize for the harsh language

[rawr]

Now that the K-1 II has it, and thus has entered into the world of professional use, and what this whole thread demonstrates, is that Ricoh needs to release a white paper on the Accelerator chip - what it's purpose is, how it works etc.

They have provided some notes on the subject on the K-1 Special Site, but they could go further.

Based on the knowledge and data obtained through the development of the PENTAX K-70 and KP, we knew that we could upgrade image quality in the PENTAX K-1 by coupling an accelerator unit with the PRIME imaging engine. It was more complicated than simply adding the accelerator unit, however. To keep with PENTAX’s standards for imaging, everything had to be rebuilt from the ground up after combining the PRIME engine and the accelerator unit with the 35mm full-frame sensor. The effort required to achieve our goal of exceeding the PENTAX K-1’s already exceptional image quality was harder and more intense that was initially imagined.

First was a repeated process designed to find the optimum combination of numerous factors, including the bokeh (defocus) effect, color reproduction, resolution and noise. Dramatic changes of the image as the sensitivity level was adjusted step by step would not be acceptable. PENTAX demanded the seamless reproduction of high-quality images at all sensitivities, regardless of the subject or scene. The goal was to provide images that are clearly superior, from the lowest sensitivity to the highest.

Realizing the ideal imaging concept included extremely detailed activities, because PENTAX places as much importance on the human perception of beauty as the company does in numerical evaluations. This resulted in a number of improvements for the PENTAX K-1 Mark II. For instance, there was a change in the quality of noise. Since the noise was identified more precisely than previously, it could be detected more clearly. The imaging concept which PENTAX has always pursued is evident in processes such as this.

Other improvements during shooting include improved resolution, more faithful color reproduction at higher sensitivities, and overall noise reduction. Since the camera lets the photographer set a sensitivity higher than standard, it expands the possibilities for a field camera throughout a wider range of applications: landscapes previously impossible to capture in handheld photography, or action scenes made possible only by varying the shutter speed. Video recording and the quality of live-view images also benefit. Image quality has been improved to a level where sensitivity as high as ISO 12800 can regularly be used. This expands a range of applications for the TAv mode, which allows the photographer to set the desired aperture and shutter speed, while shifting the sensitivity to control the exposure level of the subject.

Challengers | PENTAX K-1 Special site | RICOH IMAGING

Interesting that video is mentioned as one place where the improvements to noise performance may be felt.

[SSGGeezer]

Now that the K-1 II has it, and thus has entered into the world of professional use, and what this whole thread demonstrates, is that Ricoh needs to release a white paper on the Accelerator chip - what it's purpose is, how it works etc.

They have provided some notes on the subject on the K-1 Special Site, but they could go further.


Challengers | PENTAX K-1 Special site | RICOH IMAGING

Interesting that video is mentioned as one place where the improvements to noise performance may be felt.

Trade secrets are not for sharing with companies that like to reverse engineer lenses and software around every corner, (cough, Sigma!)

TMI has brought down many endeavors throughout history, even if whispered as a rumor.

They need to put up shot by shot comparisons and let the curious stay curious about how they accomplish their magic.

[Kunzite]

This must be the most "fan-boy" attended thread I have ever seen on Pentax-forums!

So we're "fanboys" because we don't immediately start whining about something we didn't establish if it exists in the first place?

We need a proper comparison with the K-1. Otherwise, one can see whatever he wants, if he's looking hard enough.

---------- Post added 20-04-18 at 10:54 AM ----------

They have provided some notes on the subject on the K-1 Special Site, but they could go further.

Not much further, I'm afraid.

In the Imaging Resource interview, they state about the Dynamic Pixel Shift:
"I'm sorry to say we can’t give any specific technical details about this. We still have some things that are very much proprietary, patent things, so we have to keep quiet about that, sorry to say. <laughs>"

If anyone is curious if the accelerator unit processing could be replicated by software:
"The accelerator unit initially processes the output signal from the sensor, meaning that the accelerator comes right after the image sensor. And then it conveys it to the PRIME IV -- PRIME IV is the name of our image processing engine -- and then an accessory unit does a kind of signal processing which cannot be obtained by just software processing mechanism without degrading the resolving performance of the sensor."
Ricoh @ CP+: Rewarding K-1 fans with a major upgrade, plus what comes next after the K-3 II?

[mohb]

This thread is a fine example of the dangers of pixel peeping and jumping to unfounded conclusions

[Rondec]

This must be the most "fan-boy" attended thread I have ever seen on Pentax-forums!
edit: or at least it ranks very high on the list...

I guess I'm probably a fanboy, although I haven't posted on this thread. Honestly, I have a hard time seeing the issues that some people are claiming. There may be a bit of softening, but I'm of the impression that neither photo is great nor would they look good printed at 100 percent. My feeling is that pixel peeping really high iso images is a fairly frustrating experience -- regardless of the camera. But looking at other threads where folks have posted high iso shots of night skies, I don't detect any stars being eaten or the like.

But I'm not an expert on the subject either. I'm happy with my K-1's performance and might get a K-1 II down the road if auto focus is a bit better.

[rawr]

It would be useful to provide some official before/after image examples of the K-1 II/ K-1 with/without the Accelerator Unit. At the moment they don't provide much info on that, short of this little video:

Youtu.be

Youtu.be

Not much further, I'm afraid.

The web K-1 II web page says almost the exact same as Ricoh said in the IR interview:

"To reproduce lively colors and rich gradations close to memory colors in all sensitivity ranges, the PENTAX K-1 Mark II newly incorporates an original accelerator unit, which efficiently processes image signals output by the image sensor before sending them to the imaging engine. This process upgrades both image resolution and color reproduction in a high-sensitivity range, while drastically reducing noise compared to the PENTAX K-1."

High-resolution design | PENTAX K-1 Mark II | RICOH IMAGING

After reading this, I'm particularly interested in learning what the Accelerator chip 'effect curve' looks like when the performance of the chip is measured along the ISO scale. Perhaps it does nothing up to ISO 400, then progressively starts to act as ISO increases.

[gaweidert]

Traditionally, a chip like the accelerator chip in the K1-II is to help number crunching. The idea goes back to 1980 when Intel introduced the 8987 co-processor. It worked on motherboards with the 8086 CPU. It's purpose was to speed up computation of floating point decimal arithmetic. If the accelerator is being used in the same way, then improvements due to it cannot be duplicated by firmware on the K1. For some functions it may simply be calculating the result to more decimal points for greater precision. The stand alone processor on the K1 could do this, but it would take longer to reach the same level of precision therefore the calculation is truncated in the interest of speed.

For instance, shake reduction calculations have to happen pretty quick. The accelerator chip provides this speed. Shutter shock on my K1-II is not as bad as it is on my K1 and SR seems to deal with it nicely. Autofocus calculations carried out faster with a more precision would be noticed too. Like it or not the two models will need different firmware and operational and imaging improvements due to having an accelerator chip on the K1-II cannot simply be corrected by firmware on the K1.

As to the original post, quite simply, more data is needed. I tested my K1-II at ISO 3200 with positive results and I found that hand held pixel shift works really well at that ISO too. The camera is merely a tool and no single tool will work well in all situations. So the human part of the equation is to figure out how to make the tools we have get the results we want in the situations we find ourselves in. Avoiding, or working around, the shutter shock "sweet spot" on the K1 is an example of this.

[normhead]

When I get mine, you'll see better. Just a prediction.

In part, because if I get that, I'll send it back.

I simply refuse to entertain the notion that some other camera would be better in the same circumstance without comparison images. in the same circumstances.

Even if the results are accurate, one camera is a very small sample, and the odds are probably stacked against someone duplicating the result with another mkII body, although who knows, maybe he's on to something.

[Class A]

I thought that was the intent in that test? - to underexpose the iso 100 shot and then push it to the same exposure and also get the same amount of light and noise, and then using different isos to see if the camera preprocesses the iso 100 and 800 shots differently (which it seems to do a bit?)

I'm not sure which test exactly you are talking about.

In general, yes, there are tests that compare pushed low ISO shots with high ISO shots. The idea behind such tests to examine how "ISO invariant" or "ISO less" the sensor is. It is generally a desirable property of a sensor if a pushed low ISO shots does not show more noise than a high ISO shot. Such "ISO invariant" sensors allow shooting with highlight margins (thus avoiding overexposure) and/or high dynamic range shots (which will see their shadows pushed in post-production), without paying the price of increased noise levels.

All I wanted to point out with my statement is that by pushing shots in post-production, one is increasing the effective ISO setting. In other words, you cannot underexpose a shot by three stops, push it in post by three stops and then say "look at at all that noise and mushy detail, it was only an ISO 800 shot", since what you are looking at is essentially an ISO 6400 shot. This is similar to how cropping changes focal length. One cannot shoot at 50mm, then do an extreme crop, enlarge the crop to normal viewing size and then say "Look at the perspective compression this 50mm delivers. By cropping, one increases the effective focal length.

So if anyone is talking about the noise in an ISO 800 (or so) shot, they need to be explicit about what kind of pushing in post-processing was necessary to achieve the exposure level shown.

Sorry, but I am not following you here. Are you saying that noise depends, at least partially, on the amount of light striking the sensor and that underexposure (less light) creates noise?

Yes, except that "partially" should be "mostly".

As photoptimist already elaborated upon, light is stochastic in nature. Light isn't "analogue" in the sense that you can dim it down to infinitesimally small levels while maintaining a "smooth" signal. Light is "discrete" as it interacts with matter in the form of individual photons. The lower the light levels, the more that stochastic nature of light comes to the fore. At very low light levels, the photon count becomes so low that there is not a chance of any even exposures across the sensor -- even if you had perfectly even illumination -- as the photons create a random and thus very noise tapestry of isolated exposure events. Even at very high light levels light behaves in the same random manner but as the signal to noise ratio (SNR) is much higher, the noise isn't as visible.

With modern Sony sensors, the read out noise is so low that you are mostly seeing the stochastic nature of light, not the sensor-contributed noise.

The reason why high-ISO settings provoke more "photon noise" (aka "shot noise") even in well-lit scenes is because high-ISO settings imply other settings that limit light entry. You typically have high shutter speeds and/or high f-ratios. If you didn't limit light entry in some way, the high-ISO shot would be hopelessly overexposed. So in low-light scenes, you see the stochastic nature of light directly and high ISO settings (or pushing in post) just magnifies the noise. In well-lit scenes, which would have a good SNR if recorded directly, high-ISO settings artificially limit the light levels reaching the sensor due to the light-limiting shutter speed / aperture settings, thus causing bad SNR, i.e., visible "photon noise".

[stevebrot]

As photoptimist already elaborated upon, light is stochastic in nature. Light isn't "analogue" in the sense that you can dim it down to infinitesimally small levels while maintaining a "smooth" signal. Light is "discrete" as it interacts with matter in the form of individual photons. The lower the light levels, the more that stochastic nature of light comes to the fore. At very low light levels, the photon count becomes so low that there is not a chance of any even exposures across the sensor -- even if you had perfectly even illumination -- as the photons create a random and thus very noise tapestry of isolated exposure events. Even at very high light levels light behaves in the same random manner but as the signal to noise ratio (SNR) is much higher, the noise isn't as visible.

With modern Sony sensors, the read out noise is so low that you are mostly seeing the stochastic nature of light, not the sensor-contributed noise.

The reason why high-ISO settings provoke more "photon noise" (aka "shot noise") even in well-lit scenes is because high-ISO settings imply other settings that limit light entry. You typically have high shutter speeds and/or high f-ratios. If you didn't limit light entry in some way, the high-ISO shot would be hopelessly overexposed. So in low-light scenes, you see the stochastic nature of light directly and high ISO settings (or pushing in post) just magnifies the noise. In well-lit scenes, which would have a good SNR if recorded directly, high-ISO settings artificially limit the light levels reaching the sensor due to the light-limiting shutter speed / aperture settings, thus causing bad SNR, i.e., visible "photon noise".

Thank you for the further elaboration. Clearly, the noise question is not as simple as is often portrayed in discussion and there are multiple points for potential intervention in the processing pipeline. Thanks again for the details. Photoptimist's explanation was enough (all I needed was the word "Poisson" to trigger the "aha" moment), but your additions fleshed the matter out nicely. I won't derail the thread further on this, though it might make a nice future technical article should someone be willing to write it up.


Steve

[stevebrot]

As to the original post, quite simply, more data is needed.

Ditto for the follow-up images from @aikaarska. I spent some time with the DNGs from those yesterday and was wishing for something other than wool to evaluate, the single strands of which are significantly birefringent by nature (outer sheath refracts a rainbow sheen out the sides). En masse, a bundle naturally appears like a field of artifact if even slightly OOF. The short story is that an additional stop of exposure would have been nice with those shots (all were exposed to the same LV...so much for test comparison) and that the intended plane of focus is not obvious. I am also finding that TIFF rendering varies significantly depending on software used for viewing with some viewers "normalizing" the image for display.

No, nothing is easy.


Steve

[Rico]

Yeah, so...

I took this as a part of my new series.



K-1 mk2, Samyang 35/1.4 (perfect sample), ISO800, DNG RAW, Lightroom 7.3 default sharpening settings

Closer look at lower left corner:



Horrible pixel level quality @ ISO800 (looks like cellphone image) from a sensor which gave a lot better results back in 2012 in D800(E) Nikon body. This is looking like filtering what happens with Sony A7x bodies after 3.2second exposures. Some examples of Sony filtered mess @ 100% crop:





Upper Sony crop is from a distant landscape, lower is part of a sand pile filtered to mush. That crop from K-1 MK2 is not far off. Everything in the crops are in perfect focus with lens stopped down far enough.

I am not seeing what you have a problem with in this image. The biggest factor to the image is the quality of light it was taken under.

Maybe you should have tried the hand held Dynamic Pixel Shift in this situation. At least you didn't have to worry about any movement from your subject.