[Class A]

From what I heard I would guess that the accelerator unit is performing the same kind of de-noising process as DxO PRIME; at least it resorts to the same kind of algorithms.

Thanks.

So it seems you would agree that claims regarding the "accelerator" unit using information that is only available during the time of capture are to be met with a high amount of suspicion. Correct?

[reh321]

Thanks.

So it seems you would agree that claims regarding the "accelerator" unit using information that is only available during the time of capture are to be met with a high amount of suspicion. Correct?

I was surprised to see it in the K-1ii - because those users are get least benefit from using it - but once we see it there, we should expect it on all the models regardless of our opinions; it is Japanese practice and Pentax opinion that matter, after all.

[Class A]

Apologies Class A if I unintentionally offended you.

Not to worry!

Thanks and all good!

[house]

RAW Purity doesn't exist because there is image processing, you have to demosaic the image (as it has a Bayer filter array for colours), etc. etc. etc. It's mostly a line in the sand, but I suppose most of the "accepted" RAW massaging is the "necessary evil" needed to get the digital signal into a visible-on-a-screen image. This is a fair position to have, honestly. I keep saying that it would be great if they refined the accelerator or managed to make it optional.

You aren't really understanding the issue. Those equivalences you list are false and based loosely on the idea of 'untouched' or 'pure' as understood from other contexts. That's not what its about at all. The issue is if data is destroyed or not. The bayer sensor, the white balance or the colour profile doesn't destroy data captured by the sensor. Noise reduction destroys data. How much and whether it's an issue is up for debate and use case. Discussing 'purity' in some philosophical sense is misdirection or lack of knowledge. The great advantage of raw files is that you get sensor data. A raw file doesn't even have colour or whitebalance. Those are just operations you do when processing the files and the camera helpfully supplies values that you can choose to use.

[Class A]

- noise is generated exclusively on the sensor

Where else would it be generated?

The sensor delivers digital data. There is no way any noise could be introduced outside the sensor.

Of course, there is the noise of the light itself, but I think we are both excluding that here.

- no external factors could possibly "inject" noise into the image

Of course there are external factors that influence the noise, for instance temperature.

These have nothing to do with the "accelerator" unit, though.

Just a "noisier" power supply could affect the image you're getting.

Every manufacturer knows how to build a power supply that meets the requirements of a Sony sensor.

And, again, what does this have to do with the "accelerator" unit?
Whether anyone makes certain assumptions about the power supply being able to induce noise or not, has no bearing on what the "accelerator" unit does.

- there is no way to detect anything about the noise, except by analyzing the final "RAW" image

What alternative ways are you suggesting and, again, what's the relevance for an "accelerator" unit discussion?

Or the SR system, which - as we know - it's using electromagnets.

It does that or does not do that with or without an "accelerator" unit.

The "accelerator" unit has no chance to do anything against the transients produced by the SR system (see below).

Then there's the operating conditions - temperature, exposure duration, etc.

See above. Not relevant for what the true function of the "accelerator" unit is.

If the "accelerator" unit were able to mitigate any such influences then you'd be able to achieve the same with a K-1 by just stabilising the operating conditions.

Here's some random patent which talks about such issues:

This patent describes an approach in which the "noise-reading processor reads noise signals from a pixel circuit for a pixel that is an object of noise acquisition, in parallel with the reading of accumulated charges by the charge-reading processor.". This is a good idea to deal with transients during sensor readout but it is not possible to implement this scheme with a stock Sony sensor.

All I can say for sure is that it results in better images.

No one is disputing that you prefer the processed images.

[Class A]

From my standpoint, the accelerator isn't the terrible evil that has been reported...

I agree.

...and any detail that is smoothed isn't detrimental to the final image.

For some it isn't. For others the preservation of any true information is important (e.g., for image stacking).

[Tad1952]

I know more AF points is one of the big requests from Pentax users over the years......but......

am I the only one whose first action on getting a new camera is to change the focus points from "whatever they are" to central 9 ?

Centre spot for me 9 times out of 10.

[house]

I get the impression, considering Pentax continuing focus on "finishing touches" and other in camera processing in their communications, that they greatly emphasize in camera tweaks.

[Mistral75]

Thanks.

So it seems you would agree that claims regarding the "accelerator" unit using information that is only available during the time of capture are to be met with a high amount of suspicion. Correct?

Err... I don't see the point. The accelerator unit uses (and 'works on') the data transferred from the sensor, that's for sure. From the A/D converters to be more exact.

[reh321]

For some it isn't. For others the preservation of any true information is important (e.g., for image stacking).

Do you have a link to any evidence that image stacking is affected??

[Class A]

IMO all of our cameras deliver "unpure" images. Some tend to tone cyan, some warm.

Such colour differences are caused by the camera profiles used either in-camera to create JPEG files or by raw converters.

You could create the same colour "look" of another camera by just using the same camera profile (Capture One, for instance, allows you to select colour profiles from other cameras).

Typically, you won't get exactly the same results because manufacturers tend to use their own CFAs (colour filters arrays) i.e., specific Bayer filter mosaics.

Add lenses that render cooler tones or alter the color saturation or whatever and I don't think there's anything like "RAW Purity" anyway.

Lenses do influence colour balance but either these characteristics are faithfully captured by the raw data, or there may be some manipulation.

In summary, you touch upon two issues:

1. there are influences on the image that occur before the data is created by the sensor.
2. there are influences on the image after the data has been exported from the camera.

Neither of these influence categories have something to do with whether or not the data is manipulated before it is exported from the camera.

[Class A]

Err... I don't see the point.

The point is that some here stated that Pentax claimed

"that their adjustments to higher ISO values are somehow connected to the sensor in a way that must be performed at the time, rather than being standard denoising that can be performed at any time..."

According to my technical understanding, it makes no sense to make such a claim.

The accelerator unit uses (and 'works on') the data transferred from the sensor, that's for sure.

This is my understanding as well. Your statement seems to (again) confirm my view that claims to a sophisticated noise reduction scheme that can only be enacted within the camera during the time of capture are without basis.

First, such schemes are not feasible with the Sony sensor Pentax used.
Second, it would make no sense for such sophisticated schemes to exhibit smoothing effects. They would just remove noise and not remove detail as a by product.

[phoebus]

Centre spot for me 9 times out of 10.

Me too

[Class A]

RAW Purity doesn't exist because there is image processing, you have to demosaic the image (as it has a Bayer filter array for colours), etc.

Demosaicking occurs outside the camera. Respective algorithms operate on either raw data or on data that has been processed.

The validity of a "pure raw" notion is not affected by what happens to the "pure data" after the camera has written it to a card.

It's mostly a line in the sand, ...

In a sense that is true as raw denoising is very common and where the "line" is, is determined at which ISO setting the smoothing kicks in.

The problem with the "accelerator" unit is not that it manipulates raw data -- the majority of cameras do that -- the problem is that it already does this at ISO 640. I'd say nobody cares what happens at ISO 12,800 and beyond. Those ISO settings are not conducive to high-quality, detailed images anyhow. The concern with the "accelerator" unit is that it manipulates raw data in an ISO range that is not at all uncommon for modern digital photography.

[gatorguy]

Such colour differences are caused by the camera profiles used either in-camera to create JPEG files or by raw converters.

You could create the same colour "look" of another camera by just using the same camera profile (Capture One, for instance, allows you to select colour profiles from other cameras).

Typically, you won't get exactly the same results because manufacturers tend to use their own CFAs (colour filters arrays) i.e., specific Bayer filter mosaics.


Lenses do influence colour balance but either these characteristics are faithfully captured by the raw data, or there may be some manipulation.

In summary, you touch upon two issues:

1. there are influences on the image that occur before the data is created by the sensor.
2. there are influences on the image after the data has been exported from the camera.

Neither of these influence categories have something to do with whether or not the data is manipulated before it is exported from the camera.

Part of the question would have to be: Does the Accelerator help to mitigate any of the issues attributable to the lens and/or sensor? I don't know that anyone has actually determined the limits of what the accelerator does(or doesn't) and its effectiveness have they?

IMHO the Pentax Accelerator array may actually have the least negative effect of all three variables on the RAW image.
Of note Sony too seems to believe a modicum of RAW baking can be advantageous.
Proof: Sony Picture Profiles Do Affect Raw Photo Files | Fstoppers