[Erictator]

You may notice that the thread has been given a new home.


Steve

And that is entirely fine, and why I gave the Mod's the option in the first place. Carry on...

[Class A]

What I'm saying is the accelerator isn't noise processing. It's much more than that.

That is correct. The image processing also involves colour correction and sharpening.

The common theme to all "corrections" are that they are post-processing techniques.
Even Ricoh refer to them as "image-processing".

Hence, the "noise suppression" achieved is better characterised as "denoising" (achieved after the fact) and should not be confused with actual "noise reduction" (achieved in the sensor).

The Sony sensor is a black box that uses several techniques to achieve actual "noise reduction". What the Pentax engineers get to work with is a digital output stream. They can then massage the latter with image-processing techniques to make the image look better.

With a pure approach, the data coming out of the sensor would be stored as the RAW data in order to give users the option to either apply the Pentax-proprietary image-processing (in something like the adapted SILKYPIX), or use some other techniques, potentially better ones in the future.

No matter the magic sauce applied by Ricoh with the "accelerator unit" it could be just as well applied outside the camera.
Applying it inside helps with the perception of a better camera though. I'm assuming that's the reason why the processing won't become optional (again), even though the latter would result in a win-win. I hope Ricoh will prove me wrong!

Again, for those who like the image-processing (<- not my choice of words, Ricoh's choice of words), enjoy!

Just wanting to clarify that one must not expect significantly better results through improved sensor technology. As @beholder3 put it:

It's important for noob buyers to understand that sensors have not made any significant progress during all the last years other than for some fringe topics.

BSI, dual-gain, and stacked-sensors were the recent sensor technology advances and we won't see anything beyond that in the K-new.

Unless the K-new uses a dual-gain BSI sensor, improvements over the KP will come from the image-processing.
It is a judgement call as to whether one should, in this case, declare "improvements".
Some will see it like that, some won't.

[Class A]

If the camera has one big display or one big display plus a tiny one doesn't affect the sensor requirements.

It does in the sense that with the "tiny display" camera one is essentially forced to adopt OSPDAF (on-sensor PDAF) unless one wants to end up with Panasonic-style DFD focusing.

The "big display" camera, on the other hand, may make use of sensors that are not marred by the striping and banding that can be caused by certain OSPDAF technologies. In particular Sony sensors suffer from OSPDAF-induced artefacts, so I'm hoping Ricoh will not try to compete with MILCs with respect to LiveView AF. In my view they should, like Panasonic, prioritise the integrity of the image data, and live with slower LiveView AF, but still benefit from excellent PDAF (when the optical viewfinder is used).

[Rondec]

Unfortunately, I do think (as others have said) that sensor development has stagnated. That is to say, the priorities that Sony puts on new sensors are (1) more megapixels (2) faster read out (3) better video and (4) more PDAF sensors. Those are wonderful things, but they don't necessarily lead to better still image quality on an SLR, which is what I am interested in.

I am sure Ricoh has already selected the sensor and since this camera has been in development for a couple of years it is likely the 26 megapixel one. It should have faster frame rate and better video, but I guess we will see what improvements Pentax can bring to allow for top end still images. I know that it is sacrilege to suggest an Accelerator V 2.0, but the big things I am looking for are things like cleaner shadows, better dynamic range, and more usable high iso range. To be clear the issue with high iso isn't noise, it is lack of dynamic range -- even on a K-1, most images look pretty flat when you get over iso 6400.

[bobmaxja]

Most effort on sensor are for cellphone and auto mobile camera, that were the volume is and competition. And because Canon use their own sensor, that leave about 50% of the market for Sony sensor, not a big market anymore.

[thibs]

Most effort on sensor are for cellphone and auto mobile camera, that were the volume is and competition. And because Canon use their own sensor, that leave about 50% of the market for Sony sensor, not a big market anymore.

I think there's more to it.
Researching sensor advancement is nice and all but until you find a way to really advance in a field (let's say DR), well you'll try to push things in other fields. I can understad that.
If Sony is stuck regarding DR (and I think they are); they'd be VERY stupid not to reserarch and market around fps, on sesnor PD, video etc.

[bobmaxja]

Who do you think are the largest Sony customer? Do you thing when Apple and Pentax or Nikon call, who will get priorities?

Sony’s huge presence in the image processing market will certainly push this technology to clients at a huge scale. Hanson notes that the company has more than 60 percent market share, and shipped about 1.6 billion sensors last year. Among Sony’s customers is Apple, which uses the company’s sensors in its iPhone line.

Source: Sony?s first AI image sensor will make cameras everywhere smarter - The Verge

---------- Post added 2nd Aug 2020 at 20:59 ----------

Here more to point that DSLR or MILC are peanut market now for Sony imaging

"The growth in sensor sales over the past three years has been nothing short of remarkable. This year Sony will generate more profits from imaging than any of its other business lines. Sensors are on track to generate $1.92 billion in profits… 10% more than Sony’s long-established gaming arm.

Sony absolutely dominates the image sensor industry, accounting for over 50% of global sales.

The battle among smartphone firms to make better cameras has been a boon for Sony. It controls over 70% of the smartphone sensor market. And it’s been the exclusive maker of image sensors for every iPhone since 2010.

And when it comes to quality, Sony is in a league of its own. Its image sensors are so far ahead, it charges 2X as much as its closest competitor.

Right now, Sony’s world-class imaging business is flying under the radar. But as the computer vision boom takes off, I expect the stock to attract a lot of hype. And with image sensors becoming a larger part of Sony’s business, it could easily soar 300%+ in the coming years."

Source: One Investor's View on Sony CIS Business - F4News

[Rondec]

I think there's more to it.
Researching sensor advancement is nice and all but until you find a way to really advance in a field (let's say DR), well you'll try to push things in other fields. I can understad that.
If Sony is stuck regarding DR (and I think they are); they'd be VERY stupid not to reserarch and market around fps, on sesnor PD, video etc.

I think mirrorless created some problems. Auto focus really was terrible on early models. The solution was creating sensors with auto focus points on them. As time goes by, they add more and more of those. This isn't really a problem for SLRs, but they do have the ability to create artifacts. You fix one problem and add another one.

On the bright side, they allow MILCs to focus fast with lots of auto focus sensors that cover most of the sensor width.

[Jonathan Mac]

I really hope that they don't use a sensor anywhere close to 43MP. That's far too much resolution for the vast majority of applications and will take up too much space and take too long to process.

I imagine that the new APS-C camera will have one of the newer 26MP sensors which don't perform any better than the older 24MP ones except in readout speed.

[thibs]

I really hope that they don't use a sensor anywhere close to 43MP. That's far too much resolution for the vast majority of applications and will take up too much space and take too long to process.

I imagine that the new APS-C camera will have one of the newer 26MP sensors which don't perform any better than the older 24MP ones except in readout speed.

Storage and processing power are no problem at all (relatively speaking to e.g. 10MP K10D era RAWs). Though many softwares, Adobe in particular, are very badly threaded and can't take advantage of modern CPU/GPU power.
If Adobe (and others) get their acts together, once most people will finaly upgrade their PC (or MAC), it will be just fine.

[thibs]

I think mirrorless created some problems. Auto focus really was terrible on early models. The solution was creating sensors with auto focus points on them. As time goes by, they add more and more of those. This isn't really a problem for SLRs, but they do have the ability to create artifacts. You fix one problem and add another one.

On the bright side, they allow MILCs to focus fast with lots of auto focus sensors that cover most of the sensor width.

That too indeed.

[Serkevan]

Storage and processing power are no problem at all (relatively speaking to e.g. 10MP K10D era RAWs). Though many softwares, Adobe in particular, are very badly threaded and can't take advantage of modern CPU/GPU power.
If Adobe (and others) get their acts together, once most people will finaly upgrade their PC (or MAC), it will be just fine.

Considering they've had a good decade to realize that we're all using multicore CPUs, I think "if" is a very important word indeed

[normhead]

Considering they've had a good decade to realize that we're all using multicore CPUs, I think "if" is a very important word indeed

My 10 year old i7 quad core uses all four cores in Aperture. By comparison my current duo is slow and unresponsive. Aperture of course was cancelled.

[Serkevan]

My 10 year old i7 quad core uses all four cores in Aperture. By comparison my current duo is slow and unresponsive. Aperture of course was cancelled.

My 2012 i7 was (and is, whenever my father turns it on) chugging along quite happily in AutoCAD using several cores. Lightroom meanwhile behaves like an asthmatic snail while my 2017 i7 is sitting there like "eh, dude, just so you know, I'm basically free". The GPU is, of course, chilling in the background.

[Smolk]

Apart from a brief dalliance with Sigma DP Merrills (which only ended because of the lack of a proper viewfinder), I still work in film only. But with a serious scanner (Screen Cezanne) that only runs on my 1999 Mac G4 Tower because of the SCSI connection... It delivers serious file sizes, 38-41 Mb Tiff for 35mm film and up to 100Mb for 6x6. I experimented with larger (and smaller) files, but they did not return more (or as much) detail. So I have no issues with large file sizes. I must pray my scanners keep working... Some say that a lightbox plus stand and a good dslr can do the same trick (by stitching), but I've got no dslr yet so I cannot confirm.

I am tempted by the i-Knew, although I expect the to be released later FF KC to be even better suited to my tastes (focal length in lenses and viewfinder). For now, I'm still doing with my KXs and Z1 and a battery of lenses, or my Rolleiflexes, all in film. I do miss some shots every now and then (Z-1 decides that something is not in focus, or the lighting has just changed, and refuses to release for example) but hey, failure is part of the fun. There are some shots for which I need IBIS, however, so I am monitoring the Pentax developments very closely.

I love Pentax's dedication to the OVF, and it has reassured me that Pentax is not going to wither. So the K-new or FF K-newer will probably be my next purchase.