[rawr]

This remains to be seen still

I just wish they would hurry up and get something out. It's mid-September already!

[beholder3]

Here is an explanation of the auto-focus system:
[0026]
When the release button is pressed halfway, the focus is adjusted. The system control circuit 40 executes AF processing based on the pixel signal read from the image sensor 32. That is, the amount of defocus is calculated to shift the image plane to the in-focus position. In addition, AF processing by multipoint ranging can be executed.
[0039]
In order to form a pair of divided images according to the pupil division phase difference method, the focus detection pixel pair PB is formed with light-shielding films SL and SM that block half of the pixels. The light-shielding films SL and SM are formed at positions symmetrical with each other so that a pair of divided images are formed from each focus detection pixel pair.
[0040]
The focus detection pixel pairs PB, PG, and PR are not limited to the configuration in which they are arranged adjacent to each other in the oblique direction, and may be adjacent to each other in the row direction and the column direction. The light-shielding films SL and SM may also be formed at positions according to the arrangement direction of the focus detection pixel pair so that a pair of pupil division images can be obtained.
[0041]
For AF processing by multipoint distance measurement, a plurality of (nine here) divided distance measurement areas are defined for the entire image pickup area IM of the image sensor 32 ( see FIG. 5 ). When the AF process is performed, the focus detection process is performed based on the focus detection pixel pairs PB, PG, and PR belonging to the divided ranging area in which the target subject is projected.

If that "image sensor 32" really is something sitting on he prism where the light metering chip used to sit you have found a real big innovation here.

It would mean that they actually do what I wondered about long ago: why not cram a small smartphone image sensor on the prism.


The highlighted piece above in my view means that they indent to do Canon style Dual Pixel AF off of this small secondary sensor on the prism and can forego the smaller dedicated item in the bottom of the camera.

As a consequence I think they could drop the secondary mirror and thus the semitransparency using up light. This would lead to manual SLR style bright viewfinder.

It would not only allow pretty good mirror down subject tracking but depending on where they put these dual pixel thingies it would really improve AF coverage and make definition of AF points pure software (size and position).

And it would allow pretty easy AF fine calibration, because you could automate that: The camera just needs to check itself wether the depth information from the AF sensor (mirror down) is the same as when mirror is up and the main sensor can check.

In case the main sensor already has PDAF pixels (assuming the Fuji X-T4 sensor), you really have a lot of stuff the engineers can play with.

This does sound like something really new in the industry.

Now I am more excited about what we'll be presented in the beginning of November.

[Serkevan]

If that "image sensor 32" really is something sitting on he prism where the light metering chip used to sit you have found a real big innovation here.

It would mean that they actually do what I wondered about long ago: why not cram a small smartphone image sensor on the prism.


The highlighted piece above in my view means that they indent to do Canon style Dual Pixel AF off of this small secondary sensor on the prism and can forego the smaller dedicated item in the bottom of the camera.

As a consequence I think they could drop the secondary mirror and thus the semitransparency using up light. This would lead to manual SLR style bright viewfinder.

It would not only allow pretty good mirror down subject tracking but depending on where they put these dual pixel thingies it would really improve AF coverage and make definition of AF points pure software (size and position).

And it would allow pretty easy AF fine calibration, because you could automate that: The camera just needs to check itself wether the depth information from the AF sensor (mirror down) is the same as when mirror is up and the main sensor can check.

In case the main sensor already has PDAF pixels (assuming the Fuji X-T4 sensor), you really have a lot of stuff the engineers can play with.

This does sound like something really new in the industry.

Now I am more excited about what we'll be presented in the beginning of November.

If they truly manage to do On-Sensor (just not the main sensor ) Phase Detection AF with an OVF Pentax will have a depth charge of a camera. It might still need some AF fine tuning (automated as it may be), but it would be a massive advancement on a "dead technology from a century ago" (lol).

The PDAF pixels on the XT-4 come from the mask on the sensor, though, don't they? The K-New probably just won't have such masking, as is already the case with other sensors. It wouldn't make sense to reduce the usability of Live-View (due to no flipping screen) and put emphasis on the OVF (Have I mentioned I'm excited yet?) to then compromise IQ in all modes just for some AF gains in a mode they are quite explicitly sidelining.

[bencoskater]

Interesting. If by "not using the SAFOX sensor anymore but the RBG sensor" you are saying the metering sensor completely replaces the PDAF sensor in K-new for autofocus duties, I'm not sure that all the bits of info we have so far add up to that conclusion.

Early images of the K-new with no lens installed apparently showed the reflection of a larger (than in K-3) PDAF sensor under the mirror box, implying a larger coverage area. The cutaway drawing showing the new prism arrangement also shows a large area below the mirror box - I don't see the purpose of that space if not for a PDAF sensor.

I think prior Pentax cameras used the metering sensor to assist autofocus in tracking objects (by color, maybe shape), essentially giving the system cues on which PDAF point to use as object moved across the frame. My take on this may be completely wrong.

The sensor in the patent seems to be a Foveon-like sensor. Maybe it can be used together with a PDAF sensor to enhance the AF system? Or maybe the patent is for a future camera's AF system? Anyway, I'd be surprised if K-new doesn't use PDAF. I have no idea if they'll call it SAFOX or something else. But this is all guesswork by me based on what I think I know - again I may be completely wrong.

Everything is explained in the patent. The metering sensor is used to do the PDAF focus too by using the half of pixels (on each color). Regarding the patent (I didn't read everything in details) there's a double check of focusing (2 flowcharts are present). That would be a heavy task to use the SAFOX + the metering sensor for the processor in my opinion, that means there is a triple check for the focus. Unfortunately it appears that camera has only one processor (regarding the patent schema and explanation). The second CPU is used to control the motor of the lens nothing more.

It might be something for the future however Ricoh/Pentax says in their video that the prism, auto-focus and processor (I'm not surprised for the processor, it's an evolution of DSLR) have been improved, plus there is a new joystick which is be useless for a SAFOX sensor limited by the number of focusing points (in my opinion).

Keeping SAFOX (how many control of this part by the processor?) + double check from the metering RGB sensor will be a pain (sorry for being impolite) in the ass for the processor :/


I might be wrong, we will see sooner anyway.

---

Last edited by bencoskater; 09-16-2020 at 06:48 AM.

[ffking]

The second CPU is used to control the motor of the lens nothing more.

A good implementation of that would be very handy in the focusibg process, of course.

a pain (sorry for being impolite) in the ass for the processor :/

well, I've learned something I didn't know about processor architecture

[bencoskater]

If that "image sensor 32" really is something sitting on he prism where the light metering chip used to sit you have found a real big innovation here.

It would mean that they actually do what I wondered about long ago: why not cram a small smartphone image sensor on the prism.


The highlighted piece above in my view means that they indent to do Canon style Dual Pixel AF off of this small secondary sensor on the prism and can forego the smaller dedicated item in the bottom of the camera.

As a consequence I think they could drop the secondary mirror and thus the semitransparency using up light. This would lead to manual SLR style bright viewfinder.

It would not only allow pretty good mirror down subject tracking but depending on where they put these dual pixel thingies it would really improve AF coverage and make definition of AF points pure software (size and position).

And it would allow pretty easy AF fine calibration, because you could automate that: The camera just needs to check itself wether the depth information from the AF sensor (mirror down) is the same as when mirror is up and the main sensor can check.

In case the main sensor already has PDAF pixels (assuming the Fuji X-T4 sensor), you really have a lot of stuff the engineers can play with.

This does sound like something really new in the industry.

Now I am more excited about what we'll be presented in the beginning of November.

I think the sensor will remain at its place (at the bottom of the camera). Using a smartphone sensor would be great, there are only 2 problems:
- Pentax must create new optical path adjusted to the tiny sensor
- the other problem is the amount of pixels on this sensor that the processor could not handle for a fast response :/

Yes Fuji have it directly from the main and unique sensor. People must see the mirror-less like a DSLR with an active live view

---------- Post added 09-16-20 at 02:34 PM ----------

A good implementation of that would be very handy in the focusibg process, of course.



well, I've learned something I didn't know about processor architecture

I could be wrong on what I am saying. Like said by @rawr the SAFOX may remain on the DSLR to its place (in the bottom of the camera, we can see the place in the picture shared by Pentax) and the metering sensor will complement to help the focus system. I don't know where the metering sensor is positioned in Pentax DSLR. Some brands put it on the top of the prism of the OVF.

In any case, I hope this is going to really improve the auto-focus system in speed and especially when the subject is moving toward us.

I'll wait on real review (from real Pentax users not DPR or whoever from) of this new camera to see if it's a big change.

[rawr]

As a consequence I think they could drop the secondary mirror and thus the semitransparency using up light. This would lead to manual SLR style bright viewfinder.

That certainly would brighten up the viewfinder image. But to me the K-new cutaway image strongly suggests that the structure is still there for a PDAF assembly, which of course needs that secondary mirror.

[bencoskater]

By looking the side view of the camera (K-1 vs K-new), it's still there:


So @rawr seems to be right on the assumption. They won't ditch the SAFOX easily.

[MMVIII]

That certainly would brighten up the viewfinder image. But to me the K-new cutaway image strongly suggests that the structure is still there for a PDAF assembly, which of course needs that secondary mirror.

I agree, there is a lot of space below the mirror box and I would also assume that some kind of sensors are implemented right there. I have no idea if it is possible to implement a measurement sensor, in combining CDAF (or the variant described in the patent) and the light sensor in that. If the coverage of the AF active area will increase, and especially to get light measurements for the whole image, this would also mean an increase of the secondary mirror.

Another possibility would be that the information overlay really might have moved to the light path at the back of the prism and the technology from the other cited patent is used. If the application of this new high resolution combined light meter with AF-capabilities can really be applied to one of the reflective surfaces of the prism this could be a dual AF-system, as some have already pointed out, or at least support tracking and object recognition. This could be the space (prism surface 64) where now AF-points display is, that would not be needed with that.

---------- Post added 09-16-20 at 04:13 PM ----------

By looking the side view of the camera (K-1 vs K-new), it's still there:


So @rawr seems to be right on the assumption. They won't ditch the SAFOX easily.

Regarding your sketch with metering system: No, the indicated places are not the light meter but a "superimposing optical system", probably replacing AF-point indicators and the bottom viewfinder info display in one system.
See the details in this cut model: [Edit: follow the link to view an image with descriprion of the components.]

https://www.pentaxforums.com/articles/photokina-2012/pentax-k30-k-5-cross-section.html

[bencoskater]

you're right (by looking the path of the light) so there is no more space for the metering sensor on the top of the new camera. Which means it has been moved somewhere else, I don't see any place except the place used for the SAFOX. Either it's been added in the path of SAFOX optics which means less light for the SAFOX and struggling to focus or SAFOX is replaced by the RGB metering sensor (what I and @rawr assumed from the beginning). In this case that's the end of SAFOX and the focus joystick make sense here, otherwise there is no reason to install a joystick for 33 or even 52 focus points in my opinion.

[repaap]

you're right (by looking the path of the light) so there is no more space for the metering sensor on the top of the new camera. Which means it has been moved somewhere else, I don't see any place except the place used for the SAFOX. Either it's been added in the path of SAFOX optics which means less light for the SAFOX and struggling to focus or SAFOX is replaced by the RGB metering sensor (what I and @rawr assumed from the beginning). In this case that's the end of SAFOX and the focus joystick make sense here, otherwise there is no reason to install a joystick for 33 or even 52 focus points in my opinion.

There is quite strong rumour for 100+ focuspoints with larger coverage than what it is now.

---------- Post added 09-16-20 at 17:48 ----------

I just wish they would hurry up and get something out. It's mid-September already!

I agree. They should spill the beans already!

[FozzFoster]

I just wish they would hurry up and get something out. It's mid-September already!

"A delayed camera can eventually be good.
A rushed camera is forever bad."
-Shigeru Miyamoto (if he made cameras)

[texandrews]

Just as a comment from a new moderator: These last few pages are a model of a great speculative discussion. Excellent back and forth, supported by good info and ideas. Might use this in the future to point people to. Thank you all very much.

[bencoskater]

"A delayed camera can eventually be good.
A rushed camera is forever bad."
-Shigeru Miyamoto (if he made cameras)

I agree with you, they better have to take the time to finalize it. Developing a camera reminds me how long it takes to develop an air plane, it takes couple of years and thus used older technologies compared to date technologies.


The camera is almost ready to go in production, even the model name is already defined. If you look closely to the published pictures of the new camera, there is a black tape hiding it. The only thing that could be improved and debugged is the software part, all physical part is ready for the production (excluding minor electronic issues). Only major physical issue could impact the delay of production.

This is how I see things (that could be different workflow for Ricoh/Pentax).

---------- Post added 09-16-20 at 04:20 PM ----------

Just as a comment from a new moderator: These last few pages are a model of a great speculative discussion. Excellent back and forth, supported by good info and ideas. Might use this in the future to point people to. Thank you all very much.

Thanks @texandrews. I understand your point. All the time I read a thread, people don't stick to the subject and I'm frustrated. This is the reason I am 99% a reader than contributing.

From here I'm sticking to the subject to do not make a new thread in the thread lol.

---------- Post added 09-16-20 at 04:32 PM ----------

There is quite strong rumour for 100+ focuspoints with larger coverage than what it is now.

I hope too. Even if we got 70 to 90 points with a better and accurate auto-focus would be a huge improvement. I was close to buy a Nikon due to the auto-focus. Thanks to the rumors I saw, I am waiting on the tests in real life like I said from real Pentaxians.

PS: The only flaw I see is the quality of the joystick which may fail in the time.

[rawr]

so there is no more space for the metering sensor on the top of the new camera.

I think there probably is, looking at K-5, KP and K-1 cutaways. It is likely the K-new sensor will be in much the same place and alignment. It doesn't take up too much space above the viewfinder eye-piece lens assembly.

K-5 cutaway from PF

And if necessary, instead of standing it vertically, Pentax could lay the RGB sensor flat down and put a small prism below it, for example like the Canon 7D II does for it's own 150k RGB sensor.