[microlight]

Pixel shift for even slow moving subjects in its current implementation could only be feasible if the frames were taken very much more closely together in time, it seems to me. PS as it stands is very much a serial mechanical operation because the sensor has to move between frames, so we're up against hardware limitations. The KP's accelerator chip didn't appear to speed up the PS write rate so unless Ricoh has something up their sleeve apart from their arm, PS may not change substantively in the near future. An amount of motion is already coped with by motion correction, which works well in DCU5 and appears to be a lot less fiddly than RawTherapee to get very good PS output.

[Kunzite]

Yes, they know that they are doing a 50mm prime when people have asked for 20mm and 85mm primes because there are non available.
Result is, the poll about who will buy the DFA*50mm says that 80% of Pentaxians will no buy it.

Yes, they know what they are doing - because they're doing it, not just talking about it on a forum. And as you must know, the D FA* 85mm f/1.4 is planned as well, together with a D FA* wide and a D FA ultra wide.
Are you talking about the poll that says 20% of Pentaxians will buy it on release?

[biz-engineer]

20% of Pentaxians will buy it

On release date or later.

[Kunzite]

"Poll: Who will buy the new DFA*50mm on release?"
It's made clear that if e.g. you're waiting for reviews, prices, are in any way undecided you should vote "no". And it still got 20%.

Poll: Who will buy the new DFA*50mm on release? - PentaxForums.com

[fsge]

On release date or later.

Do you really think this kind of poll is something they shall trust? Serious photographer not only Pentaxian already has a 50mm (or 55mm). For those who prefer zooms this focal lens is covered by 18-135, 16-85, 17-70 in a very acceptable way. This DFA* main target is K1 owners who want to upgrade glass in my opinion...

[ogl]

No, I have no plans to buy it at release or soon after. I have no plan to buy ever.

[photoptimist]

Pixel shift for even slow moving subjects in its current implementation could only be feasible if the frames were taken very much more closely together in time, it seems to me. PS as it stands is very much a serial mechanical operation because the sensor has to move between frames, so we're up against hardware limitations. The KP's accelerator chip didn't appear to speed up the PS write rate so unless Ricoh has something up their sleeve apart from their arm, PS may not change substantively in the near future. An amount of motion is already coped with by motion correction, which works well in DCU5 and appears to be a lot less fiddly than RawTherapee to get very good PS output.

You are right that pixel shift is a serial operation, but the limits for pixel shift are not mechanical. Nor are they related to CPU performance or anything on the digital side of the camera electronics (buffer, UHS class, SD card class, etc.).

Unfortunately, PS is limited by the frame rate of the sensor (plus the photographer's chosen shutter speed) which is determined by the analog electronic's ability to read a clean analog signal out of millions of pixels of the sensor and convert all of them to digital values. Once the frame is grabbed, it can just sit in the buffer and wait for processing after all the frames are done. The camera does not need to do any processing to a frame before doing a shift and taking the next frame. The mechanical motions are so small (only 1 pixel) that they are nearly instantaneous -- probably only about 1/1000 of a second.

The technology in Sony's A9 (which can do 24 MPix @ 20 FPS) is the current state of the art in portable consumer device performance could, in theory, do pixel shift where the subject only needs to be totally motionless for about 1/6th of a second. Red's digital video cameras are faster (60 fps), but they cost on the order of $79,000 and need a battery that weighs about 1 kg.

[Rico]

Pixel shift for even slow moving subjects in its current implementation could only be feasible if the frames were taken very much more closely together in time, it seems to me. PS as it stands is very much a serial mechanical operation because the sensor has to move between frames, so we're up against hardware limitations. The KP's accelerator chip didn't appear to speed up the PS write rate so unless Ricoh has something up their sleeve apart from their arm, PS may not change substantively in the near future. An amount of motion is already coped with by motion correction, which works well in DCU5 and appears to be a lot less fiddly than RawTherapee to get very good PS output.

Using a 3 layered sensor would solve the pixel shift movement problem. Pixel Shift using a multi layered sensor could sample for even more colors for more faithful color fidelity and even greater clarity and sharpness. Of course Pixel Shift would introduce the movement problem again.

[dafbp]

- 28mp (on the assumption that I'm right about the above)
- 4K video (don't see how they can leave it out on a high-end model in 2018 and all the chips they use should support it)
- All fundamental specs at least as good as the K-3ii and probably better
- The accelerator chip from the KP
- Body features from the K-1 (monitor, illumination, 3rd dial)
- Released by late summer, along with the new wide APS-C zoom already shown
- Price probably above $1,500

As long as dynamic range and ISO performance don't get worse than Pentax's best [in APS], and the AF in the new camera becomes Pentax's best, I have nothing "against" these specs, although:
-16MP (K5iis) is already nice, and I wouldn't mind having more, but I'd rather trade the extra resolution for card write speed, DR, ISO IQ, and maybe some better jpg quality.
-no change in card format, only interface speed, please
-I'll use my phone for video, 4k or whatever, so don't really care
-3rd dial is useful, and the other stuff too, but some of the current dial options don't seem to benefit much from it (opinion)
-what do you mean probably ? You have more chances of it coming out before Summer than that happening


I'd buy that camera immediately (if before July), or next year (airshows are in the Summer...)

[credos4u]

I know some like the 3rd wheel, but for me, a 3rd wheel is literally a 3rd wheel.
I also think the wheel is too fujiesk and makes the whole design look "clumsy" somehow...
but i like having the front e-dial in the vertical position and the grip system found on the PENTAX KP.


Also note that video is a hot-topic for many people, and always was a reason for many people to switch away from pentax to other brands which can do more...
...i would say 4k video and better bitrates as well as betterframerates will have to be improved, or the market share of pentax K-mount cameras will shrink to MEANINGLESS, what will make the brand disappear within app. 5years or so.

What would mean: no more new lenses and absolutely ZERO 3rd party accessories for PENTAX cameras will be available in the future... support for the brand is already thin.
I am convinced popular features(people look after such features when they buy new gear, if they use it or not) could change that trend,
what would lead to more users, what would lead to more 3rd party support for accessories as well as for lenses.
(its a fact, that most people buy gear only if it offers a variety of popular features, you cant talk that away by sayin: i dun use video.)


and since someone said faster write speeds would be preferred, one should welcome the change to a more capable memory-card format.(XQD or CFExpress)

Having a real fast memory bus system and such capable cards, will not only make fast and nearly endless bursts possible they will make your material more safe.
That is, because the XQD format is slightly bigger than SD...
What not only means: Its more easy to handle and the cards dont fly around like tiddlywinks chips...
it also offers more space for the memory ... what means less heat gets generated and your safed data is more secure.
that is the main reason why you find Cfast ports in pro-cameras. (One can also try to talk away the safety factor of bigger media.. but in the end we all know, its better.)

And despite the fact, that SDXC cards already run hot at write speeds of less than 150MB/s (when in excessive use),
SD-Format will also reach the ceiling when it comes to memory space....
that will become important when you want to do video with your camera also.

whilst XQD-format already offers target read and write speeds from 1 Gbit/s (125 Mbyte/s) to about 4 Gbit/s (500 Mbytes/s) and storage capabilities beyond 2TB.
transfer speed with XQD 2.0 could get as fast as 1000MB/s... and will offer far more space.
and CFExpress(the format that will succeed in the end I guess, and which is said to be "backwards" compatible to XQD) will offer even higher transfer rates.
V1.0:app. 2000Mbyte/s
it is rumoured that following versions may offer transfer speeds up to 8000Mbyte/s.

And since digital evolution just doesnt stand still and resolutions seem to increase ... it would be a wise move to have a foot in that door sooner than later.


not that i would say, memory card format should be ranked top-priority....
there is still a bit space for speed and size increases in the SD format and
Autofocus for instance is far more important to improve!
but it would be an improvement that would make sense and it would confer a more professional claim to our cameras.

[photobobhunt]

Most of Pentax's cameras over the last few years have been marketed to entry level and those who like gimmicks such as multi-color"LED's and different body colors, and don't forget the Marc Newsom designed K-01 ( I own a yellow one converted to IR) .
Consider how many camera models were sold with the aperture block problem,(MyKS2) no recall ,no updated part. Same for the SDM motors.
With the SDM problem Ricoh could have provided a firmware update to allow using the screw focus ( In the DA*lenses)........ but no. It took savvy people to hack the firmware. My 17 - 70 f4 had the SDM motor fail twice ,just after the warrantee expired.
Ricoh inherited some of these issues, but in my opinion, the company should have made an effort satisfy customers who purchased unreliable products.
The real question is how committed is Ricoh to the Pentax brand and to the advanced user . Just saying.

[beachgardener]

we are closing in on the next possible announcement time, Pentax I think needs to give at least a teaser, if there is actually something on the horizon.

Come ooonnn... Pentax

oops, I'm probably watching far too much tennis

[Larrymc]

but i like having the front e-dial in the vertical position and the grip system found on the PENTAX KP.

I for one hope they do not depart from the E-dial configuration of the K-3/K-3II its ergonomically perfect the way it is. The Gimmicky front E-dial on the KP was designed that way to accommodate the interchangeable grips which was a not a great implementation to begin with. Don't mess with perfect ergonomics!

Larry

[reh321]

We don't want Pentax to become involved in XQD - Nikon experts tell me that the D500 can write to some SD cards fast enough to keep up with their processor. In addition, the XQD design is owned by Sony, and they always keep the price up by keeping the supply down, so I don't see an upside to them.


I've been told that the D500 and D850 each has a special processor to handle AF- that is where Pentax should put some effort!

[luismpg]

We don't want Pentax to become involved in XQD - Nikon experts tell me that the D500 can write to some SD cards fast enough to keep up with their processor. In addition, the XQD design is owned by Sony, and they always keep the price up by keeping the supply down, so I don't see an upside to them.


I've been told that the D500 and D850 each has a special processor to handle AF- that is where Pentax should put some effort!

I would say, as a K3-II user, that:

- better AF system,
- increase low light capacities would be OK
- add a litle flash

would be a great improvement, to turn the camera more versatle.


The other improvement, as more MP, buttons, changing the sensor. aren't in my idea necessary

so lets wait and see