[Aristophanes]

Don't these two contradict?

No. They occupy separate price points.

I see FF at over $1,500/body for the better part of a decade.

APS-C DSLR's will occupy the other 90% of the market and may drop per body into the $399 category all the way up to $1,299.

I do not see mirrorless erasing lower-end DSLR's from the map because the DSLR is versatile, fast (AF especially), suits people who want a larger camera with grip, has a real OVF, has terrific sunk cost ROI, and are have an enormous legacy already in the pipeline.

I think people greatly overestimate how much people are willing to pay for camera bodies. The market shrinks dramatically once you get over $1,000 per body. That leaves the vast majority of the market to APS-C.

Just my predictions.

[Clavius]

It depends if you read my whole post instead of skim reading it. In all probability amateur photographers aren't all going to buy FF cameras - especially if there is a cheaper APS-C camera available that has excellent, compact lenses like the DA limited lenses to use on it. But professionals like me are going to be more interested in the FF and 645D cameras because of the FA limited lenses and the legacy glass we have. Pentax would survive longer as a company if they offered both APS-C and FF cameras - with the 645D at the apex.

No, I didn't skim read it. You said APS-H would be a half arsed solution that only prolongs the demise of the company. Then you said it's going to be FF or bust. Then, contradicting, you claim that Pentax can continue to market excellent APS-C cameras... How are they going to do that when they have met their demise by not marketing an FF camera then?!?

[Digitalis]

No, I didn't skim read it. You said APS-H would be a half arsed solution that only prolongs the demise of the company. Then you said it's going to be FF or bust. Then, contradicting, you claim that Pentax can continue to market excellent APS-C cameras... How are they going to do that when they have met their demise by not marketing an FF camera then?!?

Pentax can continue to make APS-C cameras which are excellent which is something we can both agree upon. However, if they ever want to move forward in the photographic industry and be considered by other prospective photographers then they will have to make a FF camera placed above the K5 - (or its sucessor). The bottom line is that an APS-H simply won't attract as many photographers as a FF camera would - and in the end that is what matters, a business has to make a profit somewhere. Hence my use of the phrase: Full Frame or Bust - either they make a full frame camera or they won't, there is no room for half measures. If they make an APS-H camera it would be a huge waste of effort for very little gain*, and losses like that is something that pentax simply cannot sustain.

* though if Pentax could chuck an APS-H monochrome sensor into a K-01 type camera, with removable IR filter- that could be fun - but it wouldn't attract the attention of a wider audience, not the way a FF DSLR would. I always saw the K-01 as an experimental camera, it isn't designed to compete directly with other mirror less cameras. That makes it so pentax can do some really weird things with it. Perhaps put in a 25mm square sensor, or make an ultrafast short backfocus wideangle lens specifically for it... the people who are interested in those kind of things will buy it.

[Clavius]

Pentax can continue to make APS-C cameras which are excellent which is something we can both agree upon. However, if they ever want to move forward in the photographic industry and be considered by other prospective photographers then they will have to make a FF camera placed above the K5 - (or its sucessor). The bottom line is that an APS-H simply won't attract as many photographers as a FF camera would. Hence my use of the phrase: Full Frame or Bust - either they make a full frame camera or they won't, there is no room for half measures. If they make an APS-H camera it would be a huge waste of effort for very little gain*, and losses like that is something that pentax simply cannot sustain.

* though if Pentax could chuck an APS-H monochrome sensor into a K-01 type camera, with removable IR filter- that could be fun - but it wouldn't attract the attention of a wider audience, not the way a FF DSLR would.

Ah yes, I fully agree to this. Especially the bold part. That is why I proposed them to make use of their SR system to do what Hasselblad does: shifting the sensor to creater images larger then the sensor format. That is something that the Pentax hardware probably already/almost supports, and can be used in their future FF body as well. The latter would give them a slight but extra edge over the other brands.

[Digitalis]

That is why I proposed them to make use of their SR system to do what Hasselblad does: shifting the sensor to creater images larger then the sensor format.

I have used these 200Mp Multishot cameras myself - and they can only gain the full advantage of the piezo shift with unmoving static subjects, also the sensor shift piezo system needs maintenance to keep it within spec. Since pentax has built its image upon rugged durable go-anywhere ergonomics and no nonsense design I would say using this technology wouldn't gather much following because it is only practical under very specific circumstances.

[Clavius]

I have used these 200Mp cameras myself - and they can only gain the full advantage of the pezio shift with unmoving static subjects, also the sensor shift pezio system needs maintenance to keep it within spec. Since pentax has built its image upon rugged durable go-anywhere ergonomics and no nonsense design I would say using this technology wouldn't gather much following because it is only practical under very specific circumstances.

Very specific circumstances, yes, much like the Pentax SR system that is only of use in very specific circumstances. There are even guides on how to get the most out of SR. Explainging when to turn in on or of. Explainging when the body itself turns it on and off. And explaining when it is an advantage, and when it has an adverse effect. So I think it suits Pentax perfectly. You don't have to use sensor shift "stitching", just like you don't have to use the SR.

Just saying, if the hardware already supports it, why not just develop it further?

However, requireing maintenance does seem like a drawback. If Pentax could solve that...

[Digitalis]

much like the Pentax SR system that is only of use in very specific circumstances

Actually the Pentax SR system is able to be used under more general circumstances than a piezo driven multishot shift camera can. When you are using a multishot camera nothing in the scene you are photographing can move, which means using flash or powerful continuous lighting with the camera bolted to a tripod - otherwise you will encounter artefacts and ghost images. And i'm not sure if the Pentax Magnetic SR system is capable of the sub-micron levels of accuracy that the hasselblad piezo driven system can achieve. The Pixels in the K5IIs are approximately 4.8 µm - that requires some very high tolerances and some expensive engineering.

However, requireing maintenance does seem like a drawback. If Pentax could solve that...

unlikely, if Hasselblad couldn't figure out a way around it: no one can - remember hasselblad engineered and developed the cameras that were used at the moon landings.

[Clavius]

Actually the Pentax SR system is able to be used under more general circumstances than a piezo driven multishot shift camera can. When you are using a multishot camera nothing in the scene you are photographing can move, which means using flash or powerful continuous lighting with the camera bolted to a tripod - otherwise you will encounter artefacts and ghost images. And i'm not sure if the Pentax Magnetic SR system is capable of the sub-micron levels of accuracy that the hasselblad piezo driven system can achieve. The Pixels in the K5IIs are approximately 4.8 µm - that requires some very high tolerances and some expensive engineering.

Dang, I wasn't aware of that difference, and it's implecations.

But I don't think the artifacts and ghost images are a reason for this feature not to be of any interest. With such an argument, exposure bracketing should be left out as wel then? No way!

unlikely, if Hasselblad couldn't figure out a way around it: no one can - remember hasselblad engineered and developed the cameras that were used at the moon landings.

Someone else tried and failed, so no one else should do any more attempts? You're not working for a boss, I figure? Attitudes like that get employees fired.

The moonlanding might be impressive if my phone didn't have over 100x the computing power then the equipment that they brought along on those moonlandings. I have respect for that achievement, but is history.

Let me put it this way...

I have a simple free android app on my phone with which I can do handheld panorama's. I activate it, and hold it in front of me, with stretched arm, swiping the camera over the scenery. Immediately it snap pictures, and stitches the pictures together, live on the screen. When it can't match the pieces of the puzzle, it highlights the areas that I have to repeat. The pictures are of course of phonecam resolution, but there are no stitching errors. It's flawless. This tiny little free app, developed by 1 single (but brilliant) person, in his hobby-time doesn't make this stuff feel like rocket-science. Nor does it seem like Hassy-grade-expenses are required.

Now... Why can't any prosumer or pro grade camera do this many times better? If this is possible with a static sensor on the end of my streched arm, then in-camera managed SR rapidly taking 4 pictures in each corner of it's SR system and sticking them together doesn't seem al that hard anymore. It doesn't even need to recognize image features or stitch them together, because it already knows the 4 positions of the sensor.

[Digitalis]

You're not working for a boss, I figure? Attitudes like that get employees fired.

I do actually - For my day job I work as a toxicologist for cancer researchers, and one important thing about research is looking at data gathered from previous and current experiments and work out what is feasible and what isn't. Hardware engineering follows a similar approach, see if similar mechanisms have been made - and use data collected about their construction and the problems encountered in making them so you can make alterations or improvements when designing your own.

I have a simple free android app on my phone with which I can do handheld panorama's. I activate it, and hold it in front of me, with stretched arm, swiping the camera over the scenery. Immediately it snap pictures, and stitches the pictures together, live on the screen. When it can't match the pieces of the puzzle, it highlights the areas that I have to repeat. The pictures are of course of phonecam resolution, but there are no stitching errors. It's flawless. This tiny little free app, developed by 1 single (but brilliant) person, in his hobby-time doesn't make this stuff feel like rocket-science. Nor does it seem like Hassy-grade-expenses are required.

There is a massive difference between hardware engineering and software engineering, while this could theoretically be done in software there would also have to be alterations done to hardware as well - also don't forget that a DSLR has considerably higher resolution and shallower DOF than what your smaller camera has - so you would have to use small f stops and that will mean lower shutter speeds ( or higher ISO) which will mean the SR/image processing system will have even more demands placed upon it.

[Fogel70]

Let me put it this way...

I have a simple free android app on my phone with which I can do handheld panorama's. I activate it, and hold it in front of me, with stretched arm, swiping the camera over the scenery. Immediately it snap pictures, and stitches the pictures together, live on the screen. When it can't match the pieces of the puzzle, it highlights the areas that I have to repeat. The pictures are of course of phonecam resolution, but there are no stitching errors. It's flawless. This tiny little free app, developed by 1 single (but brilliant) person, in his hobby-time doesn't make this stuff feel like rocket-science. Nor does it seem like Hassy-grade-expenses are required.

Now... Why can't any prosumer or pro grade camera do this many times better? If this is possible with a static sensor on the end of my streched arm, then in-camera managed SR rapidly taking 4 pictures in each corner of it's SR system and sticking them together doesn't seem al that hard anymore. It doesn't even need to recognize image features or stitch them together, because it already knows the 4 positions of the sensor.

This is something completely different from what Hasselblad does. Hasselblad shift the sensor a tiny, tiny, tiny bit, so it basically add resolution to the existing image. Stitching images to create panoramas is much easier, but using SR will create very small panoramas compared to the panorama function Sony and Samsung has in their ILC. so I'm not sure its the best way to go.

Most advanced users that create panoramas probably will not want to rely on a automatic function in camera, but prefer to do it manually. Just like most advanced users that create HDR images don't use HDR functions in camera. So automatic panoramas and HDR functions is probably are best suited for entry level cameras.

[RonHendriks1966]

So this CEO of Aptina has also a vision about larger sensors. I think the aps-h is also one

Sensor size is what matters and the trend is for larger, says Aptina: Digital Photography Review

[Digitalis]

So this CEO of Aptina has also a vision about larger sensors. I think the aps-h is also one

APS-H isn't the only format of interest here, perhaps we might see some camera maker come out with a digital version of the Hasselblad X-PAN.With a 24X65mm variable format sensor would be interesting indeed.

[RonHendriks1966]

APS-H isn't the only format of interest here, perhaps we might see some camera maker come out with a digital version of the Hasselblad X-PAN.With a 24X65mm variable format sensor would be interesting indeed.

That is a panoramasensor. I don't know, is that trick now more for software and gigapan-equipement? That sensor would even fit in a 645D_panorama version, since the lenses do cover that aerea!!!!!!!!!!

This could also be done for K-mount. When you hold the hight off the sensor the same as in aps-c you can make it 39mm wide.
Creating a sensor:

• 3264x8208=26,7 megapixel sized as in K-5
• 4000x10000=40 megapixel smaller pixels as in Nikon D7200
• panoramaratio 2.5:1

Or probably wen as wide as possible is the way to go then the hight off the micro4/3 sensor is a good start, the same goes for as many pixels as possible:

• sensorsize 13x41mm
• 3333x10512=35 megapixel
• panoramaratio 3.15:1

[RonHendriks1966]

From another threat:

I used Pentax film SLRs from about 1967 to 2004. (Then I switched to digital). I still have a number of (no longer used) screw-mount and K-mount film SLRs. I recently looked through their viewfinders to judge whether I still liked them.

My conclusion was that the older the camera, the less I liked its viewfinder. I now wear spectacles when using a camera, and I struggle to see the whole of the frame on the older cameras. But this isn't the same problem with later film SLRs. It appears (but is it true?) that the angle subtended at my eye by the later cameras is smaller, and this makes them more usable.

And I find the "grain" of the viewfinder of the older cameras a bit obtrusive. It probably makes them more useful for manual focusing, but also more distracting. Mostly I don't manually focus. I used to have two LXs (before they were stolen!) They had interchangeable viewfinders and screens. I didn't make much use of that, but I probably would now.

Question: how much is the effect of the "larger view finder" purely to do with the screen size, and how much to do with the eyepiece? Could a different eyepiece make the viewfinder of APS-C sensor cameras appear wider? (That would presumably also make them a bit darker). Could a different eyepiece on an FF camera make the viewfinder appear smaller for people like myself? Or have I got this totally wrong?

ps: I recently used my Pentax Z-1p to take lots of photos with my DA and DA* lenses to judge their image circle, hence usability on an FF camera. (I've posted the link to the results here). I feel that the Z-1p viewfinder is smaller than those for my older film SLRs, but it was also harder to use for manual focusing. Unless I am mistaken, just having an FF camera doesn't dictate all the aspects of viewfinder size or ease of manual focusing. The nature (not just size) of the screen matters, and I think the eyepiece does too.

Clicking along you come to this:
Cutting the 10-17mm fisheye
This makes it almost work without black borders on APS-H! Only the 10-11 mm give some heavy vignetting


Do I see aps-H advantage!!!!!!!!!!!!!!!

[Fogel70]

I don't see much use in APS-H sensors though.
I don't really see the point of optimizing sensor size for a few APS-C lenses, when you can use a sensor format that can be optimized for all lenses that has been used on K-mount.

Cropping to a smaller image circle is much more flexible than limiting sensor size, and Pentax could implement cropping on a FF sensor as a users selectable crop mode for all modern lenses. So if you choose 1.3x crop mode for DA 10-17 the camera will remember that every time you use that lens, and the OVF will show a frame on what part of it is in use. Maybe even a crop mode that can vary with focal length on a zoom lens so IQ could be maximized through out the whole zoom range. And Pentax need to do something like this to minimize the disadvantage of having so many APS-C lenses on the market, so maybe they even could add it as a pre-programmed user selectable (vignetting crop) option, just like distortion correction.

And with a 36MP sensor there could be at least 10 crop modes varying between 1.1x - 3.0x, and it could also be implemented as a digital zoom for times when wanting a crop straight out of camera.