The Next Generation?

[rparmar]

Originally Posted by benjikan

What will the next generation of DSLR's be? Will the pixel race ever end and will the consumer accept Quality over Quantity. When is enough enough? The numbers game has worked to date, but will the majors start communicating other aspects of what is important, to slow down this endemic procession?

I don't think it will happen. In the p'n's arena Fuji had a line that bucked the trend for a while, trading more MP for better quality, especially in low light. But the F10, F11 and F30 (of which I own two!) were succeeded by more traditional market-driven cameras.

Sigma shows they are willing to try something different, but their efforts have been disappointing in terms of IQ and are priced too high.

What I want from Pentax is a no-mode camera. Just dials for aperture, shutter and ISO like on an LX or something. Everything at one's fingertips with dials that don't change what they do based on arbitrary modes. Strip off RAW buttons, SR switches and other junk. But put on a custom function mode so we can add whatever we want to the four rocker switches. Make it smaller and lighter so even the M8 folk get envious.

I have been wondering why Pentax produce such quality and beauty in the Limited lenses when there is no Limited camera to go with them.

[Tom M]

Originally Posted by rparmar

I have been wondering why Pentax produce such quality and beauty in the Limited lenses when there is no Limited camera to go with them.

And if they did, I truly believe they could make a Leica M8 killer..

[*isteve]

Originally Posted by creampuff

Important thing is will the next camera be out by Christmas?
Pentax should abandon rolling out lenses with the APS-C sized image circle pronto.

Why? Since 95% of pentax users will never go FF, why should they pander to the 5%? APS specific lenses are cheaper and/or sharper for the same focal range. Pentax's great selling point as far as I am concerned is that they are the only APSC SLR maker who is taking the format seriously.

You will be able to buy FF lenses so whats your problem?

[rparmar]

Originally Posted by Tom M

And if they did, I truly believe they could make a Leica M8 killer..

Exactly so. And I imagine it might cost less than four grand.

Another thing that might help is if they stopped with the ridiculous model names. I cannot imagine how any marketing genius thought it wise to make the consumer count zeros to determine which body is which. And that is alongside the insanely ill-thought out "*" designation.

Originally Posted by rparmar

Sigma shows they are willing to try something different, but their efforts have been disappointing in terms of IQ and are priced too high.

Not going in a bayer vs foveon sensor discussion if Sigma DP1 excels in one department this must be IQ. Unfortunatelly it is a half baked product with very small buffer and nonresponsive in general use, atrocious frame to frame timmings, not that good LCD, slow AF, poor battery life. Nothing that reminds even remotely in use to a DSLR and by some accounts not even a modern digicam.
Too bad that Sigma didnd't do better with this cam in theory this is a niche with some lucrative prospects.

Radu

[nathancombs]

Originally Posted by RaduA

Not going in a bayer vs foveon sensor discussion if Sigma DP1 excels in one department this must be IQ. Unfortunatelly it is a half baked product with very small buffer and nonresponsive in general use, atrocious frame to frame timmings, not that good LCD, slow AF, poor battery life. Nothing that reminds even remotely in use to a DSLR and by some accounts not even a modern digicam.
Too bad that Sigma didnd't do better with this cam in theory this is a niche with some lucrative prospects.

Radu

i was looking at the DP1 but @ 800$ i would be soon able to get a K20D LOL so i found a cheaper altuntive that will do the same thing i bought a Pentax auto 110 LOL
and it has pretty good reviews so now i have a good pocket camera w/ all the trimmings for 100$

[creampuff]

Originally Posted by *isteve

Why? Since 95% of pentax users will never go FF, why should they pander to the 5%? APS specific lenses are cheaper and/or sharper for the same focal range. Pentax's great selling point as far as I am concerned is that they are the only APSC SLR maker who is taking the format seriously.

You will be able to buy FF lenses so whats your problem?

Problem? I've got no problems I can think off.

The majority of Pentax users will never go FF because firstly there are currently no FF cameras in Pentax's line-up. True, many may own lenses which are APS-C specific. But that does not mean APS-C lenses won't be usable on FF cameras in the future (look at the example of the Nikon D3).

However there is nothing to say that people won't make the transition to FF cameras when they are eventually released. How fast or how slow it will be no one can say for sure. Obviously a FF camera won't be cheap initially simply due to the initial production cost for the sensor but like computer chips, the cost will progressively come down such that FF might well be within reach of consumer grade cameras.

APS-C lenses are sharper and cheaper? This is certainly debatable. There is nothing special about APS or FF lenses per se. The difference lies simply in the image circle projected. It is no more expensive to produce an APS or FF lens.

Pentax cannot afford to stick to APS-C when it's competitors are going FF. I would rather Pentax develop FF than remain strictly at APS-C. I don't want Pentax to go the route of Olympus 4/3, which is a digital dead-end.

My hope for Pentax to develop lenses with an image circle coverage for FF is for forward compatibility into the future. This does not equate to the abandonment of APS-C.

[xand]

Originally Posted by creampuff

The majority of Pentax users will never go FF because firstly there are currently no FF cameras in Pentax's line-up. True, many may own lenses which are APS-C specific. But that does not mean APS-C lenses won't be usable on FF cameras in the future (look at the example of the Nikon D3).

He obviously meant a slightly longer term than "now". Say 5 years.

And - will you really pay for a FF camera body and use existing DA (*even) lenses?

However there is nothing to say that people won't make the transition to FF cameras when they are eventually released. How fast or how slow it will be no one can say for sure.

except for what you pointed out below (emphasis mine):

Obviously a FF camera won't be cheap initially simply due to the initial production cost for the sensor but like computer chips, the cost will progressively come down such that FF might well be within reach of consumer grade cameras.

Sure, but there will never be a case where FF reaches price parity with APS-C - unless the APS-C format dies altogether.

APS-C lenses are sharper and cheaper? This is certainly debatable. There is nothing special about APS or FF lenses per se. The difference lies simply in the image circle projected. It is no more expensive to produce an APS or FF lens.

Of course it's more expensive (in general) to produce a lens which has in sharp focus a larger image circle.

Pentax cannot afford to stick to APS-C when it's competitors are going FF. I would rather Pentax develop FF than remain strictly at APS-C. I don't want Pentax to go the route of Olympus 4/3, which is a digital dead-end.

The more time passes, the less of a "dead end" 4/3 becomes. APS-C will die only if consumer demand stops or the sensor manufacturers decide.

Consumer demand is likely to stop (and sensor manufactures will probably kill it) only when APS-C vs. FF sensors are so similarly priced (or have such similar manufacturing costs) and the incremental costs of a larger viewfinder prism, etc, necessitated by the FF sensor are so minor, that the price differential to a consumer between APS-C and FF cameras would be low hundreds of dollars. I don't think this will happen for the next 5 years.

My hope for Pentax to develop lenses with an image circle coverage for FF is for forward compatibility into the future. This does not equate to the abandonment of APS-C.

You'd be better off hoping that 645 image circles are brought back. Why stop at "FF".

[falconeye]

Originally Posted by benjikan

When is enough enough?

Hi Ben, my view on this is this (my physicists's long-term view, not my photographic need's view for the next generation -- so you may want to ignore it ):

A pixel site should'n get smaller than about 3 micron (because of noise and the optical limits of fine glass like the FA 31mm which is still good at 170 Lp/mm). But not larger either, because our human eyes are so amazingly sharp at their point of sharpest vision (~81 MPixels when we are allowed to point them wherever we want on an image...).

So, we end up with FF and 96 MPixels. For rounding and marketing reasons, this means:

The DSLR pixel race will go up to 100 MPixels and stop there.

[rparmar]

Originally Posted by falconeye

A pixel site should'n get smaller than about 3 micron (because of noise and the optical limits of fine glass like the FA 31mm which is still good at 170 Lp/mm). But not larger either, because our human eyes are so amazingly sharp at their point of sharpest vision (~81 MPixels when we are allowed to point them wherever we want on an image...).

So, we end up with FF and 96 MPixels.

I do find this interesting but in the nearer term there are more practical concerns. The K20D already shows up the limitations of bad or low resolution lenses. A potential camera with more megapixels or a larger sensor would need even better glass. Consider the cost of your kit when the 31/43/77 is only the starting point -- not everyone can afford this.

Thus a camera that resolves more than the K20D is going to have progressively less of a market. That's why development at this end of the scale will be slower. (Though I think it will come.)

I really hope that bodies made to high standards have more appeal. Besides the Limited Body how about an underwater body? Or a crop sensor body with replaceable and hence upgradeable digital backs?

Or bodies in different designer colours?

[regken]

Originally Posted by falconeye

A pixel site should'n get smaller than about 3 micron (because of noise and the optical limits of fine glass like the FA 31mm which is still good at 170 Lp/mm). But not larger either, because our human eyes are so amazingly sharp at their point of sharpest vision (~81 MPixels when we are allowed to point them wherever we want on an image...).

So, we end up with FF and 96 MPixels. For rounding and marketing reasons, this means:

The DSLR pixel race will go up to 100 MPixels and stop there.

You have aroused my curiosity. Is there a way of looking at the Lp/mm of a lens and be able to tell if it is capable of out resolving a sensor with a specific MPixel count? Don't get to technical for us sales type people please.

Thanks,
Ken

[RiceHigh]

Originally Posted by benjikan

I am sure we will get there, but at what cost? To date, the K20D has served me well and for all intent and purpose outperforms the K10D significantly.

Would you elaborate more in what aspects does your K20D "outperform your K10D *significantly*"?

IIRC, before you had your K20D, the K10D had being looked perfect to you for what you had told us here.

Just very curious.

[MikePerham]

Originally Posted by *isteve

Why? Since 95% of pentax users will never go FF, why should they pander to the 5%? APS specific lenses are cheaper and/or sharper for the same focal range. Pentax's great selling point as far as I am concerned is that they are the only APSC SLR maker who is taking the format seriously.

You will be able to buy FF lenses so whats your problem?

Me too! I think the FF movement is misplaced. Only a small segment of the market is there and we have established how good APS-C can be. The current CMOS sensor is great and I am hopeing for a bit more camera than the K20D ...but will stick with APS as that is where my lens investment is.

Mike.

[creampuff]

Originally Posted by xand

He obviously meant a slightly longer term than "now". Say 5 years.

That's just an assumption on your part. 5 years is a long time in consumer electronics.

And - will you really pay for a FF camera body and use existing DA (*even) lenses?

Why not? That's backward compatibility. Friends of mine who have the Nikon D3 are still using non FF lenses because they still have APS-C bodies.

Sure, but there will never be a case where FF reaches price parity with APS-C - unless the APS-C format dies altogether.

Again that is an assumption on your part. The price differential may be closer than what it is at present. FF may never reach parity because of market segmentation by the manufacturers - FF for pro/serious enthusiast; APS-C for consumer/entry level.

Of course it's more expensive (in general) to produce a lens which has in sharp focus a larger image circle.

Then by that logic our venerable FA series of lenses should cost far more than DA lenses. FYI camera makers have been making lenses with an image circle to cover the 35mm format for decades and not any more expensive than modern digital only counterparts.

The more time passes, the less of a "dead end" 4/3 becomes. APS-C will die only if consumer demand stops or the sensor manufacturers decide.

4/3 & APS-C formats were born at a time when it was prohibitively difficult and expensive to produce a sensor equivalent in size to 35mm film. Sensor development has progressed to the stage that FF sensors are already a fact. APS-C might well be a transitional format in the evolution of digital sensors. There is also nothing to prevent FF & APS-C to co-exist.

Consumer demand is likely to stop (and sensor manufactures will probably kill it) only when APS-C vs. FF sensors are so similarly priced (or have such similar manufacturing costs) and the incremental costs of a larger viewfinder prism, etc, necessitated by the FF sensor are so minor, that the price differential to a consumer between APS-C and FF cameras would be low hundreds of dollars. I don't think this will happen for the next 5 years.

Your guess is as good as anyone's on this. But the price of the D3 is coming down since it's launch. So the price differential is narrowing. It could get much narrower in the foreseeable future with higher production volumes.

You'd be better off hoping that 645 image circles are brought back. Why stop at "FF".

Because even in the era of film cameras, few would want to carry such a big camera, duh. Hence the general and widespread popularity of the 35mm camera over the superior quality of the medium format cameras.

[falconeye]

Originally Posted by regken

You have aroused my curiosity. Is there a way of looking at the Lp/mm of a lens and be able to tell if it is capable of out resolving a sensor with a specific MPixel count?

Funny to see the word "arouse" in this context. As a non native speaker, I thought it would have some reserved other meaning

Well, if you know the Lp/mm of your lens (line pairs per mm, or lines per mm) then you know how many lines the lens resolves with 5% (or 10%) contrast (which is not much -- black and white both become very similiar shades of gray here!). You better know the lens' MTF which tells you how many lines for 70%, 50% or 20% contrast... Also, some tests quote the MTF50 figure (i.e., the figure for 50% contrast). But MTF50 is not normally the figure defined as resolution.

Anyway, assume you know the lens' Lp/mm figure for some accepted contrast figure.

Multiply by 2, then multiply by your sensor height in mm, to obtain the corresponding pixel row count.

Example:

Lens: 150 Lp/mm
Pixel rows: 150 x 2 x 15.6 = 4680
K20D rows: 3104

So, this lens would still yield some low (but acceptable) contrast between pixels.

Zeiss claims that their K mount lenses resolve about 300 Lp/mm. Such figures are normally obtained using laser beams and not normally being publshed. Review magazines cannot do this.

You can inverse the computation.

For a 14.6 MPixel sensor, you get 99.4 Lp/mm (also known as Nyquist frequency).

The better the contrast at 100 Lp/mm for a given lens, the better the image. Contrast figures beyond 100 Lp/mm don't matter for the K20D.


Most tests I see quote about 1300 Lp/IH (per image height) for the K20D and use the DFA 100mm Macro. This is below 1550 Lp/IH (82 rather than 99 Lp/mm) this lens should easily deliver. I think those tests use the MTF50 value and contrast with the DFA 100mm Macro at 100 Lp/mm is slightly below 50%. Because contrast can be boosted by the firmware, MTF50-based resolution values aren't very interesting and Pentax regularly looses here against the competition despite them having less pixels...