Pentax Q with K mount converter: weird possibilities

[DrugaRunda]

well - I am one waiting to try this out...

the camera is also diffraction limited at about f2.4... still a stop or two down from that should not be too terrible, and one should still be able to make a nice A4 in good light even at f5.6, not to mention a web size shot... therefore using something like 300mm DA at f4 - you do not need the lens to stop down as it would not really yield anything better... but hey - I have few other lenses to try out on this... and tamron 90mm macro as 500mm lens or Voiglander 180 as 1000mm equivalent... mmmm

this is not about "wasting" good glass, as it is already "wasted" on K20D ... it is about giving it new and wonderful ability...I would bet that something like 90mm Tamron macro is a lot better as 500mm equivalent than usual superzoom glass... not to mention that it should be able to do 5:1 ! macro as a bonus... with f16 (FF) amount of DOF at f2.8 (so low iso usage on the tiny sensor is possible too)

while we do not have 390MP FF sensors to crop, we have the Pentax Q... once I get it and the third party adaptor, I'll let you know how it works out
--
common sense is anything but common

 

[Faintandfuzzy]

Imagine the Pnetax Q with a k mount teleconverter and a DA* 300 f4. That would give you an equivalent of a 1680mm lens!!! Now if only there was a way to hold it properly

Imagine the moderators finally banning 13 year old kids from posting in forums....

 

[Roland Karlsson]

No, it won't be so.
Light gathering ability of the lens and the size of its glass won't change.

When you use a smaller sensor, most light will fall outside the sensor.

The light intensity will still be the same of course, but on a smaller area.

Therefore less light is used to make the image. Therefore you lose light gathering capacity.

In practice this means that the pixels needs to be smaller on the small sensor. Therefore less photons are hitting each pixel. Therefore the light sensitivity of the system goes down.

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html

 

[godfrog]

That will be an expensive way to find out just how poorly suited lenses with image circles designed for 135-format are for use with a tiny 1/2.3" sensor. Lenses designed for compacts are much, much sharper in terms of lp/mm, but of course the image circle is also much smaller.

You are also forced to use electronic shutter which lowers image quality further, since the Pentax Q has no mechanical shutter.

If you want to see what it looks like right now, just buy a cheap film body, shoot some microfilm/TechPan/Acros100 and have it drum scanned, and crop an 6x4.5mm area. It's mush with a 300mm lens.

--
My Flickr:
http://www.flickr.com/photos/36164047@N06/

 

[Uloo]

Acording to your way of thinking. So what your saying is that if you using a full frame f2.8 SLR lens and use it on a APS size sensor wich is smaller, it will loose light and it wont be a f2.8 anymore and the image quality will be lower...... ?
mmmh

 

[Roland Karlsson]

So what your saying is that if you using a full frame f2.8 SLR lens and use it on a APS size sensor it will loose light

Yes - it will fall outside the sensor. In practice this means you have to have smaller pixels if you want the same amount. Therefore FF often have more pixels than APS-C. BTW - a zero size sensor behind any lens will gather zero light.

and it wont be a f2.8 anymore

No - it will still be F2.8. Its here the confusion starts. F2.8 is not a measure of light gathering. Its a number telling focal-lengh/aperture-diameter.

and the image quality will be lower...... ?

Yes - you can make sharper lenses for APS-C if they dont have to cover FF.

You have to remember though that FF or APS-C is not all that big difference. APS-C is just a crop factor of 1.5. So - its quite usable.

Using APS-C or FF lenses on Q sensors though ...

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html

 

[Roland Karlsson]

You are also forced to use electronic shutter which lowers image quality further, since the Pentax Q has no mechanical shutter.

Good point! Very good point!

That was clever of Pentax. It will force you to buy Q lenses. Thats 100% the opposite to what Pentax usually do - trying to be backward compatible.

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html

 

[snostorm]

No, it won't be so.
Light gathering ability of the lens and the size of its glass won't change.

When you use a smaller sensor, most light will fall outside the sensor.

The light intensity will still be the same of course, but on a smaller area.

Therefore less light is used to make the image. Therefore you lose light gathering capacity.

In practice this means that the pixels needs to be smaller on the small sensor. Therefore less photons are hitting each pixel. Therefore the light sensitivity of the system goes down.

Hi Roland,

I think you're wrong here. Consider the probable ISO range of a APS-C 12 MP Exmor R sensor (probably 100-25600) and compare it to the range of the Q (125-6400) I think that this indicates that the smaller pixel size is taken into account.

If your thinking was correct, smaller sensors would need proportionally larger front elements to get to attain the same light gathering capability (f stop). This obviously is not the case.

I could be wrong, I'm no camera or optics engineer. If my thinking is off, I'd be open to an explanation.

Scott

 

[snostorm]

So, my FA*250-600mm f5.6 would become like.........DANG!....a freaking monster. :-D

I'm sold. :-D :-D :-D

Hi Ron,

It's fun to think about the possibilities -- even if it doesn't work well in reality. . . but without thinking outside the box, where would either of us be???

Scott

 

[brandrx]

So, my FA*250-600mm f5.6 would become like.........DANG!....a freaking monster. :-D

I'm sold. :-D :-D :-D

Hi Ron,

It's fun to think about the possibilities -- even if it doesn't work well in reality. . . but without thinking outside the box, where would either of us be???

Vey well said Scott. I agree.

Ron

--
Ron - 'We don't have time to go take pics this afternoon Carl.'
Carl - 'What do you mean? It will only take 1/1000s.'

'Keep your eyes looking forward. However, glance back now and then to see where you've come from. It will put a smile on your face.' ~ brandrx

 

[snostorm]

That will be an expensive way to find out just how poorly suited lenses with image circles designed for 135-format are for use with a tiny 1/2.3" sensor. Lenses designed for compacts are much, much sharper in terms of lp/mm, but of course the image circle is also much smaller.

You are also forced to use electronic shutter which lowers image quality further, since the Pentax Q has no mechanical shutter.

If you want to see what it looks like right now, just buy a cheap film body, shoot some microfilm/TechPan/Acros100 and have it drum scanned, and crop an 6x4.5mm area. It's mush with a 300mm lens.

Hi godfrog,

You make some very good points.

I'd still like to try it though. Going against conventional wisdom has given me some surprisingly good returns so far, so I tend to not dismiss ideas unless I actually try them myself.

If I decide to buy a Q, I'll definitely give it a go. If it works, I'll have a lot of fun with it, if it doesn't, not much lost.

Scott

 

[jeanphilippe Goube]

And with a 600f4 !!!!!!!
--
jpgoube

 

[JacobSR]

This how I see it:

a lens with F2.8 from FF or aps-c camera has a bigger aperture than the f2.8 lense made for the Q. (the physical opening of aperture is much bigger) and the lenses that are made for the Q are much smaller with tiny aperture (opening).

So when a FF lens is mounted on the Q (wich has much bigger F2.8 opening than the F2.8 made for the Q) should allow more light and not less.

I know most of the light will fall around the sensor, but that loss of light is to cover FF image and is not needed for the Q sensor anyway. It's a crop sensor and it should have its fill of light. The unit of light per area should be the same or even more on the Q sensor.

The only thing that would make sense to me is a loss of light because of the greater distance of the optics from the FF lens to the Q sensor.

I am saying this knowing that I could be completley wrong, but this is how I see it, sombody please tell me what I'm missing.

Jacob

 

[snostorm]

This how I see it:

a lens with F2.8 from FF or aps-c camera has a bigger aperture than the f2.8 lense made for the Q. (the physical opening of aperture is much bigger) and the lenses that are made for the Q are much smaller with tiny aperture (opening).

So when a FF lens is mounted on the Q (wich has much bigger F2.8 opening than the F2.8 made for the Q) should allow more light and not less.

I know most of the light will fall around the sensor, but that loss of light is to cover FF image and is not needed for the Q sensor anyway. It's a crop sensor and it should have its fill of light. The unit of light per area should be the same or even more on the Q sensor.

The only thing that would make sense to me is a loss of light because of the greater distance of the optics from the FF lens to the Q sensor.

I am saying this knowing that I could be completley wrong, but this is how I see it, sombody please tell me what I'm missing.

Hi Jacob,

I've been thinking about this since my last post (see under the "wrong" subthread of this thread). The amount of light is actually the same. There aren't different standards for different formats. f stop values have to correspond with Exposure values so all the rules of photography work. There are, of course, sample variations and calibration variations for metering systems, and sometimes these are possibly tweaked by mfgs so they might be able to market a quality -- like when a particular model's stated ISO setting is actually faster than the actual value to make the noise performance look better at a given setting. They all have to be within some certain tolerances to be credible though.

Think about this:

You have a 12MP FF (24x36mm) body and a 12MP APS-C (16x24mm) body plus one f2.8 lens that covers the 135 frame and has a mount that fits both. Both bodies expose exactly the same with the same aperture, shutter speed, and ISO setting, as you'd expect. The APS-C has a sensor that has less than half the area of the the other, and the same number of photosites since they both are 12MP. The lens and register distance is the same for both bodies or the lens wouldn't focus from the same MFD to infinity on both bodies.

The "Laws of Physics" (I'm being sarcastic. . .this is an overused and specious term as usually applied on photo fora -- it's actually the Rules of Math) dictate that the APS-C sensor would need to have photosites that are at most proportional (less than half) the area of those on the FF sensor each (assuming that the position, shape and size of the photosites are proportional), so if the amount of light was cut roughly in half by the smaller photo sites, why do the cameras expose the same scene correctly with the same shutter, aperture, and ISO? Let's say you also have a 12 MP P&S camera with an f2.8 of the same FL equivalent lens and a 1/1.23" sensor, and when you meter the same scene and take the shot, it also exposes correctly at the same shutter speed, aperture, and ISO with only that tine sensor. Pretty weird right? -- not really, because the photography industry adheres (within tolerances) to the same standards of Ev and exposure.

The FF lens has a larger physical aperture than the P&S sensor lens only because it must cover a larger image circle. It's also farther away from its respective sensor (due to the longer registration distance used by the lens' and body's design) than the P&S Cam's lens. Neither of these make a difference, because the intensity of the light that gets to the sensor with equivalent Ev and f-stop values is the same between the respective cameras, or they would be out of tolerance for the industry standard and the rules of photography.

Any differences in the number of photons of light that reach each photosite for a given Ev are adjusted for in the design of the camera. Max Aperture value is dependent on the lens's FL and Aperture (diameter of the front element of the lens or entrance pupil) with no other obstruction of the light path. Smaller Avs for the lens are calibrated from the Max Av, with each stop reducing the area of the opening in respect to the previous by factor of 2 (if you have an f2.8 lens, then at f4, the aperture blades close down to the point that the area of the hole left is 1/2 of the area wide open, and f5.6 is 1/2 the area of f4, and so on. . .). Shutter speed is measurable against the international standard. ISO is calibrated against a measured standard light source using the calculated Av and the measured shutter speed and the values are set in firmware.

The designers of the camera decide the ISO range available in the camera given the sensor and their standards for IQ after test images are evaluated after processing by the camera. This is why the Pentax K-5 and the Nikon D7000, both with essentially the same sensor, have different available ISO ranges.

Sorry for the long winded reply, but wanted to try to explain this thoroughly given my understanding.

Scott

 

[Roland Karlsson]

That was a long post!

I read it all and I think you are right.

In a simplistic view - then the light gathering capacity of the camera, if you use the sam F-number is proportional to the sensor area.

But - then we make two assumptions.

1. Both sensors use the same technology level.

2. Both systems are made for the same quality output.

But - in practice this is not true. Therefore - the actual difference is probably somewhat smaller.

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html

 

[Roland Karlsson]

Hi Roland,

I think you're wrong here. Consider the probable ISO range of a APS-C 12 MP Exmor R sensor (probably 100-25600) and compare it to the range of the Q (125-6400) I think that this indicates that the smaller pixel size is taken into account.

If your thinking was correct, smaller sensors would need proportionally larger front elements to get to attain the same light gathering capability (f stop). This obviously is not the case.

I could be wrong, I'm no camera or optics engineer. If my thinking is off, I'd be open to an explanation.

You have some points. I would not say that your points make mw wrong

The problem is complex. A big and a small sensor do not use the same technology and they do not have the same design criteria.

Here are two extremes.

1. Lets ay that we had a small 10 MP sensor and just made it bigger - still a 10 MP sensor - but the pixels are just scaled up and so are everything else - also the electronics. Thats a simplification - but let us assume its possible. Then I would be correct. Then a twice as large senor would just collect four times the amount of photons at the same F-stop.

2. Lets say instead that we scale up the number of pixels only. The same size pixels - but much more of them. Lets us also demand that if we crop down the larger sensor to the smaller, then we shall have the same quality. Then people saying that F2.8 is always F2.8 are totally right.

In practice there is a compromise though. If you have a larger sensor - demanding larger lenses etc - then you expect some improvements.

Moreover - if you have a smaller sensor I assume you can afford a more sophisticated technology.

So ....

But ... I have assumed 1 above as the working model. Same number of pixels. And trying to get the same image quality. Looking at small and big with equal eyes. And if you set a price of $800 - then I think thats the correct view.

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html

 

[awaldram]

this seems so wrong

The F number is a ratio between the diameter of the aperture in the lens and the focal length of the lens.

So a 50mm lens will have the same aperture (iris) whatever sensor its fitted to it a constant ratio.

I think your all confusing Virtual focal length (crop factor ) with the physical characteristics of the lens.

a 300mm Lens on a 1/2.3 sensor has the same aperture as a MF 300mm lens it makes no odds as the Aperture is defined by the focal length/F number.

--
My PPG

http://www.pentaxphotogallery.com/home#section=ARTIST&subSection=1471087&subSubSection=0&language=EN
My Photo Stream
http://www.flickr.com/photos/awaldram/

 

[Roland Karlsson]

this seems so wrong

The F number is a ratio between the diameter of the aperture in the lens and the focal length of the lens.

So a 50mm lens will have the same aperture (iris) whatever sensor its fitted to it a constant ratio.

I think your all confusing Virtual focal length (crop factor ) with the physical characteristics of the lens.

a 300mm Lens on a 1/2.3 sensor has the same aperture as a MF 300mm lens it makes no odds as the Aperture is defined by the focal length/F number.

Sorry - I dont understand what you say is wrong.

That F2.8 always is F2.8 no matter how you crop - thats of course true. And I have not said anything else.

What I say is that if you crop then you lose otherwise catched photons. What you normally do when cropping is to scale up the image to compensate for the smaller size. When doing that upscaling you have a higher demand on image quality which means that you usually have to lower the ISO number (and maybe also, if possible, have smaller pixels). And its there you lose the sensitivity of the camera by cropping - or by using a smaller sensor - which is the same thing.

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html

 

[Greg Lovern]

How are you calling it F2.8 when most of the light is wasted and is falling outside of the tiny sensor? Wide open the lens would be something like F16.

That's like saying that if you crop your photo, it gets darker, and the more you crop, the darker it gets -- because more and more of the light is "wasted", falling into the cropped-off area.

Greg

--
Brand loyalty is a character flaw.

 

[Roland Karlsson]

That's like saying that if you crop your photo, it gets darker, and the more you crop, the darker it gets -- because more and more of the light is "wasted", falling into the cropped-off area.

Myari has a point.

The cropped image has to be scaled up to fill the same photo frame.

To be able to do that, with the same quality, more pixels and photons are needed.

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html