Personal View site logo
Make sure to join PV on Telegram or Facebook! Perfect to keep up with community on your smartphone.
Please, support PV!
It allows to keep PV going, with more focus towards AI, but keeping be one of the few truly independent places.
F Stop Flame
  • 119 Replies sorted by
  • @John_Farragut

    Sure is. Except the cows have come home and it's still going. I'm so glad @chazzmoe has come up with these Oakley light meters. That'll help for my M4/3 camera problems but I'll still have to chuck out my 1912 Daguerre and Leica collection. Seems their exposures were wrong all along...

  • rofl.. funny stuff. :)

  • I have seen this stuff on other forums, I mean, what is the point? Don't feed the trolls.

  • @DrDave I don´t think I´d describe Aki Hartikainen as a troll.. (in that I don´t think he was out to fight - he genuinely believes he has a case). More like an extremely stubborn guy who doesn´t listen.

    If memory serves it seemed like there was a similar situation in the topic of using magenta filters e.t.c and not just on this forum.

  • @drdave Generally if the guy types out more than 2 sentences per reply, he doesn't classify as a troll.

  • To answer oscillian's question about equivalence between the FZ200 and the GH2, yes, the FZ200's f=4.4 mm f/2.8 lens has roughly the same angle of view and aperture size as an f=12.5 mm f/7.8 lens on the GH2. The GH2 would need an ISO setting 8 times as high to give the same brightness in the output image. You can make a more accurate calculation by finding the exact sensor widths of the two cameras in 16:9 video mode.

  • Correct me if I'm wrong:

    1. Take three 12 Mpixel sensors (1/2.3", m43, FF).

    2. They all have specific lenses designed for them to produce similar FOV resulting in different focal lengths on each system depending on crop factor (9mm, 25mm, 50mm="normal"FOV)

    3. Each lens have a f-stop of say: 2.8

    4. Being different focal lengths, each lens produce an image with different spacial compression (background being closer or farther away, facial features flatter or more distorted) and different DOF from deep to shallow at the same f-stop while retaining the same FOV placed at equal distance from subject. (Edit: probably no difference in distorsion/compression @RRRR)

    5. Put adapters on the 1/2.3" and m43 cameras and place the 50mm f2.8 on all three cameras.

    6. We need to back away the cropped sensor cameras to retain same FOV. This changes distortion/compression for the crop sensors (making the 50mm a telephoto or extreme telephoto lens) The perceived DOF is shallower on the FF system due to the subject actually being closer to the lens.

    7. All three sensors need the same shutter speed and ISO to get the same exposure since the resulting 12Mpx image looks the same on all cameras.

    8. All three sensors have the same number of photo sites (pixel wells) but the individual size of each pixel gets larger with increasing sensor size.

    9. Signal to noise ratio is better with larger photo sites and allows higher ISO with less noise.

    10. Results

    1/2.3" sensor: deep DOF, bad noise performance.

    M43 sensor: "filmic" DOF, good enough noise performance.

    FF sensor: shallow DOF, excellent noise performance.

    So when Panasonic releases the FZ200 and I go: "Wow, that's a fast zoom lens with a constant f2.8" I should be thinking: "Everything's gonna look video-ish with deeper DOF, blown out highlights and I can't go over ISO 400 in a dimly lit church unless Panny has done some real magic with their noise reduction"

    Please chime in :)

    1. If the lens manufacturer manages to create completely identical optical features at each focal length for each sensor (to create the same FOV - not necessarily focal length) then there is no difference in terms of distortion or spatial compression, but indeed a difference in how the DOF appears on each sensor.

    Normally though, it is very difficult to create two different focal lenghts so as they appear similar in optical features (distortion, ca a.s.o.).. needless to say it would be even more difficult to calculate for such differences for different sensor sizes.

    And yes, noise is an issue when pushing many photosites into a small area. I can´t say much of the technical side of that – I´m sure someone else can chime in how it works, exactly.

  • @oscillian wrote

    All three sensors have the same number of photo sites (pixel wells) but the individual size of each pixel gets larger with increasing sensor size.

    Not true. Google pixel pitch. Pixels on sensor are like drawing small dots on a paper. There are more empty spaces. Welcome to the digital world.

  • Pixels on sensor are like drawing small dots on a paper. There are more empty spaces. Welcome to the digital world.

    This is extremely strange statement.

  • All those FF having more dynamic range is B.S.

    Canon 5D has about less than 11 stops DR. 5DmkII has less than 12 stops. Olympus OMD E-M5 has close to 12 stops. E-PL5 is rumored to have higher DR than E-M5. FF doesn't necessarily have better DR than smaller sensors.

    I see some benefits of FF though. Better low light performance for photo but not for video. Shallower DOF. A large selection of lenses.

  • @Vitaliy I don't think pixel pitch dimension is filled up with 100% actual pixels. Isn't there empty spaces between pixels? I recall a tutorial link posted by you. He was the inventor of CMOS technology. I think he said something like that there are more spaces between pixels.

    More about pixel pitch.

    How pixel size does matter.

  • @stonebat

    Just check citation that you used. Again, talk was about same resolution sensors. Of course, each pixel is larger and well is larger for larger sensors.

    DR is different thing as it is ratio between maximum real well capacity divided by electronic noise (error induced during reading actual electrons contained in well). So, if you have higher noise, you can have lower DR on same size sensor. Also worse microlenses can result in worse real well capacity.

    Modern sensors do not have gaps between microlenses, hence gapless term used in all recent announcments :-)

    Data you are deferring to arise from DXO measurements that require careful manual checking an reading all their procures description.

  • I thought the talk was digressed to dynamic range. Too bad the poor guy can't post for a month. But enough time for him to do more googling :)

  • oscillian said

    So when Panasonic releases the FZ200 and I go: "Wow, that's a fast zoom lens with a constant f2.8" I should be thinking: "Everything's gonna look video-ish with deeper DOF,

    Yes, but only because the lens has a small aperture.

    blown out highlights

    Yes, but the dynamic range is small because the pixels are small, and maybe also because it's a cheap sensor with poor dynamic range.

    and I can't go over ISO 400 in a dimly lit church unless Panny has done some real magic with their noise reduction"

    Yes, but only because the small aperture lens is capturing less light. The sensor itself could have similar ISO noise performance.

    Please correct me if I'm wrong.

    And so you see there are trade-offs for whichever direction you go with the sensor size if you want a big zoom range. On a large format, that lens will be slow, or huge, heavy, and expensive. On a small format, the aperture won't be very large. There's a happy medium, somewhere in the middle. It's part of the reason why 2/3-inch is the common ENG/EFP format.

    Stonebat, a full-frame sensor could have better ISO noise performance for video, if it uses large pixels or a proper downscaling or binning algorithm is employed. It's when line skipping is used that the benefits of a large sensor for noise performance are thrown out.

  • @balazer (October 14)

    "If I take a piece of 100 ISO film and cut it in half, is it still 100 ISO?"

    Probably it's still 100 ISO. However, I don't know, since it depends on definition of ISO for analog film. However, if you try to make a print of the same size taken with entire and cut film you would probably see the difference. If the final size and overall brightness of the print is made the same, you'll have a feeling that the picture made from cut film was taken with about 200 instead of 100 ISO film (nominal size film) due to higher grainess. Therefore, the print would seem more "noisy". I might be wrong, though, since I haven't made this experiment yet (and probably won't).

  • My first post here, and thanks to Vitaliy for all the great work you've done, and to all the members for the posts. I've read things here often.

    I wanted to toss aside the technical talk and to try to help clear this argument up, and shed some light on the subject. I believe a common misconception is that gathering more total light on a sensor transfers to more total perceived light in an image. This is the important thing to consider in Photography and Cinematography, perception. While it is true that you will collect more total light with a FF(full 35mm sized) sensor using, let's say, a 50mm f1.4 lens specifically designed for that FF sensor, this will not, even though it may seem counter intuitive, translate to less perceived light when adapted to a smaller sensor. The only difference, as it pertains to perception, is that the smaller sensor will seem to be a telephoto when compared to the FF.

    In any argument one must not confuse apples for oranges. I heard talk of focal reducers, darkroom photo enlargers, and actual magnification within a camera.

    I must applaud the talk of focal reducers as I find them very interesting, but alas it is confusing apples for oranges. In the case within this argument, if a focal reducer was used the FF sensor would vignette(this being a small circle of image surrounded by darkness in the image) and retain the same amount of total light, and no useful gain in perceived light, although if only concentrating on looking at the image and throwing away the black surrounding it, you would indeed perceive their to be more light(I.E. cropping the image). Now if we were to use this same setup on a smaller sensor we would get less or no vignette at all and if we compared it to the cropped image I spoke of in the last sentence, the amount of perceived light would be exactly the same....there is no X factor here, the same amount of light is there, no matter if you aim the light at a brick wall, a piece of paper, my forehead, or an image sensor....nothing can change this.

    Now then for the argument of a photo enlarger. This again is confusing apples for oranges. If we take an image with both a FF and smaller sensor with exactly the same lens, then convert both into real life photos or negatives, then use a photo enlarger on both, we will again see that both images have exactly the same amount of perceived light. Furthermore a real life photo/enlarger is actually a very bad example because the photo itself does not generate light, but simply reflects the light you turned on. We cannot and should not confuse digital information with actual physical properties...the two do not intermix. A digital zero is always a digital 0. A photon is always a photon....they are separate things.

    Now for magnification within a camera...this simply is the exact same argument as a focal reducer, so see that paragraph to clear that up.

    At the end of the day we must always ask ourselves "How is it perceived", and never "What is the math". We are Artists who depend on math and physics, not Physicists and Mathematicians depending on art.