I did comparition between D700 and ZDb , both runnin' at 14 bits, a while back, and ZD still would pull out a bit more "umph" in 3D feel of image.
As does 645D, from what we seen so far. You cant beat simple thing called "physics".
Bigger sensors / bigger film wins on larger prints.
It would be great if you could simply read a spec on the number of bits of conversion in the A/D filter and that would tell you everything.
It IS an important part of the puzzle. But the quality of the A/D convertor (not all 12 bit converters are equal, not all 14 bit converters are equal, not all 16 bit converts are equal), the heat sinking, electrical isolation, and quality of wiring between the sensor and A/D convertor matter, the use (or non use) of dark frame calibration and the quality of that dark frame calibration matter, the type/quality of demosasicing, noise reduction, and sharpening math matters (most raw developers start with relatively generic math which is open-source or university-researched but spend lots of time and money on tweaking them).
So sure, you can blame the bits (it does play a big role) but the dynamic range, and importantly how natural the outer range of the dynamic range can be rendered (tonal smoothness, shadow color accuracy etc) cannot simply be reduced to one thing.
Doug Peterson (e-mail Me)
Head of Technical Services, Capture Integration
Phase One Partner of the Year
Leaf, Leica, Cambo, Arca Swiss, Canon, Apple, Profoto, Broncolor, Eizo & More
National: 877.217.9870 *| *Cell: 740.707.2183
Newsletter | RSS Feed
Buy Capture One at 10% off
I taught a photographer how to use LR a couple of days ago. He's a big name in this town but he tells me he's a photographer not a technician and changed to digital thinking he could shoot and develop like film and got a rude awakening so he employed me to show him how to do it properly.
I calibrated his screen for a start. I then showed him how the different tools worked, when to use what for which usage, etc. He kept asking me how I knew what numbers to dial in, how far to push the slider. He kept asking me whether the histogram was correct. In other words he'd been reading too much.
The simple answer was 'when it looks right'. You use the tools for the correct purpose but when you know what to use when, just twiddle it until you like how it looks and screw what people tell you should be right.
There was an interesting article by Andrew Rodney on LuLa recently about calibrating a screen to print. He shrugged off all the conventional wisdom and put it in a nutshell 'when it looks right'.
I remembered this when I read Marc's post. The numbers don't mean squat. It's what it looks like and how it resonates with you personally for your needs. Anything else is just measurabating.
One persons needs cannot dictate your specific needs, what you try to achieve. I like good tonality and facial tones but couldn't give a bleep about accurate colours elsewhere. For someone else this is essential. Heck you could find a million other examples.
That article by two people who are neither pro photographers, neither skilled reviewers and neither experienced with the systems or software they used told me nothing. They weren't sure it even told themselves anything at the end of the day!
Bottom line, only opinion that counts is when you've shot it your way, processed it your way, printed it your way and you say whether it 'clicks' for you or not. Anything else is just an entertaining use of time reading academic studies of an subjective opinion about a camera based on the users specific personal needs, chips on shoulders and personal bugbears.
I like DPReview, he tells me what features the camera has with screenshots (nothing else). LuLa is amusing and entertaining but it tells me practically nothing about the fulfillment of my own needs.
I am not a painter, nor an artist. Therefore I can see straight, and that may be my undoing. - Alfred Stieglitz
Ben my theory numbers really count when you DON'T have a calibrated system otherwise they are nice guides to go by than do what looks the best. I ALWAYS tell folks on the workshops when they are processing WB means very little if you like a certain look in your files go with what feels good to yourself. People it is obviously a little more critical but again even after WB i will sometimes adjust to my liking. This is not to say though you should not know your raw processing better than you know yourself far from it become a master at it. If you know the software there is always a little gold token at the end of the process. LOL
Photography is all about experimentation and without it you will never learn art.
Marc, Pentax is 14Bit as I remember.
It would have cost far less to produce a 10 !
No it was a missed management which in return can raise up financial issues now. The sales of the M9 are done and the S2 looks like a flop, hard times ahead probably.
Lets not turn this into another S2/R10/Leica/shoulda woulda coulda thread.
We have enough of them already.
My experience is different. I went to 5dmk2 from 5d in the middle of my senior session and I didn't see loss of info or banding, but I don't claim to be an expert in that kind of thing. I just compare what I've done with one camera to what I'm doing with another camera.
I'm not sure my H3d2-39 at 800iso is better than my old Canon 20D. Certainly not better than my old 5d. I'll have to do a comparison during my winter slow season.
I was not referring to the 5D to 5DII transition, what I was meaning to say is that with the 5DII (for example) at the higher range of iso settings (and to a lesser degree the lower although it is still present) one can see a considerable amount of color smearing. This is one of the tricks of the trade that engineers do to improve the apparent high iso performance of the camera. The trade-off is that if they had not done that, and it is a deliberate choice, then the over-all look of the images would have been better but the high iso specsmanship would have been harder to assert.
Think about the physics for a moment, the photon flux at the sensor limits the potential signal to noise ratio based on the number of levels of quantization desired, the size of the pixel, the exposure/sampling time, the well saturation limit, and the quantum efficiency of the sensor. Quantum efficiency of back lit sensors can approach 90% but is more typically about 70% and front lit sensors are about 20%. A 20% Qe sensor effectively throws away 80% of the light falling on it. When sensors are "pushed" they really don't get more sensitive, they merely crank in more amplification (a combination of digital by shifting and analog by increasing gain). The base sensitivity is determined by the flux density and the number of photons necessary to saturate the sensor cell. So as you peer at your images, looking for the evidence of noise, just do it with the knowledge that each time iso sensitivity is doubled, you lose a bit of quantization level. A 100 iso base sensor that produces 12 "real" bits of information that is pushed to iso 16400 has only 6 bits left. The result is banding, maybe you don't see it because noise tends to obscure it, but it is there and can be measured.
Last edited by Bob; 18th November 2010 at 01:05.
I can't speak for M9 camera sales, but I can say that ANY of the Summilux M lenses are very hard to find new - production is not keeping up with demand.
If this is a flop, bring it on!
Regarding Leica, to quote Mark Twain, "The rumors of my death have been greatly exaggerated"
Actually, I think that applies to Pentax too.