RobertLynn wrote:
I like how people are comparing cameras that don't have full rawsamples out, and like "oh 1dx sucks better get a d4".
Do people even think before they type?
I know it's a gear forum, but I didn't think we were neurotic.
+1
D4 released and there are 5 topics already on the Canon sub-forum about people going: "oh noes, will teh D4 beats teh a$$ of teh 1DX, what is going to do ?!?!? OMFGKSDAFJAJ"
What I find most ironic that most of the discussion is probably carried out by people who either can't afford the D4/1DX or who won't buy it/consider buying it...
I'm a 5D II owner and I know my league (compared to 1D4,1Ds III, D3X/D3s etc)... PRICE WISE....
So I find all these discussions...futile..
PetKal wrote:
D4 is apparently supposed to AF quite nicely even with f/8 nominal aperture lens combos.
I think the Canon "f/8 birders" will have a bird now.
If only Nikon had a 800mm that would be my dream combo.
ausemmao wrote:
But as current technology stands, IQ isn't bound by QE, not by a long way. QE + shot noise bound only limits SNR at full well capacity, and yes, there read noise means much less. At every other luma value and especially in shadow areas, all the other sources of noise become more significant as they comprise a greater portion of the reading, until they become the same magnitude. There is absolutely a long way to go there.
This is the reason why you'd find it hard to tell a D3 from a D3s from a typical ISO 200 image, but wouldn't find it hard at all at 12800....Show more →
I don't know your background, but it is fundamentally wrong that QE/shot noise only influences noise around full well capacity. The QE related photon shot noise is the square root of the the number of captured photons. You will be surprised by how deep into the shadows this noise dominates.
At ISO 6400, mid gray level, the 5DII sensels capture on average about 150 photons. The QE related noise (photon shot noise) is the square root of that, about 12 photons. In comparison, the read noise is equivalent to ~ 3 photons.
4 Ev below mid gray at ISO 6400, the read noise will be about as large as the photon shot noise, and an improvement will be visible. Still, if the read noise is reduced to ~1 photon, the total noise will only be reduced from ~4.2 (sqrt(3*3 + 3*3)) to ~3.2 (sqrt(3*3 + 1*1))
Ralph Conway wrote:
But D3 offers the same IQ D700 does. And this is just a fast breath better in ISO 3.200 & 6.400 compared to what 5D MK II gives a shooter.
We'll have to agree to disagree here. My buddy shoots with the 5DII and I shoot with the D3, and there is a world of difference to me, much more than 1 stop, and I am not even getting into the noise which exists at base ISO in the 5DII shadows.
sperraglia wrote:
If only Nikon had a 800mm that would be my dream combo.
I think it is quite likely that within a couple of years Nikon will come up with an 800mm f/5.6 VR lens of their own.
At that point in time many Canon wildlife photographers will be faced with a real temptation to switch to Nikon, particularly so if their equivalent camera/lens offerings are cheaper and their pro grade cameras are better performing.
I have no loyalties to any brand whatsoever, and I have enjoyed seeing the aggressive and effective stance Nikon have taken in the past few years.
I also think Canon needs to feel the heat of some competitive force bearing on them. I sure hope 1DX performance will give to many of us a solid good reason to stay with the Canon system.
alundeb wrote:
I don't know your background, but it is fundamentally wrong that QE/shot noise only influences noise around full well capacity. The QE related photon shot noise is the square root of the the number of captured photons. You will be surprised by how deep into the shadows this noise dominates.
At ISO 6400, mid gray level, the 5DII sensels capture on average about 150 photons. The QE related noise (photon shot noise) is the square root of that, about 12 photons. In comparison, the read noise is equivalent to ~ 3 photons.
4 Ev below mid gray at ISO 6400, the read noise will be about as large as the photon shot noise, and an improvement will be visible. Still, if the read noise is reduced to ~1 photon, the total noise will only be reduced from ~4.2 (sqrt(3*3 + 3*3)) to ~3.2 (sqrt(3*3 + 1*1)) ...Show more →
Aside from the 3 e- read noise being the minimum read noise, (it is not constant for any sensor, and tends to rise with ISO), at full well, read noise doesn't matter - 2 3 e- vanishes compared to shot noise.
Mid grey, it starts mattering, in the shadow areas it really matters. Even at ISO 3200 you're looking at more like 20e- for read noise on a 5D2 (so that right there is 3 stops of DR lost compared to a constant 3e-). That's why a D7000 looks unquestionably better than a 5D at 3200, and more so the further you go, and why the D3s looks better than any other DSLR at high ISO. It's also why, provided the sensor structures can be reproduced properly, increasing resolution reduces image noise.
Edit: I need to check the figures some - the differences in the chain between light hitting pixels and a value being written to file between the cameras means they are measuring different things, and as such it may not be a fair comparison. (The D7000 for example sends digital values off the chip, so has no noise associated with information transfer, where the other cameras do).
carstenw wrote:
We'll have to agree to disagree here. My buddy shoots with the 5DII and I shoot with the D3, and there is a world of difference to me, much more than 1 stop, and I am not even getting into the noise which exists at base ISO in the 5DII shadows.
The funny thing about this: The 5D2 isn't actually much noisier than other cameras in the shadows. It's that the noise isn't randomly distributed that makes it objectionable.
What many people here are forgetting is the 1dx claims to have gains in ISO for there .jpg lesser for there RAW. Nikon on the other hand has the gains for there RAWs so if you want to use canon in camera processing and shoot .jpg then Canon may equal Nikons D3s in high ISO. As for the D4 it,s rummored to be 1 stop above the d3s. Regardless both cameras will be amazing and those who shoot them will be amazed yet the debates will still continue.
One thing which hasn't been mentioned here is that the Canon noise tends to have more colour in it than Nikon noise. This can make a perceptual difference which isn't necessarily reflected in pure test results.
^ 12mp vs 21.1mp. Everything seems to be more evident with the high resolution image files.
It's called detail. It's something Nikonian has yet to grasp it.
And speaking of ergonomics.... by the looks of the introduction of a joystick, Nikon is finally following Canon. That italian designer guy should stay with his day job. The design of the D4 is too fussy.
Buttons everywhere. 1dx and even the d3s had a cleaner implementation.
D4 reminds me of the dials and knobs in the old Michael Keaton's Batmobile. In the world that demands sleeker and streamline designs, WTF?!
ausemmao wrote:
Aside from the 3 e- read noise being the minimum read noise, (it is not constant for any sensor, and tends to rise with ISO)
Maybe just a typo on your part, but read noise tends to go down as ISO goes up. This is, after all, the only good reason to increase your ISO setting when you can no longer saturate the sensor!
ausemmao wrote:
Even at ISO 3200 you're looking at more like 20e- for read noise on a 5D2
The figures I've seen say just over 3 electrons (with no improvement beyond about ISO 1600). Where did you get 20 electrons from?
ausemmao wrote:
That's why a D7000 looks unquestionably better than a 5D at 3200
I would question that even if you really mean 5D (your previous sentence was about the 5D Mark II). And a D7000 doesn't look better than a 5D Mark II at ISO 3200, by almost any definition of better except perhaps visibility of faint pattern noise. It arguably looks better at ISO 100, of course, since it has far lower read noise than a 5D Mark II (or any Canon SLR) at base ISO; though obviously the D7000 still suffers from higher shot noise (i.e. noise above deep shadows) than a 5D Mark II.
Closer to the original topic, the D4 will disappoint people hoping for a stop of improvement at high but sensible ISOs. It might conceivably be a stop better, by some limited definition of better, at ISO 100,000, but how many people care? It may also have much cleaner shadows at base ISO, since the D3S had high read noise at base ISO, and a high base ISO. People will eventually care about this.
From ISO 3200 to ISO 12800, no Bayer-based FX-sensor camera will ever be "one stop" better than the D3S. We no longer have room to improve by one stop in this range.
carstenw wrote:
Err, 24MP D3x with a weaker AA filter? Anyway, that has nothing to do with colour vs. luminance noise when looking at 100% or anywhere near that.
Uhm... was referring to your comparison of your buddy shooting with a 5d2 and you with a d3.
carstenw wrote:
We'll have to agree to disagree here. My buddy shoots with the 5DII and I shoot with the D3, and there is a world of difference to me, much more than 1 stop, and I am not even getting into the noise which exists at base ISO in the 5DII shadows.
carstenw wrote:
One thing which hasn't been mentioned here is that the Canon noise tends to have more colour in it than Nikon noise. This can make a perceptual difference which isn't necessarily reflected in pure test results.
There's no real reason this shouldn't show up in tests (DxOMark shows it, for example).
The reason for this is that Canon used weaker colour filters in the Bayer array to boost the signal, thereby lowering luminance noise. (Nikon did this with the D3S too, compared to the D3; hence the higher chroma noise of the D3S at moderate ISOs.) After demosaicking, the colour needs to be boosted more accordingly, introducing more chroma noise.
The wild success of the 5D Mark II shows no-one cares about colour accuracy if they get a bit less high-ISO luminance noise in exchange for it. Of course, the manufacturers should have known that from the earlier success of certain E6 emulsions that flattered rather than revealed.
