Are the D3s of the world killing fast lenses?

[THORHAMMER]

hopefully they can give the lenses a little challenge. Id like to see some of the faster lenses go down a little in price.

 

[Iron Flatline]

...but you can use a Noctilux at f/1 in bright daylight with a film camera. You can use it at f/1 for very long exposures in bright daylight. What's new, apart from being able to use it at shutter speeds higher than 1/1000 (nominal)?

Not much - I guess I'm just pointing out that the way you can use gear is incremental, but it doesn't really change photography that much. Being able to shoot in extremely bright day light with a Nocti isn't that big a deal... so that goes for high ISO as well.

The lens of a camera is its bottleneck. It doesn't matter how much you can crank up the sharpness or the contrast in post, if the details didn't pass through the lens. It's like if you connect a portable music player to an amplifier and turn the volume all (or almost all) the way down on the music player. No matter how much you amplify the incoming signal, if you've got nothing to amplify, then all you get is noise.

RAW is considered to be the raw data - everything is there, you just need to fish out what you need and adjust it how you wish. But even RAW isn't completely raw. The real raw material is the world at the other end of the lens. Once it passes the first element, it ceases to be raw, and becomes an interpretation or rendition.

There's a lot of people who argue that a sensor still can't really resolve the potential sharpness of some lenses... but sharpness and contrast were once the gear-head yardsticks, just like speed (f-stop) or now things like pixelcount or ISO. Yes, once the image is captured, you can't really add in sharpness... but you can play with the resolution at the edges and give the appearance of sharpness - and that's often just as good.

 

[Garbz]

Does it really produce a signal with no noise? That's the first time i hear that.
In any case, what i meant was that whatever registers on the chip, first passes through the lens, obviously. And if you've got a cheap lens, or a cheap filter that renders some of the details unresolvable, then the quality of the chip doesn't really matter.

Regarding the range of ISO, i didn't know the D300 and the D3's lowest ISO was 200. And i thought that my 30D's 100 was too high...Does anyone know the reason for this?

Not noise as far as electronics goes anyway. As far as I understand it a lens just bends light through. Photos pass through in a way and hit the film plane on the other side. Now I don't think in this case it is photons coupled with some random noise. I believe all of this is generated by the sensor itself. So if you receive an order of magnitude photon less because of a lens which lets less light through or because less light is being generated IF and only if you can increase the sensor sensitivity without inducing added electrical noise then you still get a good picture. If there were something like light noise which is a concept I've never heard of then my guess is those 30second night time exposures (at ISO100 they are relatively free of electrical noise) would look far worse than they do, not to mention 6 hour photos on film which does not get noisy from long exposures.

As for the ISO200 bit I am not entirely 100% sure on this either. {scratch that I was in the middle of typing when i realised my reasoning didn't make sense} I'm open to explanations of why too. It must have something to do with the Analogue to Digital conversion.