Accused of "manipulating" your shot?

I agree that more noise would be a problem for Foveon or any other sensor, I just think its premature to say that Foveon's design is inherently noisy.  I also wouldn't say that sensitivity it less important in UW still.  Perhaps for macro.  Good sensitivity implies low noise.

The Foveon sensor is inherently more noisy compared to a conventional silicon sensor. That's really in the physics of how these things work. It is perhap a little speculative to say they will never figure out a way to compensate for it. I don't think they can, but that possibility can't be ruled out.

Sensitivity is less important in UW still photography than other applications. How many still photos have you taken at above ASA 100. I've never done it. I'm not aware of any good UW photos taken with highspeed film or taken with a digital camera at a high ASA setting. The prefered method is to bring a big enough strobe to provide the needed light.

I mistakenly thought that Bayer filtering was achieved through a difraction grid rather a passive filtering array, so I see now why you draw an analogy between the eye and the Bayer CCD. Your spectral response plot of the cones is considerably different than I am used to seeing, though. I don't think actually impacts any theory (but sure would effect the implementation!). I actually understand that violet can be made to work. It's just glossed over for the most part. You're right, the "blue sensor" could really be violet. There are many ways it could be done.

Regarding sensitivity and film speed, I agree that overwhelming force is desirable for macro. To the extent that faster films could help DOF, they would be desirable. The problem is that you can't separate film speed from other factors that make a film desirable, and many photographers believe that shooting high speed film won't produce the results they desire, right or wrong. You won't see and good UW photos using 400 speed film if no one is shooting it. Any mixed light shot is different entirely.

My UW experience is largely with video, although I'm setting up a D100. High sensitivity enables a very wide white balance range in both video and digital still. This only comes into play in ambient or mixed light shooting. Fact is, sensitivity is one of the things that makes the VX-2000 such a great UW camera. Depending on what you do, this may not be important.

I understand why you claim Foveon is inherently noisier but I'm just not sure of that. Still, I'm as skeptical as you about their technical viability. I'm not even sure you would prefer Foveon sensors even if they could solve the noise problems they reportedly have. They good news is that we have great imagers and great cameras getting better all the time. If Foveon succeeds, even better.

Wow, this is getting serious now guy's... We are getting into doppler-shift!

How many still photos have you taken at above ASA 100. I've never done it. I'm not aware of any good UW photos taken with highspeed film or taken with a digital camera at a high ASA setting. The prefered method is to bring a big enough strobe to provide the needed light.

Herb,

This is slightly the wrong way around.

I regularly shoot film at high ASA. Colour to 1000 and BW to 1600, you would rarely require a strobe at all and why it makes carrying a 3rd camera practicle.

The quality of the shots are reduced at higher speeds through grain but the sharpness remains giving an overall balance. This technique is especially good for old ship wrecks and silver things like barracuda and even sharks.

The encosed shot was taken using a Nik V down at 12m with settings of 1/500 @ f11 without a stobe using the Fuji 1600 BW film.

A digital unit could not cope with the conditions as all of the graduation would be lost in black umbra. Shadows and black graduations are where pixels fear to treadeth...

Thanks Bob, I didn't know that.

I agree the consumer digicams would have no chance because of their limited dynamic range. Do you know how well one of the DSLR would do?

I think they are exactly the same...

If you look into the frame Eric took of the photopro and try and map the graduation from arms to abdomon. Sorry Eric, only example to hand!

This I think is why two strobes are a must in general digitography.

So to sum up...

Sensitivity is not important but dynamic range is, and all digital cameras have exactly the same dynamic range.

Sensitivity, noise, and dynamic range are all interrelated, and there are big differences between digital SLR's and digital P&S in this regard. In the video world this is one of the biggest differentiators between consumer and pro lines.

Is it also true that all slide scanners have exactly the same dynamic range? I think not.

Incidently, "no chance" of what?

So to sum up...

 

Incidently, "no chance" of what?

No chance of getting the details in the shadows shown in Bob's photo on a shot that also includes a bright sun, 8 bits of dynamic range is not enough. My question was whether it is possible with a DSLR that has closer to 12 bits of dynamic range.

Yes, I think so. I don't know the exact amount of bits or stops that a DSLR can reproduce, but I'm sure it's more than a consumer digicam. Heck, it's probably in one of the reviews somewhere on depreview.com

DSLR's also have MUCH less noise at higher ISO's than consumer digitams. The S2 actually takes very good pix at 400 and 800 although I don't use that unless I really have to. Same w/ the D60.

Cheers

James

Haven't seen any tests confirming the actual dynamic range of the imagers, but the D1 series and D100 have 12 bit ADC's. I'm confident the Fujis and Canons do as well. The VX2000 has 12-bit ADC's that give it great white balance range.

One of the big things when the D1 first came out was its large imager and its 12 bit ADC's.

Yeah, but the 12 bits only refers to colour I think!

I don't think I understand this comment. It is 12 bits per channel RGB. There is no separate luminance channel.

Maybe that's why!