dmitriyk

<i>Dmitry's untrue statements are, I believe, based on his lack of understanding about

lenses for UV and IR. He may want to read this thread: http://www.photo.net/bboard/q-

and-a-fetch-msg?msg_id=009xUN</i><br>

<br>

Oh geez. You had to go there, didn't you?<br>

<br>

I try to not do this a lot, but this time I'm genuinely curious. Which of my statements is

untrue? To recap:<br>

<ul>

<li>Your fearsome lens and filter collection cannot be used as a justification for saying

that Hoya digital filters block IR and UV</li>

<li>Regular dSLR lenses transmit 0 UV</li>

<li>IR filter packs reduce recorded IR by many EVs</li>

<li>The non-transmission of UV and the relatively low IR sensitivity results in a sensor

that is well-suited for visible-light photography without use of IR and hot-mirror filters</

li>

</ul>

<br>

I ask this, honestly, not to be an ass, but because I've read the linked thread and the

Nærfoto article linked within, and I just don't see how the above statements are untrue.

Also, instead of throwing out broad generalizations, I'd like you to point out specific

untrue statements made by me in the course of this discussion.<br>

Do note that there is a difference between an untrue statement and speculation.

Vivek: yes I'm sure your filter and lens collection is quite impressive, but that still doesn't

justify your assertion that these "digital" filters block IR and UV.

Lex: of course the D70 is used for this kind of photography; it was actually part of the

reason I bought it. But Bjørn has to use quartz lenses for UV and IR filters (that push

exposure times down by many stops) for IR photography. In fact, if I remember correctly,

his main IR workhorse is a D1 that had its IR filter removed.

And while the D70 is obviously sensitive to IR and UV, I haven't seen anything to indicate

that this sensitivity presents any issues for regular visible-light photography that would

require additional filtration from hot-mirror and UV filters.

<i>First, the digital filter is a real deal. A DSLR like D70 has a very poor filter infront of its

CCD sensor. This makes it very very sensitive to UV and IR. With normal lenses it affects

the color balancing big time. If you use the Hoya filter you mention, the unwanted UV/IR

light will be minimized.</i><br>

<br>

I don't buy it. First of all, unless you're using a UV-Nikkor (or another quartz lens), you're

going to get approximately 0 UV light reaching the sensor. Lenses, as it turns out, make

the best UV filters. Second of all, something makes me doubt these filters block IR. A

respectable manufacturer like Hoya would not label and sell a hot-mirror filter as a UV or

protection filter. Not only is additional IR filtration not needed on dSLRs, but it's simply

not economical - hot-mirror filters are a lot more expensive. So unless you can provide

some actual evidence that these digital filters block invisible light, I'm going to have to go

with Gary's theory on this one.

<i>There are only three color sensors in the human eye therefore a 4-primary color

system has one too-many degrees of freedom and tends to not be well defined. Really it's

just a matter of math.</i><br>

<br>

*ahem*<br>

<br>

To get the math out of the way, the fact that color space is a three-dimensional vector

space (with a basis set of R, G, and B), does not preclude it from being well-

defined by a four-element spanning set of C, M, Y, and K.<br>

<br>

<i>yeah, and that math has nothing to do with the eye's sensors. Look up "additive color"

versus "subtractive color" sometime.</i><br>

<br>

The subtractive color model still only has three elements - C, M, and Y. The reason we

need K is because in the real world C, M, and Y aren't perfectly subtractive. C does not

absorb all of R, M

does not absorb all of G, and Y does not absorb all the B. We just can't make inks that

well. So we need a fourth color to completely absorb light, and that's why there's K. It has

nothing to do with the dimension of color space or additive vs. subtractive color theory.

It's just a hacked-together engineering solution. :)<br>

<br>

At any rate, Mr. Rodney is completely correct. I'd just like to add that we also work in

wider-gamut color spaces because it reduces round-off errors, keeping nasty things like

posterization at bay.

If you convert from a wide-gamut profile (like AdobeRGB) to a narrow-gamut profile (like

sRGB), the colors that are not in the narrow profile gamut will be altered, as the narrow

profile just can't accommodate them. It's not something that should impact a typical

photograph in a meaningful way, but the potential is there.

Furthermore, if you convert from a narrow-gamut profile to a wide-gamut profile and your

bit depth is too low, you may get posterization. This error should be pretty small, and the

easy workaround is to work in 16-bit mode.

For purposes of raw conversion, you can consider the camera space to be slightly wider

than AdobeRGB with about 12 bits of depth. Of course, this varies from camera to

camera.

Neither of these two scenarios is that big a deal, so for the most part the colors will look

the same. It's just something you should be aware of when you start digging deeper into

color theory.

The ISO speed is the third exposure parameter which you can adjust on digital cameras (or

film cameras by swapping film). This setting changes the sensitivity of the sensor - if you

increase it, the sensor will require less light for the same exposure. That is, if you

increase the ISO sensitivity by one stop, you can increase your shutter speed by one stop.

One stop of ISO sensitivity is a doubling of the ISO number - 100, 200, 400, 800, 1600,

3200, etc...

Noise increases as the ISO is increased. Smaller digital cameras especially suffer from this.

Experiment with your camera and decide what's acceptable to you.

The D2Hs is designed for newspaper photographers. For them, 4MP is already overkill.

Anything more just slows down the camera and reduces storage capacity. If you want 6MP

at 8fps, get a D2X and use it in 2x crop mode.

But the D70s is fairly underwhelming (not that I as a D70 owner care, really). Here's

hoping we'll see a D200/D80 soon.

The D70s adds a wired remote shutter release, a new battery, and 2" LCD screen (vs. a 1.8"

screen).

Still 6MP, still no mirror-lockup.

Meh.

Just remember that what I did yesterday was very, very stupid. Don't expect your camera to make it through something like that unscathed. I just wanted to point out that the D70 is sealed better than your average point-and-shoot.

You can take a D70 out in the rain, just don't be stupid about it.

I was flying internationally with my D70 a month ago and (against my protests) had to check my carryon with the camera bag inside. When I pulled the camera out, the UV filter on the lens (an expensive B+W MRC one) was shattered. It must have taken one hell of an impact to do that, but both the body and the lens were fine.

Yesterday, I also spent 15 minutes shooting in pouring rain. There was water and snow pooling up in the various cracks and crevices of the camera, but it didn't miss a beat.

For a prosumer camera, it seems to be pretty damn durable.

Here's an action shot:<div></div>