Live and Relax

The 14 bit raw file is new feature introduced by Nikon when they introduce the D3 and D300.  The D700 also have this feature built in.  Is the 14 bit raw file have any advantage over 12 bit?  Let’s do some trial to find out.  I have set up the D700 with 70-200 to shoot a gradient pattern displayed on screen:

(more…)

The ISO Lo 0.3 to 1 setting in D700 is said to be equivalent to ISO 160 to 100.  But equivalent means not the same.

The usual way for the camera to adjust the ISO value is by controlling the gain of amplifier of signal comming out from the sensor before analog to digital conversion.   If our sensor have unlimited dynamic range, there is no problem.  But as our sensor have a limited headroom.  A photo site is limited by a fixed quantity of photon it could receive before it is full.  If more photo is received after a photo site is full, it lost track of how many more photon is received and we got highligt clipping.

Because of that, when using ISO lower than the base ISO value, the base ISO value of a sensor is the ISO level at which the signal comming out from the sensor to do need an amplification (the amplifier gain is 1 in this case), we run the risk of higlight clipping.

So, is the ISO 200 the base ISO value of the D700?  I have tried to shoot the a scence with ISO 200 with manual exposure control so that part of the image have highlight clipping.  For the same scence with the same expsosure value at ISO Lo 1, I observed that the image have exactly the same highlight clipping area.  Therefore, it could be concluded that the base ISO value of the D700 is 200.

What is the consequence?  That would means that when in situation that needs longer exposure time, it would be better use the ND filter to acheive the effect than using this ISO Lo 1 setting.  If using this setting is not avoidable, we should be paying more attention to the highlight clipping problem.

A simple thing to test, just turn the auto image size on the D700 to OFF and plug in the Nikkor 12-24.  You got a lot of black border (It is just not vignette at this stage) when at 12mm.

(more…)

Ever wonder why the middle grey card is 18% grey.  Why middle is not equal to 50%?  This have something to do with our eyes non-linear response to the light power it receive.  Something like the following chart:

The Y axis is the perceived lightness and the X axis the the light power our eye receive. The CIE mathematical representation of this is:

L ^* = 116(Y / Y_n)^{1 / 3} − 16

Y/Yn is the normalized luminance(light power) and L is the perceived lightness.

You could take a look at the following from wikipedia:

http://en.wikipedia.org/wiki/Lightness_(color)

Update on Nov 23

Add review link to dcresource.

http://www.dcresource.com/reviews/panasonic/dmc_g1-review/index.shtml

Update on Sep 17
Am I missing something?  Reading thru the detail specification of the G1 and I cannot find the movie recording mode! When the movie taking capability is becoming standard on DSLR (we see this feature in both Nikon’s D90 and Canon’s 5D Mark II), we missed this in the G1.  What’s in the mind of our Pansonic engineers and product designers?

Finaly, we see something more creative in product design other than the mega pixel race.  The first micro 4/3 system camera is announced.   From an operation view point, it is a true alternative choice other than the bulkier and noiser DSLR design.  Of course, we have to wait till how these class of camera really perform in the real world.  But sure, there is something to be expected there than the next DSLR camera comming from Nikon, Canon or whatever.  They are so so boring.
The only down size I could guess from the use of EVF only design is that this G1 will not perform well at low light.  In some really dark situation, picture composition is not possible as the EVF will be just dark.  Other than that it would be really good choice for street and news photos where is lighter, smaller and quieter camera is important.  You could find some more information about the Panasonic G1 in imaging resource:

http://www.imaging-resource.com/PRODS/DMCG1/DMCG1A.HTM

Just in case you are not aware, there is a free software to help calibrate the color displayed on you monitor:

http://www.hex2bit.com/products/product_mcw.asp

I don’t think it could be a replacement of a color calibration device for more serious work but it do help make the display on my monitor less “cool”.

I don’t have an english version of this article as there is already very good material in the internet.  You could see one of it here:

http://www.luminous-landscape.com/tutorials/understanding-series/understanding-mtf.shtml

In order for the sensor in digital camera able to record color images, the photosites on the sensor need to have the ability to recognize the color in the first place.  In general, this is done by placing a color filter on top of the photosites so that each photosite could only recognize one of the primary colors. There are 2 different methods in arranging the color sensitive photosites:

The first one is each pixel is constructed as 3 overlaping layers of photosite each corresponding to a different primary color.  At present, only Foveon produced such kind of sensor and Sigma is the only camera manufacturer that applied this type of sensor in their products. 

Another method is to arrange the red, green and blue photosites alternatively (Bayer Pattern) on a single layer as follow:

(more…)