Sun, May 8, 2005
I wasn't going to upgrade to Photoshop CS2 because I didn't think that I need it.
But that was before I found out about the new support for High Dynamic Range
(HDR) and 32 bits per channel floating point support. Michael Reichmann
over at The Luminous Landscape has an article talking
about the new "Merge to HDR". I was going to how current cameras only handle
6 stops and all (and what a stop is) but Michael does a good job of that in his article.
I'm excited about this feature because I think that it will enable me to capture images
that I haven't been able to before. I've often shot bracketed exposures with
the thought that I would merge them together by hand later. The results rarely are
as good as I imagined when I took the image. For instance, shooting a sunrise
of a mountain where the top of the mountain is in the sun but the rest is in shadow
can be very difficult. Having a tool that approaches this in a more methodical
way is exciting. On top of this, I'm excited by the possibilities of this combined
with panoramic photography. Wide angle and pano photography has a problem whereby
you are taking in so much of a scene that you are much more likely to have parts of
the image be very bright and others be very dark. One of the skills that I've
been developing is the ability to pick a scene where I won't run in to these problems.
But, as great as the new Photoshop might be, there are other parts of the digital
photography system that need to catch up. The first is on the display side.
While the data in a photograph might contain lots of dynamic range, today's displays
can't show it all. That is going to change over the next couple of years.
On the software side, there is plumbing being laid to support 32bpp through the Windows
desktop and to the video driver. (The 3D APIs and graphics cards are support
a lot of this already because it can be useful in 3D rendering). The displays
are going to get better also. Over the next couple of years, expect to see wide
gamut and HDR screens coming out. Sunnybrook is one of the companies that has
been pioneering this LED based
display technology. Where normal LCD or CRT displays can get a contrast
ratio of 300-400:1, they claim their LED technology can get 60,000:1. That means
the bright spots are brighter and the shadows are darker. I haven't seen one
of these displays myself, but I'm super excited by the technology.
The other side that needs to improve is on the image capture side. I have not
specific knowledge of any plans from any camera manufacturers, but it is obvious that
the megapixel race is slowing down. We are reaching the limits of the glass.
Another example is the fact that the new prosumer level Nikon SLRs are still only
6 megapixel. It is clear that the camera makers are going to compete on more
than just the number of pixels. They will start competing on the quality
of those pixels. This is going to be very exciting. We are entering a world
where digital cameras are not just matching the abilities of film cameras, but
they are on the verge of surpassing them. A camera that can capture higher dynamic
range and a wider gamut? Sign me up!
The big question in my mind is the print making side of the equation. When I
print my photos, I usually send them out (WHCC or WCI)
to be printed on a Chromira printer. This is a printer that uses LEDs to expose
regular photo paper. This photo paper is then printed traditionally. The
gamut of these devices (and that of the ink jet processes) can be larger than current
displays but is still fairly limited. The dynamic range of all of these output
devices is pretty low compared to what can be done with an emissive display device.
The bottom line is that the printing technology isn't keeping up with the rest of
the system. What does this mean in the long term? I'm not sure but we
will start to see a vast divergence between the quality of an image when displayed
on a good monitor compared to they quality of an image when printed.
It certainly is an exciting time for photography...