On Oct 25, 2006, at 3:47 AM, 360Precision Email List wrote:
On 25 Oct 2006, at 05:44, email@hidden wrote:
I guess it *has* been over a year since I explained the idea of a
"multi-resolution reflection sphere" to you (in Venice). If the
entire sphere was to have the resolution that the front glass or
hood of the car can sample reflections from (say, in a leafy
environment), then it would be a 2.7 Gigapixel pano! (Try to stuff
that into a renderdrive!) This technique for putting "resolution
where it's needed" isn't new to experienced 3D professionals, but
when I explain it to photographers it can take a really long time
for them to wrap their heads around it. Hence I can understand why
some studios simply tell the photogs to "shoot 10,000 pixels
across or use a Spheron" (as I think you said), and then they fake
whatever they are missing in retouching.
The problem with this approach is for animations or object flyby
(object type) movies retouching 1500 frames for example is not the
cheapest method.
I think you have the concepts of *animation* and *image for print*
mixed up. ;-)
The agencies our customers use can charge thousands of pounds per
day so the cheapest method is to shoot ultra-highres panos. I
recently tested a Hasselblad H2 with the new PhaseOne P45 (39mp)
digital back and shooting 35,000 x 17,500px panos is no more
difficult than shooting a multi-row with a D70. But the results are
spectacular and can then be used as is.
Yes, I'm sure the resulting pano would be spectacular, but that is
still less than half the effective resolution of what our multi-
resolution spheres have and wastes immense amounts of resources (both
front and back-end) to put pixels where the shot doesn't require it.
You could make the argument that having continuous resolution all
around the sphere means you can render at high res from all sides,
but the angle of the shots, and therefore the general angle of flat
surface reflections, are determined by the photographer when they
shoot the background plates for the print piece (or are we back to
animation again?). It really makes sense once you actually start
doing the 3D end of things and working at those resolutions, as
opposed to reading about it in PDN. (By the way, for those in the US,
there is an interesting interview on all of this in the November
issue of PDN! ;-) )
Besides, my "work" HDR panos are not typically shot using an 8mm
(and even then it's 8k x 4k), unless it's on a set where the union
lighting crew can't spare more than 2 minutes after a shot has
wrapped, or outdoors under fast fading light. My "fun" HDR panos,
and the ones I sell on some websites *are* shot this way, and it
seems to be more than enough for most consumers, even for print
needs. It's all about angle of incidence and how well someone can
use their 3D tools. Heck, some people are still using chrome balls
and getting decent work done!
Fair enough but you always seem to pushing the point that using
your quick and cheap set-up is enough or the industry standard,
when in-fact it's not even close.
Huh? Let me know where I've said my "quick and cheap set-up is enough
or the industry standard" for some specific project or even industry
- I'm curious as to when I've given that kind definite consultation
without seeing the project. If I "seem to push a point," then it
would be my general concept that experience should be your guide to
determine what you need, and if you have no experience then get
started and don't think you are limited from the start. There are
many, many ways to capture a full-spherical HDR pano for 3D IBL, and
it is the 3D artist and the project requirements that dictate what to
use and what is sufficient. Or is this somehow getting lost in my
enthusiasm?
There's a huge gap between the typical 3D use of HDR spheres and
the high-end use by agencies in major print campaigns. The biggest
problem is art directors are now 3D/HDR savvy and they can spot low-
res/poor work a mile away. So what was once acceptable even 12-18
months ago isn't any longer if you still want to get decent jobs as
a photographer.
What you see as a "problem" (art directors understanding 3D and HDR
IBL) is what I've been promoting and facilitating for 5 years now.
I'm glad that they can spot low-res/ poor work a mile away - it makes
it easier for me and my clients to get those jobs! The only thing
standing in the way is how so many people read a few articles or
visit a website and think they know enough about how 3D and
photography can be integrated with this "new panoramic image-based-
lighting thing," and end up talking out of their behind to the great
detriment of those who are actually interested in *working* with this
technology. I mean, the 3D for print market isn't new, and printing
presses (and film imagers) have not made exponential leaps in
resolution in the 18 months, have they?
Perhaps I'm just a curmudgeon about this, since my first "3D/
photography composite for print" job was put on national billboards
(US) back in 1993, and I did a 13k x 10k 3D rendering for a japanese
backlit billboard in 1996. (That took everything that Alias and
Scitex could throw us to get it imaged!) I've stuck with this,
developing the technology and techniques of high-res 3D for print,
despite many years of misunderstanding and wild claims by "non-
practicing experts." Plus, I try very hard to get more people into
this market and excited about exploring it from all angles (pun
intended). To use the most expensive equipment reviewed on
luminouslandscape.com sounds great and sexy and I'm sure it has
plenty of applications, but it is not a measure of what is
"acceptable" once the people involved start to understand the process.
-Mark
_______________________________________________
Do not post admin requests to the list. They will be ignored.
QuickTime-VR mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/quicktime-vr/email@hidden
