Native Resolution Digital Cinema - Page 2 - AVS Forum
Forum Jump: 
Closed Thread
 
Thread Tools
post #31 of 33 Old 03-12-2001, 07:45 AM
Advanced Member
 
Huckster's Avatar
 
Join Date: Jan 2001
Posts: 872
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Mr.Wiggles,

And not reading someone's post is just as dangerous.
Quote:
<font face="Verdana, Arial" size="2">The current crop of movies being shown on DLP Cinema screens is stored as 10 bit 4:2:2. Using 10 bits in the output path would leave no space for error terms. Using a 14-bit output path allows gamma correction and color space conversion to have some over/underflow space and should allow smoother color gradients and better black level detail.</font>
Ever see this before? It was in my second post. Does it ask for higher bit depth INPUT sources? No. Does it suggest higher than 10-bit data (or output) paths? Yes.

So all this disagreement was caused by *me* not being clear on my first post (the last one you read) about using more than 10 bits in the data path, not the input source? Geesh.

Tom,

It is great to hear that this projector is not using a 10-bit data path. Since you mentioned 10-bits per channel, without mentioning a larger data path, in your first post, I incorrectly jumped to the over/underflow problems. Sorry. It seems that JVC's engineers agree that over/underflow is a concern and have expanded the data path beyond 10-bits to accomodate these problems. Excellent! When can I see one in the local cineplex?

------------------

Huck

Huck
Huckster is offline  
Sponsored Links
Advertisement
 
post #32 of 33 Old 03-12-2001, 10:46 PM
Advanced Member
 
Gary P. Gibbs's Avatar
 
Join Date: Jun 2000
Posts: 538
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
This brings up an interesting discussion concerning what we need in terms of A/D bit resolution and manipulation, etc. This is similar to the old CD quality discussion: Is 16 bit 44.1 kHz enough. The way I see we should work against two end points. The minimum end point concerns the transfer of the information that we have (e.g. 10 bit color space) all the way to the screen. The other end point is the minimum resolution that our eyes can perceive. In other words what is the best our eyes can do.

Minimum:
The minimum bit resolution that we should use in our projectors is that which will not loose quality from the stored data format all the way to presentation. This includes any scaling, contrast, brightness, or color space transformations.

Maximum:
Now for the maximum necessary based on performance of the eye (which has nothing to do with the source). I have been reading a bit on this for fun and according to HP's book "Using Computer Color Effectively" and was surprised to see that the range of light visibility of the eye is ~10^8 ! From 10^-3 lux (starlight night) to 10^5 lux (bright sun light). Granted this dynamic range (160 dB) is accomplished with a combination of Iris movement and cone/rod sensitivity. We would not want our projector to present images from 10^-3 to 10^5 lux, but I think what we want for any given scene is to assume a somewhat fixed Iris and let our cones/rods sense the difference. I think the ideal projector would have contrast ratios and addressable brightness (resolutions) for each of the R, G, and B that is equal to or greater then our cone/rod sensitivity. The big question which I have yet to see answered but I am sure has been tested and presented in the appropriate journals is: What is the dynamic range and resolution of the cones and rods in a typical eye? What are the differences in those values for black and white images versus color images?

So the best we should shoot for is a projector that has a dynamic range and resolution better than our rods and cones. From the above referenced book it also presents a bit of work on angular sensitivity of the eye. For RGB colors (depending on the test method) the eye has a resolution of ~0.2 - 0.1 degrees. For black and white it is something like 0.05 degrees. Thus we can determine what resolution projector would produce this angular resolution at give screen dimensions and viewing distances.

Something to think about http://www.avsforum.com/ubb/smile.gif




------------------
Gary

STOP DVI/HDCP!
DVI/HDCP! ~= HD-DIVX!!!
DO NOT SUPPORT JVC or anyone else who supports this!

Gary
Gary P. Gibbs is offline  
post #33 of 33 Old 03-16-2001, 10:15 PM
Advanced Member
 
KFung's Avatar
 
Join Date: Apr 2000
Location: Ottawa, Ontario, Canada
Posts: 879
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Gary,

Good point about perceptual limits on resolution and contrast. Actually, while the dynamic range of the eye is enormous (10^8! like you say), we can only perceive a contrast ratio of about 100:1 in any given picture. Projectors need a larger contrast ratio than that because the picture shown on the screen often doesn't use the entire dynamic range of the projector. How much is enough? Possibly 1000:1 (full-on full-off) is, I'm sure there's some kind of SMPTE study out there on this.

Resolution-wise I think you're bang on, it's around 1/60 of a degree. That equates to some pretty large numbers, something like 2k*2k at one screen height away from the screen.

Regards,

Kam Fung
KFung is offline  
Closed Thread Digital Hi-End Projectors - $3,000+ USD MSRP

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off