AVS Forum banner
Status
Not open for further replies.
1 - 20 of 21 Posts

·
Registered
Joined
·
1,425 Posts
Discussion Starter · #1 ·
Imagine that you are a camera that is mounted on a "x/y" positioning device. Your field of view is an area about the size of big grapefruit. The positioning device can point you at any part of the screen. You know where the image should be and you have control of the convergence and geometry circuits. Which projector would be most compatable with this situation? Only the part of the screen that the camera is "looking' at moves for geometry/convergence adjustments. Not like when a "skew" adjustment is made where the whole screen moves. Is there such a thing?
 

·
Registered
Joined
·
18,503 Posts
I know the NEC XG and PG series has 100 points of convergence for each color. That's far more than any other set.


But I've never seen the NEC ACAT camera, it was a rare beast. THe later MCAT used 16 points that you converged manually, then the set mathematically calcualated the other points. THat worked quite well.


Curt
 

·
Registered
Joined
·
8,079 Posts
I think Robro's going to tell you about the NEC GP-3000/5000 models ;)
 

·
Registered
Joined
·
4,916 Posts
TSE,

I'm not exactly sure what you are getting at here but I can tell you this. The G90 has seperate side key and pin control for each of the colors including G. So you can key and pin the top, bottom, left and right. And these controls are available as a course and fine adjustment.

The XG comes close to this with seperate side key and pin for R & B but lacks on the G which relys on key/key balance and pin/pin balance combinations.

The G has zone/point for 24? zones and the XH has a lot more zone control but is problematic when used to extensively, especially on the G control.


Terry
 

·
Registered
Joined
·
1,425 Posts
Discussion Starter · #5 ·
I've seen old Mitsubishi advertisements that mentioned 200,000 points of adjustment. That had to be interpolated points. But, still. Which projector would work best with a camera system that could only "see" a grapefruit size part of the screen at a time?
 

·
Registered
Joined
·
18,503 Posts
Never having worked with a G 90 or Mits (yet, huh Terry?;)), I'd say the NEC.
 

·
Registered
Joined
·
1,425 Posts
Discussion Starter · #7 ·
Quote:
Originally Posted by Chuchuf
TSE,

I'm not exactly sure what you are getting at here but I can tell you this. The G90 has seperate side key and pin control for each of the colors including G. So you can key and pin the top, bottom, left and right. And these controls are available as a course and fine adjustment.

The XG comes close to this with seperate side key and pin for R & B but lacks on the G which relys on key/key balance and pin/pin balance combinations.

The G has zone/point for 24? zones and the XH has a lot more zone control but is problematic when used to extensively, especially on the G control.


Terry
What I'm looking for is information on projectors that make adjustments only at one place on the screen. Not like when a "key" adjustment is made and one whole side of the picture moves. Imagine that you are a camera and can only see a grapefruit size part of the screen. You have control of that part only. When you make adjustments that is the only part of the screen that moves.
 

·
Registered
Joined
·
18,503 Posts
I know exactly what you mean. THe NEC XG fits the bill, not sure on the G 90 as I've never seen one.

Yet.

Until next week.

Yahoo!


:)
 

·
Registered
Joined
·
217 Posts
Quote:
Originally Posted by Chuchuf
TSE,

I'm not exactly sure what you are getting at here but I can tell you this. The G90 has seperate side key and pin control for each of the colors including G. So you can key and pin the top, bottom, left and right. And these controls are available as a course and fine adjustment.
Can't you do this manualy with a Sony 10XX?
 

·
Registered
Joined
·
2,041 Posts
You need a unit with as many point locations or zones as you can get.

The GPs have about 160 points, but they are low brightness low reolution PJs compared to other PJs with the same tube size.

Don't know how many point locations the Mitsus have.

How many would you need?

Roland
 

·
Registered
Joined
·
6,125 Posts
The nec's will do it but only to a point. If point convergence is used extensively, stability suffers. Is there a problem that you are trying to resolve? Watta-u-gut, a painted wall for a screen with a bump in the corner?


Chip
 

·
Registered
Joined
·
1,425 Posts
Discussion Starter · #12 ·
I'm trying to think of a way to do automatic geometry (as well as convergence) with a camera that is mounted to an X/Y positioning device. For the camera to have enough resolution to do precision alignment it has a small field of view and cannot "look" at a large part of the screen. So the standard way of setting the geometry with the analog keystones, pincusions, bows, and skews won't work because the camera doesn't see every part of the screen that is changing while the adjustment is being made. I'd like for the camera to see only the one point that is being changed.
 

·
Registered
Joined
·
2,041 Posts
You would want to see at least one neighbour point for geometry reference...

Roland
 

·
Registered
Joined
·
27 Posts
Quote:
Originally Posted by tse
I've seen old Mitsubishi advertisements that mentioned 200,000 points of adjustment. That had to be interpolated points. But, still. Which projector would work best with a camera system that could only "see" a grapefruit size part of the screen at a time?
The number 200K sounds to me more like the size of the memory "window" they are dealing with for the math interpolation. IE. it fit into a 256K by 8 memory chip. You were not under the impression that you manually step thru 1000 horz zones and 200 vertical zones, are you? In math the zone itself is probably like 1000 "points" wide or so at a time. At 13 by 5, the zone would still cover 3076 "points" if if you started with 200K. Marketing figured out a clever way to say the same thing to sound like they were better than everybody else that doesn't say it the same way.


I think a more important point would be to use more resolution like a 12 bit dac instead of an 8 bit dac like alot of people use. You might get a more uniform raster

and less "line pairing" type look.


My PJ does 13 by 5.

My PJ does 200K points.

My HT amp goes to 11.


Sorry for the comment and no real data transfer.
 

·
Registered
Joined
·
1,425 Posts
Discussion Starter · #16 ·
Quote:
Originally Posted by albertsacoustics
Hello

The Mitsubishi 1281 has the most adjustments. All the adjustments like the NEC XG´s

and up to 13x15 point-convergence



Piet
I suspect that you are right. Do you know how many point-convergence zones the NEC's have?


Thanks
 

·
Registered
Joined
·
1,425 Posts
Discussion Starter · #17 ·
Here's a part of the Mitsubishi advertisement. They are claiming 200,000 points of adjustment. The grid is 13 x 15 but there must be alot of interpolation going on to get 200K points.
 

·
Registered
Joined
·
6,125 Posts
tse, is this some top secret area 51 experiment you are working on ;)? Perhaps if you shed a little light on the subject, some brainiack here could help you in your quest for grapefruit size zone convergence :cool:


Chip
 

·
Registered
Joined
·
2,041 Posts
Its just 13*15 adjustable points and a digital fliter with 32-fold oversampling.

Roland

P.S.:

The GPs have 17*13 point locations, but I do not know out of my mind how high their oversampling factor is...
 

·
Registered
Joined
·
1,425 Posts
Discussion Starter · #20 ·
And with the GPs it is possible to do all the geometry adjustments with only the point adjustments? Not using skews, pins, keys, etc?
 
1 - 20 of 21 Posts
Status
Not open for further replies.
Top