AVS Forum banner
Status
Not open for further replies.
1 - 20 of 28 Posts

·
Registered
Joined
·
6,988 Posts
Discussion Starter · #1 ·
When determining CRT throw distance, should it be to the deepest part of the curve, or to the outer, closer edges?
 

·
Registered
Joined
·
6,988 Posts
Discussion Starter · #3 ·
NEC XG135LC. Setting it to the deepest part of the curve makes sense. I'd have to be conscious of not moving it closer to maximize raster size, since the edges would already be closer.
 

·
Registered
Joined
·
4,916 Posts
Robert,

I think you get the idea. While the center of the screen is the throw distance, what you have to be careful of is that the corners can be up to 11" closer than the center to the PJ. This in turn could lead to you being a bit to close if you are using max raster.


Terry
 

·
Registered
Joined
·
6,988 Posts
Discussion Starter · #6 ·
I'm thinking maybe 6 inches for the horizontal curve, and 3 inches for the vertical curve, for a 102 inch wide screen.
 

·
Registered
Joined
·
4,916 Posts
That will give you a total of 9" center to corner.


Terry
 

·
Registered
Joined
·
13,003 Posts
Robert,

Terry would know better if that is to much. Is that 4:3 or16:9?


Be sure you check out Vince's Torus Lite thread . I think the G70 lenses and the XG LC lenses are close if not the same. Looking at his numbers, you may be close to the limit.


Ericglo
 

·
Registered
Joined
·
4,916 Posts
XG/LC and G70 are the same exact lens.

I think Vince's problems had to do with pin and bow if its the fellow I'm thinking about who set up a G70 on a Torus in Savannah GA. The XG's have a bit better range of control for the bow and pin than the G70's.


Terry
 

·
Registered
Joined
·
6,988 Posts
Discussion Starter · #10 ·
It sounds like I'm in the ballpark. It would be a 16:9 screen, and I'm thinking of changing the curves to 6.5 inch deep horizontal, and 2 1/16 inch vertical.
 

·
Registered
Joined
·
502 Posts
Robert, I've just finished mine at 96" wide and horizontal curves at 5" deep, 1.75" on the vertical. I don't know if I'd go much more than 5.5", I think even at 5" you could get your 102" screen.

Color shift will be minimal if you adjust for it in the xg's menu. I found without adjusting your white balance then color shift was unnacceptable. My white balance is running at about 60%.

At 96" wide my contrast is at 50% and brightness at 55% compared to the 65/60 on the 80" parkland screen. Very punchy picture too!
 

·
Registered
Joined
·
13,003 Posts
I have been giving this some thought for awhile now. The curves will be somewhat determined by the gain. The higher the gain the deeper the curves. Since most are using the Pearl, you could probably follow the curves that they are using. In Vince's Torus lite thread, he seemed to be fine with the vertical curve of 1.75". The horizontal curve seemed to be a little shallow and hotspoted at 3.5".


Eric
 

·
Registered
Joined
·
6,988 Posts
Discussion Starter · #13 ·
So the tradeoff is hotspotting if the curve is too shallow, and loss of focus in the corners if the curve is too deep, yes? Vince's original curves were too deep, but he also had a significantly smaller screen than what I'm shooting for.


Sometimes I feel like I'm overloaded with information on the Torus screens. :) It seems like a variety of curves work for people. Hifi, since you're happy with what you got and are also using an XG, that's a strong recommendation.
 

·
Registered
Joined
·
1,910 Posts
The main problem I had was running out of bow control. One really good thing that I realized out of my experience is that you should check out the pjackso spreadsheet for two reasons - he gives gentle curves that will minimize your geomtry issues, focus issues, etc; and the great, great thing is he plots where to measure to setup your curve without getting out in the driveway with wire and pencils etc. (not that I know what that's like ;) )
 

·
Registered
Joined
·
3,387 Posts
I was planning to use curves from -Pjackso's spreadsheet, which (for an 85" wide screen, 10' to viewing distance, 4' wide viewing area) has 3 5/16" H curve and 1 1/16" V curve. Vince didn't say his screen hotspotted (right Vince?), but it did drop off a little bit at the edges. Maybe the deeper curve would completely avoid all hot/warmspotting, but then you run into focus issues. The shallower curves seem like a good compromise.
 

·
Registered
Joined
·
1,910 Posts
Oh I definitely have hotspotting, have you checked out my pics in the Torus Lite thread? It's not the hotspotting that remains that you should focus on, it's the compromise between a deep dish with no hotspotting and difficult geometric/focus issues and the heinous hotspotting/colorshift of the pearlbrite when it's flat - put it this way, even with the flat pearlbrite I would have to think hard if I could give up the gain, the torus fixes that issue. I meant to someday post some comparisons of the flat pearlbrite but haven't gotten around to it. The pics I linked to exxagerate the hotspotting and color shift versus what you'd see in the room, but critical viewing will pick up on it anyway.


[philosophy] Like alot of things, it's a compromise but a worthwhile one. I think many products that you use have compromises built in that you're not aware of, when diy'ing there is a tendency to get hung up on perfection - I enjoy the process of compromise and the limitations of the technology. You can't have perfect convergence, geometry and brightness consistency when using a curved screen unless you have special lenses for the curve. Even then when off axis you'll have geometry issues. Of course the fun of a curved screen also weighs into it, it's just cool. Hey, when my wife notices the pop from it, it's not subtle (she's getting better, now she can tell HD from DVD, I'm so proud). [/philosophy]


Pros -

brightness pop - you won't go back to unity gain

3-d effect - from increased CR, again a stunning difference

ambient light rejection - don't care that much but it's an improvement

really cool movie theater feel - it's a unique feature for your room


Cons -

focus - the big one - stick to pjackso curves or milder to minimize this

geometry - really not a big deal, what crt is perfect? I decided the curved screen

actually makes the geometric deficiencies of the pj less noticeable

hotspot/colorshift - again keep the curves mild

weird echo from screen - doesn't seem to be an audio problem, only amplifies

sounds back to the source


Notes -

torus is not rocket science - don't make it so.

keep it simple as possible - see my fan box design in the torus lite thread.

don't get hung up on optimizing the curves, angles, etc for your room. First build

a torus and then decide if you care about that last 15% improvement - it's not

rocket science.

dont' get hung up on fancy fan needs such as pressure switches, just put a

variable resistor in line and turn the fans down as needed - it's not rocket

science

did I mention that it's not rocket science?
 

·
Registered
Joined
·
3,387 Posts
So the torus isn't perfect -- especially if I can't get my 10-yr-old to shut up during movies :D -- but it's a big increase in pop and depth, and that's what I want. I'll be using Paul's curves so hopefully focus and geometry won't be a big problem. Gotta either dig up a few small fans from the junkpile or go find a nice quiet 120mm fan. But with an XG in a not-great hushbox, I doubt I'd hear a fan behind the screen anyway. :)
 

·
Registered
Joined
·
2,048 Posts
Hi,


With the number of cons listed above I was wondering if using a 1.5 gain screen in a torus lite frame would still have some of the benifits like added depth and a little more brightness but without as much hotspotting(due to swallower curves and the 1.5 gain) and less colour shifting. Lesser curves obviously is less strain on the geometry circuits and the convergence.


Would it be worth the effort.


Thanks
 

·
Registered
Joined
·
3,387 Posts
I'm no expert, but I would think 1.5 would be too low a gain. Low gain means the light is being dispersed in all directions, which (on a curved screen surface) means that one part of the screen can wash out other nearby areas.


Iceman's study used 2.0 and 2.8 material for the torus. I don't know if he modelled washout or not.


If you were going to use 1.5, I'd use really shallow curves.
 

·
Registered
Joined
·
2,766 Posts
Yeah the depth of curve is totally related to the gain of your material, so you could probably get away with a pretty damn shallow curve with say a 1.8 material if you were only looking for a mild gain increase.


What I'm more interested in at the moment is how a *really* high gain material will cope, and wether colour shift will be a serious problem no matter what the curve.
 
1 - 20 of 28 Posts
Status
Not open for further replies.
Top