CurtPalme.com Home Theater Forum

[VideoGrabber]

Person99 wrote:
> Always count on Tim for the cool ideas! <

Thanks, Dave! I try to toss things out from time to time to get folks thinking outside the box, about alternatives and possibilities they may have dismissed. Whenever possible, I share ideas I've tried and tested myself.

> I was counting on the extra width in the middle of the tube to get a wider 2.35 then you could 16:9. But, your method, lets it get even wider because with keystone, the widest part of the active image area is just about at the widest part of the tube. <

Yep. A nice bit of serendipity.

> It is MUCH smaller than any of us would set them up. <

True. You can partially mitigate the wear on 16:9 content by horizontal orbiting, which will distribute it laterally. Though the central portion will still wear fastest... which automatically solves your contrast-modulation problem for uniform brightness.

> the NECs tend to cook the tubes within 4000 hours. If you watch alot of 16:9 content, you would not get much more than 4000 hours of the tubes IMO. <

I was guesstimating 5-6k, but you might be right. Would you rather get 4k hours of great image out of a set of tubes, or 10k hours of dim but acceptable images? I never indicated there were NO tradeoffs. There are always tradeoffs.

> I was saying "downscale" because, although you could do it without downscaling, no 8" machine is capable of resolving the full 1920 width on an active image area about 75% the width of the tube. <

First off, I want to be clear that I have nothing against using a scaler as you suggested... as long as you happen to have one, and it can do everything else you want it to do (AAS, IVTC to 72i and 96i, etc.)

Secondly, I didn't see the part where we were restricting the discussion to 8" CRTs; and with 9"-ers, this is certainly doable, even restricted to ~83% of the width.

> I experimented with this extensively on my Onyx. <

Cool! Real-world experience. The best kind.

> The power to resolve 1920 really started dropping off once you got below about 92% of the tube width and was really bad at 88-89% of the tube width (in other words, 1-on/1-off was just a gray field). With your trick above (using half the tube which also maximized width usage), then 16:9 would probably be using something closer to 83% of the tube, but it could still not resolve the 1920. Now, using 83% of the tube face (if we could get that high), we might be able to eek out a resolution of 1650 without softening. So, for a non-soft picture, you would have to downscale the horizontal to about 1650 with P16s and much less with 180s. A good 9" *might* be able to pull off the whole 1920. <

Nicely explained. And I'm pleased to see you acknowledge the 9"-er potential. But you've made one common but important mistake in your evaluation above. You made your determinations of the onset of softness based on test-patterns. Now TPs are great, and can be used to obtain objective analytical results in many contexts. And if you want to claim that "1920 TPs just won't look good, cuz they'll be too soft", I'll grant you that.

Surprisingly though, I don't find them that intriguing for normal viewing. I like to watch films, that have native resolutions after acquisition filtering that are hard-limited to ~1770 horizontal. (HD content has to start rolling off at 35 MHz, thanks to Mr. Nyquist. Your TP generator does not.) Or even live-video captures from 1440 HDCams, which tend to look sharper than they really are, due to excellent contrast (or contrast-peaking processing). While digitally generated, transported, and displayed animation could theoretically approach 1920 (as long as it never goes through an analog stage), it still looks exceptional at effective 1600-1800 h-rezs. One of the nice things about analog PJs though, is that they'll downscale it to 1600 or 1700 for you automatically.

If instead of a TP you had used high-quality source content, you would have found that subjective softening kicked in much later than you've indicated above. And that's the key. Along with maintenance of the effective contrast, so you still have the image "punch".

> But, you are just going to barely be doing the 817 horizontal lines of the 2.35 image in the 2.35:1 raster if progressive. So, there is NO WAY you are going to pack 1080 into that space without being REALLY soft. <

Agreed. I was not advocating any more than 800p in that raster area, if you're going progressive. 1080p would not be a good thing, which is why I wasn't suggesting it.

> Since with CIH, you can't use any more height, You will have to downscale the 1080 to 817 as you note. 1080i might help this but it is still going to be pretty soft--way too soft for me. <

1080i helps a lot more than you'd think. Too bad you can't give it a try any more. Of course, in this scenario you'd have to set your stig for round dots. And while I wouldn't advocate this on 8" tubes like the 180dvb's in my 808G, I think the P16's in your Onyx could have pulled it off pretty well, due to their improved beam-spot size capabilities. Also, this is about the only way you can use interlaced mode effectively on 9" PJs, which avoids having to go to an expensive scaler and trying to get 1080p @72Hz.

> My statement is correct. If I have 1920x1080 16:9 source material, you cannot properly display it. You NEED to downscale it to 1650x817 or less. Whether it still looks good is another discussion, but you WILL have to downscale it to look good. <

Did you actually try this with source content, and observe the terrible image degradation you're referring to? Actual 1080i 16:9 content squeezed into the smaller space looked bad?

> If you push the tubes that hard with alot of 16:9 content, I'd be surprised if you get 3000 hours out of them. <

Well, it seems like I watch a lot more scope than 16:9 (maybe I just enjoy it more), but it might actually be 50/50. This will vary from viewer to viewer. I'd still say 4k hours, and if you couldn't get that, then go to a gain screen. And at 4k hours, that means 8k hours per set of tubes. Even with your pessimistic estimate, it's still 6k hours. How long would it take you to log 6k hours of viewing, before having to replace your first set of tubes? Would it be worth the $$$ to you for a CIH scope setup?

A lot also depends on just how big is BIG for you. A 10' wide screen would be HUGE for me (too big for my space, in fact), but 12-14' wide is more to some folks liking. In those cases, you must have gain, which Vern did not. The original question here wasn't, "Can I have an enormous screen, light it up brighter than the sun, and do it in CIH with a 7" CRT, and get 10-yr life out of the tubes?" It was simply, can I do a CIH scope screen with a CRT... and my take was that you definitely can. Depending on your mix of requirements and preferences, it may require a 9" unit, or a gain screen, or a balancing act with a bag full of high-tech tricks... but it can be done.

> Also, Vern's was still not that bright (remember, I live 14 miles from Vern). <

Must have been pretty d@mn bright, if you could see it from that far away!

> He switched to a Qualia the second it became available because even though the CR sucked (less than 3000:1) if did CIH way better than the NECs. <

True. One of the biggest failings of his NEC (all CRTs) was brightness uniformity. Both Vern and Art commented on the huge improvements in that dept., when they switched to digital. Corner brightness is way better.

[Person99]

[VideoGrabber]

Dave, thanks for continuing an interesting discussion.

> Again, you should know me better than that. Wink <

I do. But you're not the only one reading this (hopefully).

> 1) If a display handles TPs well, it will handle all content well.
2) If a display does not handle TPs well, it may not be able to handle content well. <

True. Although I'd say that statement (1) is a lot stronger than (2). The corollary being that a display could look putrid on certain TPs, yet shockingly good, even excellent, on real-world content. Recalling that resolution is only one factor in the PQ equation, and usually ranking down around 3rd or 4th in overall importance. Though I suspect both you and I would bump it up a notch.

> ...my subjective assessment is that actual 1080 content got softer on my P16 tubed machine at just 83% the width. <

This is very valuable information to have. Thanks!

> I acknowledge your statement is true for CRTs, but it seems to categorically assume an analog stage which is not categorically true. <

Yikes! Aren't CRTs the only thing around? Don't tell me they've got those new-fangled digital dingus's working now.

Yes, I'll grant all the comments made regarding the potential to display native digital 1920 content with digital PJs, with full rez. And they may well be the better way to go for CIH 2.35 AR setups. There's no question they'll be both sharper and brighter. I was just trying to respond to John's initial question regarding the possibility of doing scope with CRT PJs, to enable the advantages that they offer.

> I tend to dislike subjective assessments as they tend to be quite dubious... They lead to bull**** like this:
http://www.hometheaterhifi.com/volume_12_3/marigo-labs-signature-3d-mat-7-2005.html <

Yeah, that one was totally mind-boggling. You expect an "April Fools!!" to pop up at any moment. But the guy was totally serious. Scary stuff.

> VideoGrabber wrote: ...but it can be done.

I'll grant you that. Although I'm becoming more and more fond of brightness. Smile 10 ft-lamberts is a little lifeless. 14 has some punch. <

Interesting comments. I suspect I'd be completely happy with less than you're proposing, but agree that it's an opinion-call, and will vary. Few of us really know how many ftL we're getting, and many who do know have the wrong numbers! But CRTs are certainly at a disadvantage if you want big AND bright at the same time. Pick one.

> I don't see many CIH screens smaller than 45x107 and I don't think most would want to go smaller than that. <

True. That's where screen gain comes in. And a whole lot of variables to shuffle and balance.

> Lighting up that screen with the small raster takes almost as much output as lighting up a 60x107 16:9 screen and few of us would recommend doing that size screen with a CRT! <

More like 52x93 16:9 screen equivalent.

> Additionally, CRTs have the "throw distance" problem. A CRT set up for large CIH may have to be further back than the back wall! <

Absolutely! That was the case for me. When I was exploring the possibilities of how big I could go with a scope screen in my media room, and what I'd need size-wise to keep the other ARs looking good, I realized I'd have to punch a hole in the outside wall, and put the CRT outside. Of course, the solution there is magic... do it with mirrors (no need for smoke). Mount the CRT vertically on the back wall, facing upward and firing into a first-surface mirror (or in the attic, firing down, if your configuration is suitable). That extended the throw length significantly, so I could have gone to a 10' wide screen (or even more)... if I could have fit one in the room. At that size, it would have had to be a high-gain torus, or a Vutec silver screen.

Instead, in my case with a lowly 8" PJ, I decided to downsize, and do a CA screen (~24 sq-ft) instead. So roughly 80" wide for regular HD content with plenty of brightness (surprisingly, right where you mentioned here earlier that you thought 8" looked best), expanding to 90" for scope... all viewed from ~9' away. d/w of 1.2 for scope (45-deg viewing angle), and 1.35 (40-deg) for normal content. Small, I know, but combined with AAS, EBS, and NCSR (72i,96i) it worked for me.

[Active-Area Scanning, Elliptical Beam Shaping (optimized for interlaced display), and Near-Constant Scan-Rate (40k).]

[Person99]