AVS Forum banner
Status
Not open for further replies.
1 - 19 of 19 Posts

·
Registered
Joined
·
4,654 Posts
Discussion Starter · #1 ·
SXGA, LCD, MLA, 800:1 Contrast, 3000 Lumens, Slick internal processing.


That's what I want, and I'll pay $8,000 steet price, which is about $15K retail.


Why isn't Sanyo making that?


Does their new contrast ratio technology work on SXGA?


Does MLA work on an SXGA chip?


It would take the market by storm.


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

Joe


"I'm a Dapper Dan man!"
 

·
Registered
Joined
·
108 Posts
I came across this a few days ago, check this PANASONIC PT-D9610U HD out, may be it will fit your bill. SXGA Resolution, Contrast Ratio of 1000:1, 12,000 ANSI lumens. Panasonic's own professional theatre grade progressive scan and deinterlace technology. Price - .... http://www.avsforum.com/ubb/eek.gif
 

·
Registered
Joined
·
4,129 Posts
Quote:
Originally posted by Dean McManis:

Although 12k lumens is a bit much for a home theater, ...
Yes, but it would lead to the advent of the first DIY black screen http://www.avsforum.com/ubb/biggrin.gif


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

The button is labeled "Play", not "Pay". STOP the MPAA!

Our Silent Angels

Please visit The Manny Page!


[This message has been edited by Man E (edited 10-11-2001).]
 

·
Registered
Joined
·
340 Posts
I like the part about "allowing two to 100 (maximum 10 x 10) PT-D9610Us to be set side-by-side or stacked up to 10 x 10". If $139K doesn't sound like enough to you, how about a cool $14M for a 10x10 array? 1,200,000 lumens - now we're talking.


- Chris
 

·
Registered
Joined
·
4,654 Posts
Discussion Starter · #7 ·
Noah, yes, and I responded.


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

Joe


"I'm a Dapper Dan man!"
 

·
Registered
Joined
·
79 Posts
Well it's not SXGA but it does have everything else:

XGA

3000 Lumens

MLA

Contrast 800:1

DVI

HDCP

and a LAN connection!


I'm talking about a Proxima DP 8000. Estimated Street price $6200 with an expected ship data of October 15th.


This is one of the few projectors with HDCP.


The network connection is really cool. What a great way to setup and control your projector, over the network from you PC.


--sdc
 

·
Registered
Joined
·
4,075 Posts
You guys are so busy chasing the next new toy when do you ever have time to watch movies? http://www.avsforum.com/ubb/biggrin.gif


Joe, I think you deserve one of those 12,000 lumen jobbies. You can erect a cooling tower right there in the back yard and use water from your swimming pool.


Let me know when you get it installed, I'll catch the next plane out! http://www.avsforum.com/ubb/smile.gif


Dan
 

·
Registered
Joined
·
4,654 Posts
Discussion Starter · #10 ·
12,000 seems a little strong, even for my lust of brightness.


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

Joe


"I'm a Dapper Dan man!"
 

·
Registered
Joined
·
835 Posts
Joe,


One of the lower end Panasonics meets your specs almost exactly. It's a 2 bulb SXGA LCD (w/ MLA) and 3000 lumens. I can't comment on how slick the internal processing is, but it should be competent and I believe it has 3-2 pulldown. $8k street is probably not in the books though, the list is something like $18k and I can tell you that no projectors outside of "Home Theatre" projectors have the margin for a nearly 50% markdown. You'd be looking at closer to $14k.


The downside is that the contrast ratio is only ~500:1 on the SXGA version. However, that should be a real world measurement. I have tested the XGA version before and it measured out in the 650+:1 contrast range, the spec is 700:1. Considering the inflated contrast ratio of many LCD projectors, that is a *great* number and the picture was punchy with quite good blacks.


Beware claims of 800:1 contrast ratio, even in MLA LCD's, they may not be (probably aren't I should say) accurate. Of course, bad specs are a problem with many projector manufacturers.


Regards,


Kam Fung
 

·
Registered
Joined
·
3,667 Posts
I guess the real question is why SXGA panels cost so much more than XGA. I would hope that Sanyo could make an SXGA version of an XP18 or XP30 at a minor increase in cost.


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

Ken Elliott
 

·
Premium Member
Joined
·
24,995 Posts
OK, got it, thanks Joe. In the past I've gotten a separate window notifying me that I had PM.


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

Noah
 

·
Registered
Joined
·
835 Posts
Ken,


I'm not surprised it is so expensive, there are lots of R&D + tooling costs to cover. The chips have something like 1.77 times the pixels, so there's a lot more complexity there (and room for dead pixels to show up). Also, the Panasonic has a dual bulb system which is bound to cost something.


If a WXGA with no MLA costs... ~$6k, I would figure an SXGA to cost at least $8k ($10k with MLA), but with something like 1000+ lumens. I don't know how much demand there is for something like that with WXGA so common right now. For 3,000 lumens you're going to have to pay quite a premium for it.


Where's my 1360*768 LCD with MLA and 1500+ lumens? I would seriously consider $7-8k street. It could be the high brightness cousin to the new 720p DLPs.


Regards,


Kam Fung
 

·
Registered
Joined
·
3,667 Posts
Kam Fung,

Actually the cost per chip has nothing to do with the number of pixels, it's related more to the area of the chip and to some extent, the complexity. I see no good reason why one couldn't shrink the chip a little to the size of the XGA chips and make a chip that cost approximately the same.


Engineering costs at the chip level would be probably $1-2M including prototype fabrication + another $1-$2M for the projector design modification. It helps that the chip is a systolic array.


If you sell 20,000, the engineering costs dominate. If you sell 200,000, they are in the noise. Yield is a different issue, but if Intel can make Pentiums for
 

·
Registered
Joined
·
79 Posts
kelliot,


I agree that getting the chip size smaller for higher resolution is critical to keeping the cost down. There are problems focusing arc lamps on that small of a chip. New lamp technologies being developed and further development of arc lamps will solve this problem.


Some companies are currently making SXGA LCOS chips and I would expect to see some on the market next year. Hopefully at a reasonable price point.


--sdc
 

·
Registered
Joined
·
3,667 Posts
The 1.8" SXGA chip used in the PLC-EF12 and PLC-EF30 can't be that much harder to make than the 1.8" chip used in the PLC-XP45.


They could shrink it to 1.3" as with the PLC-XP30 and save costs and still get 2000-3000 lumens.
 

·
Registered
Joined
·
835 Posts
Ken,


I'm by no means an expert on chip fabrication, but I don't think a die shrink is as simple as you make it out to be (otherwise they would happen a great deal more frequently). Yields are a *big* issue, they directly influence how much a chip costs. It would seem to make sense that the more complex the chip the lower the yield, all other things being equal. The rate of rejection has a great effect on the cost of a chip. The die shrinks for the original PIII's happened over the course of years (0.35, then 0.25 with the Coppermines I believe) and they started selling them in high margin Notebooks first to test the process at lower volumes.


If you go back to your example of the PIII, it was *expensive* when it first came out, even if it is ~$100 now. The P4 2GHz is right up there in the high hundreds even with intense competition from AMD, there is a learning curve and an associated cost when you have a new process or a new chip. The companies aren't just mercilessly price gouging us! (although I'm sure there's some of that to, charge what the market will bear, right?)


You are right though, the actual cost of materials is not really related to complexity as much as the size of the chip.


$2-4m is a lot to cover at low volumes (although I imagine it might be more than that), just how many of these things can they sell? Definitely not in the 100,000's, you'd be doing pretty good to sell 10,000's.


It's not really as simple as buying a 0.35um machine otherwise everyone and their best friend would have done it! :) The current D-ILA has a pixel pitch of 13.5*13.5um that's nowhere near what you see in conventional semiconductors. The other techonologies are comparable in size or larger. The processes used are not, to the best of my knowledge, exactly the same as conventional CMOS chips. Don't worry, Moore's Law still holds and we should be seeing higher resolution projectors in the near future.


Also, it's not just the higher resolution chips, it's the chips that drive them as well, there's a lot more to it than just popping a higher resolution chip into a projector and hoping it works. You have to get a faster scaler, faster chips to do the A/D for computer inputs, faster chips to drive the imagers, higher bandwidth for all the pipes connecting everything, lower noise components to handle the higher input bandwidth, etc. Some of these may be developed in-house or purchased from others, with all the associated costs. For example Sharp did their own development for their scaling chip and it has a complexity up there with many video cards. I can tell you from first hand knowledge that the R&D for a high bandwidth ASIC like that is at least $4 million, if not significantly more.

Quote:
The real issue may be marketing. No one may believe that there is a market and that may constrain investment.
I definitely agree with you, however they might also be right and there isn't a big enough market, at least not yet!


I don't know, I believe that the people running projector companies and their engineers are not complete idiots (even if they make mistakes on occasion). If they haven't done it, there must be a good reason. Shrinking the size of a chip or making a higher resolution one has got to be a no-brainer *if* it is possible. I can only conclude that it is not right now.


Regards,


Kam Fung
 

·
Registered
Joined
·
79 Posts
Kam,


The chip complexity does effect yields but the die size is really important. Smaller is better. The LCOS technology can readily make very high resolution chips on small die sizes right now but as I stated above, the lamp technology is an issue with small chips.


The issue of pricing is another story. If a manufacturer expects a small market for a projector, the margins will be set very high. So even if the chip price is reasonable the projector may have a very high markup.


--sdc
 
1 - 19 of 19 Posts
Status
Not open for further replies.
Top