Do it sim. If you’ve got enough light there is nothing better than the impact of a large screen. It sucks you into the movie/game better. I’m running mine at 120” and it’s almost perfect for my room. With the quality of you images you shouldn’t have a problem even with no gain.

DJ

Great, than I think I will. I wish my Zoom Player Pro trial hadn't expired it's not really that clear in Power DVD..sigh...I wonder if there is a hacked version out there.... I know your all are enjoying this so here's some Nemo. BTW, the woman wanted to watch her chick flick on the big screen and there was a scene where the sun came out from the horizon, she joked and said that it was to bright

This one's for the A/G Team!

Just to clarify: What I meant was use a bigger screen. I don’t know how Brains screen will go I haven’t seen it.

DJ

Ok, then can I borrow your wall?

I just checked USPS and I’m afraid they don’t ship items over 70 lb. But I may be able to strip the paper off the plaster sheet and put it in a postal tube if you don’t mind it being in two pieces.

DJ

Sim,

How much for your dalite?

pm sent

SIM, I beg of you, PLEASE take some LOTR screens!

What's this Brain's screen you speak of?

scoodidabop
Check it out!!!

LL screens !!!

Cheers P

Ok, I gotta get my Zoom Player back on. I'm having issues with PowerDVD and what it's doing to projection. I tried Windows Media player and it's better but still I'm limited to what it can do. Here are a few examples. Still the same size 68"x92" but Windows Media Player is in smaller 16:9 of course.

Simul8r,

I don't think the image shown in post 182 has the measurement dots located in the right spots for a proper 9 point lux test. The outer spots should be 1/6th in from the outer edges. The ones shown in the image look a little farther in from the outer edges. This will slightly inflate your lumen calculations. I don't think the difference will be too much, but it may be significant (maybe 10% or so difference).

Hopefully I am wrong. But I noticed it because I just recently took lux readings on my setup and initially made the mistake of placing the outer measurements just a little too far in. Either that file is slightly off or my calcs are. But my initial mistake was to draw 3 evenly spaced horizontal lines and 3 evenly spaced vertical lines. Then I initially took 9 lux readings at the interceptions. I think this is the wrong way of doing it.

I think you have to draw 2 evenly spaced horizontal and 2 evenly spaced vertical lines. Then put the meter in the middle of the 9 squares. This method places the outer measurement spots slightly closer to the outer edges, but provides a more evenly distributed set of measurments.

As I said, I don't think the difference will be too great. Your build is still quite spectacular. I hope I didn't de-stoke you.

Not at all I really want to depict what I really have no matter how large or small. So, hold on......

Would these work, then?

Based on the test image above and remeasuring after leveling the screen enough, I have the following:

88"x67" = 5896 sq/in or 3.80sq/mtrs

64 76 53
75 95 77
57 78 63

70.89 average x 3.80sq/mtrs = 269.38 ANSI

Well, it looks like with this pattern I've dropped which means my goals have now gotten shorter.

which pattern(s) did you use for 9points in your original precondenser thread?(where you got to 185 ansi)

I've been using this all this time since the precondenser test's. At first I was taking readings 2 inches from the edges of the screen then moved to this thereafter. I placed it in desktop because it was smaller in Window's MediaViewer at one time and thought this was appropriate instead.

I think this is something that needs to be addressed. The problem comes in that some are giving ANSI for 4:3 screens while I know myself am giving them for 16:9 screens.

I've been using Paladin's pgen program (it can now be located in the 'build and calibration tools' area) for locating the test areas as his program will locate the circles differently if using a 4:3 screen or 16:9 screen. I'm not even sure if this is the correct way to do the ANSI test, but it seemed to make sense.

personally, when I record ANSI measurements, I use the pgen program and select 1024x576 custom resolution (the actual desktop resolution being run through powerstrip) and choose the 'solid color' option to get the testing locations for the lux meter.

anyone have thoughts on this ?

gs

edit: duh! ... i see SIM has posted screenshots of the pgen program above... sorry about that. Do you generally run a 4:3 ratio when doing the ANSI test ? Is it supposed to be 4:3 when doing the test, or is 16:9 ok ?

thanks.

Personally, I beleive 4:3 should be the true size since I use this for everything regardless of working or playing. When I do go down to 16:9 then I'm more or less manipulating the screen from the actual total area of the panel. Otherwise I'd might as well target the cone from the precondenser just to this center part and not the rest of the LCD and say thats the only part I'm measuring. But I'm open to any logical ideas around this. Oh BTW, I've been always measuring in 1024x768.

I’ve almost given up on my lux meter; I simply can’t believe that my LCD is less than 1% transmissive.
Something I did notice while playing around trying to get reasonable values was that my NVIDIA settings played a large roll in the lux values I was measuring. The highest lux readings I could get were with all the NVIDIA settings maxed out. So perhaps the best method would be to take measurements with the LCD turned off. That wont help me though as my LCD is black when switched off so I’ll just have to sit back and watch.

DJ

How about this idea. The transmission of our panels is something we can't control. It's really based on each type made by the manufacturers. What we can control however, is how we design and build the plans to throw as much light thru the panel itself. Therefore, if we maintain 2 seperate catagories one being with the true readings without our LCD's would depict what was truely attained by the design and or builder and the other would be left for questioning until we can get a handle on the actual true nature of tranmissive LCD's. Comments?

Yah, now I can be included an not feel left out. You guys are so kind to let me play.

Nice results

excellent suggestion sim.... and for those of us running powerstripped lcds(like myself) I huess I can make a cutout the size of the actual transmission area...and then in conjunction with the pgen program its fairly easy to work out the transmission%..

DAZZLA: your screen is black in the off state? I thought I saw a reference to you that your lcd was poly-silicon? any more info on how you know that? whats your panel number? usually on the backlight if you still have it....yours may be a PVA panel which I know are less transmissive but color is better...

You got me all paranoid so I went back through my posts. I did find this post where I made a Freudian slip, should have typed a "a" instead of a "p" . And also yoshuaspawn talked about his poly silicon LCD indirectly referring to mine. My LCD a-si TFT/PVA Sorry for the confusion.

DJ

That is exactly what I mean. Each measurement in this test pattern represents an average lux of 9 equal area squares. You're a stand up guy imo for making those changes so readily. Many of us would have argued incessently to keep the higher lumen test pattern. It's simply human nature.


I think the ratio should be the same as your lcd ratio. so if your lcd is 4:3 then your screen measurements should be 4:3. If your lcd ratio is 16:9 (let's say the 15.4 WUXA panel) then your screen measurements should be 16:9.

The reason I think this is because if you have a 4:3 panel and measure a 16:9 ratio then you will be excluding light that is leaving your projector from the measurement. When you then compare your setup with someone else who did a 4:3 measurement with the exact same setup, it will look like yours is a less efficient design/parts/build. It might cause you to feel that there are additional changes that need to be made to optimize your light engine, when in fact it is already optimized.

In addition, I am a little skeptical about imposing a white background for the measurment. This is because there are an infinite number of white colors. My own lcd can be set to 3 different whites, such as 5000K white, 6500K white and 10,000K white. The only way an lcd can change the white color is by subtraction. It simply biases the RGB filters slightly and blocks some of the light. I think the maxmum measurment and probably the truest measurment of total transmission is when the panel is turned off. Then there is no signal bias to "tint" the white towards any color temp.

I don't think the difference will be great. It may not even be significant. However, technically, an off panel should provide the maximum transmission numbers.

I guess I'm more interested in how bright our projectors are before a LCD panel is actually installed. This would (in my opinion) help others get an idea what the average is and hopefully give them ideas on how to improve upon theirs before they settle on a particular panel. Max out the efficiency of your projected light and anything between the fresnels or after it is what you'll see on your wall.

I vote for a panel out standard.
As SIM says,there is nothing we,as DIYers,can do to compare results,when using different lcd's.
We are just adding complicated mathematics,trying to measure with lcd in place,then take into account the transmissivity(? ),then remove the transmissivity(? ),to get a standardised result,which would be more accurate accross the board,if the measurements were initially taken "raw".
More people would be able to give accurate input if less complications were in place.
P

Appreciate the vote pun.

On an another note, I can't stop taking pictures of this. Somehow, I think DAZZ cursed me or tried to warn me about this So, without further adue, here's some UNREAL TOURNAMENT 2004. The other character you'll see is actually quite menacing but a lousy player.

Sim,
Yes, your suggestion is a good one... everyone could compare apples to apples this way. I think we'd see more consistant numbers between setups this way as well. We all tweek our light setups in the same manner, then apply the LCD of our choice...

Of course final on-screen lux (viewing situation) would be determined by panel make/model, LCD settings, controller settings, etc...

gs

I also agree that we need a standard of measurment to determine solid stats in our projectors. Also bare in mind that we all don't have the same meters to measure with. Not everyone has a CIE certified meter so this will throw the numbers off some.

And....here we go again. I've brought the light box closer, distance was set at 210mm instead of 240mm. Turned her on and I saw that the corners were beginning to dim...not acceptable but of course (and I should have realized this from my precondenser experiment) readings had risen. Middle read was at 123. So, I turned her off and readjusted bringing the lightbox further away to 220mm and turned her on again....much better. So here are the readings based on the white background/9 circle areas:

66"x86" = 5676sq/in or 3.66sq/mtr

62 85 53
80 116 78
58 87 58

Average 75.22 x 3.66sq/mtr = 275.31.....I gained about 6 more from last reported 269.38. Not what I'd had hoped for but every little bit is always welcomed.

Now, on the other topic. I appreciate the kudos gents and again I'm always willing to brainstorm other ideas that would work better. As I said, I'm no expert in this. The only thing I see difficult with taking readings without the panel is how are we to set 9 points on the projected square area without drawing grids on the screen and or stringing up evenly divided points without the help of a LCD and paladin's projection test program? That seems like a little more work just to get readings. Can we just take 5 points which would only include the very corners and the approximate center (which may just be the highest reading)?

Hey Sim,,,

are you still using your lamp vert, relative to the ground,,

I have been talking with some who deal with MH lamps,, and despite manufacture's statements of omni position,, I am tole that the lamps die very fast, and also with temp imbalance they explode violently

I just went down the street this morning bought a new philips,, to get around the salts issue

400w 35,000 lums, 4000 K temp

oh boy put it in, brand new fired up white,, and ten minutes later, had a vastly superior picture full screen than I ever have had,, and in daylite too !!!!

Yes elken, the bulb is still vertical and standing straight up on it's base.

I've been using the previous ushio s400dd this way for sometime especially 10 hours straight until switching to mikyd's T15 6500k with no issues. As far as it's life, I guess I'll just have to see how long it lasts if what you say is absolutely true. But then again if I do get half the life of 20,000 hours (I'm guessing) isn't 10,000 hours worth 30 to 50 US for a replacement? As far as temp, I think my 150 cfm wind tunnel is very iffiecient in dispensing with the heat quite well.

I'm not sure but I think mikyd measured his T15 at 40,000+ may have to ask him for another shout out on this again.

Excellent, great to hear your able to see in the daytime. Hope to see new pics of this at your plog soon.

Elken,

Do you have a model number for the bulb you purchased? Also, what is the ballast that you are using in conjunction with it (LL e-ballast)? Lastly, do you have measurements of light output?

Cheers,

Fulcrum

When I took 9 point readings w/o the LCD in place I started by placed post it notes on the screen in the locations the readings were to be taken (using the pgen program with the LCD in the PJ)... then I removed the LCD to take the actual measurements.

Well, I was thinking along the line where a newcomer begins his/her build without purchasing an LCD. First calibrate the max projected light from the build then decide on a monitor. This way they wouldn't have to make a huge purchase all at once but first perfect their lcdless pj before finalizing with a panel.

I guess we can come up with some average readings but how is a new builder to find out his when he can't project a grid without having a panel.

Perhaps a transparency to lcd size is the anwser here. I would be willing to help a member in this respect. It would enable the light engine to be tuned beforehand.

Great idea Chuck ! Thats mighty big of you to offer your help too as well

Cardboard cut to the size of the LCD?

.......................seriously DAZZ............I need to borrow more wall. She's already backend to the rear of the room. I'm already thinking of going to the backyard with this at night.

size is about 9'x11' at 17 and half feet away.

..........

HOLY **** !!!!
What is this magic!!
Damn SIM,that is amazing.
Check those corners too!
What are you going to do next!
Cheers P

i don't know, I'm still bewildered. maybe do another lux reading?

I'm dedicating this one specially to brainchild and the other moderators.

Great job Sim,if the pics are this clear at 240" they should really stand out once it fits your screen really impressive work.

RR

I knew it, you really a midget. Serriously though what is your current diagonal? It still looks great.
As agent Smith would say “More”.

DJ

Thats what I did for mine. I used the DIYPro background and transferred it onto a trnasparency which I mounted on a plexi-slug. Worked pretty well for tuning and playing around.

SJ