or Connect
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - $3,000+ USD MSRP › Optoma H79 review & screenshots
New Posts  All Forums:Forum Nav:

Optoma H79 review & screenshots - Page 8

post #211 of 2404
I velcro my lightmeter 'head' to my CF sensor plate which is tripod mounted so that it's in the same area. It holds it still and is easily attached/removed.

Gary,
post #212 of 2404
Thread Starter 
It's got the white cap. The 100IRE was steadied good. But I was moving around trying to get the first low reading which was half way to the PJ. I got the way to find the highest decimal, it's just after the 10x spot. This I used this morning 5am/dark. It's the reading that gives you the most numbers four it was 17.25. The other spots give a similar number just deleting the right figures though. Like 17 instead of 17.25.

One thing, I could get a higher than 17.25 if I raised the meter up a foot or so higher in the area where the screen was. I got up near 19.00, but I figured I needed to be dead center.
post #213 of 2404
Quote:


Originally posted by guitarman
I give it a try but do think the 100IRE will be way high at those closer levels. Since I already got .003 a few feet away, if at that point I can get a 100IRE reading and it's way higher, it should be a pretty high cr reading. Back to wait until dark.

The problem with .003 is it is basically 30% error either direction (since the last significant digit is usually + or - one). I figured .005 was more reasonable if you could do it.
Quote:


Originally posted by krasmuzik
How does the staged measure work - isn't the gamma going to pollute the numbers?

I haven't really felt the need to do it because my lightmeter seems to work pretty well for most of the CRs I've measured just by getting it fairly close to the projector. I did try it once and I think it came out pretty good. The gamma shouldn't make any difference though. All it relies on is that 20 IRE doesn't move and as long as you don't change the gamma between measurements (changing anything on the projector between measurements is of course forbidden) then the gamma shouldn't matter. 20 IRE was just an example I threw out because with a good 2.2 gamma the 100 IRE to 20 IRE ratio should be under 100:1 (I believe somewhere around 50:1 give or take). I would think that meter would have an accurate range that was at least 200:1. Which would put the limits of measurement at well over 10k:1 with 2 sets of readings carefully done to be within the accurate range.
Quote:


Originally posted by TomHuffman
Even if the meter is off (the ColorFacts forum suggests that it overstates luminance by about 10%), since CR is a RATIO, I don't see how this matters. I whatever the error, it would appear for black as well as full on white. I've read quite a bit about the Trichomat being an unreliable device to measure color at low light levels, but a good CR number does not depend upon a precise color measurement.

Are you guys saying that the Trichomat's Foot Lambert reading is so wildy inaccurate (a variance of greater than say +- 10%) that it's totally useless for CR measurments? If you are, then Mark Hunter needs to know about this.

A meter can be off by 10% at 100 IRE and close to 100% off at 0 (or 7.5) IRE. The EyeBeamer (or EyeOne) is horrible for on/off CR. The Trichromat isn't bad. Mark Hunter knows this. I don't think Tom has a Trichromat though. If he did, I would suggest using that at maybe 6 ft from the projector after doing a black reading with the meter completely covered (on the floor or something).

EDIT: I looked back and looks like my memory is wrong. Tom does have a Trichromat. So, here is what I would suggest. Put the sensor on the floor or a table with a dark towel over it. Take a dark reading. Set it up maybe 6 to 8 feet from the projector. Make sure it is centered on the screen with the colorfacts background being the 0% black (so it isn't getting a ton of light right before it tries to take a "black" reading). Run the Colorfacts CR wizard. Tell us the number. Repeat later and see if you still get close to the same number. Maybe even from a different distance. You can try that other meter too, but I think .003 just has too much error built into it (it could really be .0026 or .0034 and you see .003).

--Darin
post #214 of 2404
Quote:


One thing, I could get a higher than 17.25 if I raised the meter up a foot or so higher in the area where the screen was. I got up near 19.00, but I figured I needed to be dead center.

Shouldn't matter in my opinion for this ratio measurment. As long as you don't move it between measurments, and it's pointing on-axis to the lens ("looking" straignt toward the lens).
post #215 of 2404
Thread Starter 
I've been using a black DVD case drapped on the Tri-chromat with a Black IRE to cut down light. maybe I should stuff it under a couch cushion.

I've heard the sensor is off at the high readings and I can train it with the one-eye. Another thing I have to do.
post #216 of 2404
darinp2,

Your trichromat meter even though it may display a value down to 0.001 does not mean that the data is reliable ! The trichromat can not read light levels accurately below approximately .07 fL which is substantially higher then what most of the users are reporting on this thread. For you to be able to get a value that is meaningful for CR you need to place the fixtured probe in a closer proximity to the light source so that your Black level reading is higher then .1 fL (the higher the better). As long as your peak White is not putting the probe into saturation (overload) then you are in good shape and the CR value obtained will be much closer to being a valid number.

You can probably take the reading from about 2-3 feet from the lens and get a fairly accurate reading without going into saturation.

Cliff
post #217 of 2404
Quote:


Originally posted by ghibliss
Your trichromat meter even though it may display a value down to 0.001 does not mean that the data is reliable !

Where did I say it did?
Quote:



The trichromat can not read light levels accurately below approximately .07 fL which is substantially higher then what most of the users are reporting on this thread.

What equipment did you use to find this point?

I always move the Trichromat closer to the projector, but I have not found that I need to go to 2-3 feet from the projector to get results that agree pretty well with other measurement devices (CA813 and Minolta LS110 lightmeters). I've of course used those at higher levels (closer to the projector for the CA813 and by putting the projector and screen close together for reading off the screen with the LS110) and not all the way at the screen as I have explained many times.

If the Trichromat needs .07 ft-lamberts for any CR accuracy then the EyeOne must need 1 ft-lambert.

--Darin
post #218 of 2404
Guitarman:

Quote:


You can't get an accurate CR reading with a colormeter from colorfacts

JimmyR gave you some bogus information. In fact, there's been a lot of careless talk about what ColorFacts can and cannot do. As Darin points out, using the Trichomat, ColorFacts offers quite reliable light readings. You have the Trichomat and a neat wizard. For Christ sakes, use the tool you have and stop agonizing over this. This is not rocket science. Just be sure to do your dark reading first and then get a good aim point. You can use the Continuous reading mode with the Luminance instrument. Put the Trichomat on a tripod facing an 80 IRE window and adjust it until you get the highest light reading. That's your aim point. Now just run the wizard. The whole process takes maybe 10 minutes.

I thought I'd provide some real data to the black hole that is this highly speculative thread. I took a CR reading with the meter approximately 5 feet away from the lens of my HS20. Here's the result:



Then I took another reading with the meter approximately 10 feet away. Here's what I got:



As you can see the result is almost identical. So, no you don't have to have a black reading above 0.07 to get reliable results. And no you don't have to set up any fancy Rube Goldberg device for channeling the light. And, no you don't need an extra light meter. Just use the product as designed and it works fine.

One thing I learned from this exercize is that my PJ is steadily loosing light output and CR. This is no great surprise, but it's annoying to have the numbers stare you in the face.

Now, in the name of all that's holy, would you PLEASE run the CR wizard on this PJ and report the number?
post #219 of 2404
Quote:


Originally posted by guitarman
Hi,
I don't build them but Wing does and he has all the facts from going to Ti's factory. I don't think he's making it up.

Assume your contact is Wing Chung who is the Product Engineering Manager at Optoma. He would seem to be the one who should know....
SJ
post #220 of 2404
"Everybody have fun tonight....Everybody Wing Chung tonight..."

Wow, that was a great '80's tune wasn't it? Oh no, that was "Wang Chung", not "Wing Chung"!!! (Sorry, couldn't resist )
post #221 of 2404
Thread Starter 
I missed my window of opportunity, but did mess around with testing this morning. I can get things pretty dark when daylight comes but can still see around the rooms so light is in the area.

Ok the light meter I found a spot where it didn't max out, it's about 5 feet away from the lens. I steadied the meter with the tripod and Trichromat mount. I got steady reading of

100IRE 16.70
Black .001

I couldn't move the setup any close without white maxing out, at least at the 5ft distance I could read both. I think that's what's wanted two light readings at the same spot.


Here's what the wizard came up with using th Tri-chromat.

http://www.cigarbest.com/sales/h79trichromatcr.jpg

I try it again sometime at night without the kitchen and dining rooms windows daylight. The sensors are facing that way. I'll probably get a better low light reading. Not bad for the situation though, low read 0.347
post #222 of 2404
The Trichomat reading of 2710:1 sounds about right. You might get a little better reading once the sun goes down.

On the other hand, the light meter reading of 16,700:1 is ridiculous. Don't know what's going on there.

Thanks for the info.
post #223 of 2404
Thread Starter 
Quote:


Originally posted by SJHT
Assume your contact is Wing Chung who is the Product Engineering Manager at Optoma. He would seem to be the one who should know....
SJ

One in the same. Wing is in Taiwan again right now, no questions for a couple of weeks. He's been a great help to me always friendly and eager to answer questions.
post #224 of 2404
Thread Starter 
Quote:


Originally posted by TomHuffman
The Trichomat reading of 2710:1 sounds about right. You might get a little better reading once the sun goes down.

On the other hand, the light meter reading of 16,700:1 is ridiculous. Don't know what's going on there.

Thanks for the info.

It's the low reading. I did better when I had the meter closer and got .003 or .002 but then I couldn't get the white reading that close.
post #225 of 2404
Tom, I know advanced adjustments of color temp, RGB contrast and brightness are available under "Image" through the standard menu, but do you have access to the service menu? Or is there nothing in the service menu an ISF service person would need?

I'm getting a H79 next week and after doing some calibration will post what I can.

Dan
post #226 of 2404
Tom,

Can you post what the lumens and contrast ratio numbers are for the high-lamp mode with the white peaking turned off.

Thanks!

Frank T. Lee
post #227 of 2404
I was trying out my CF after training it against a friends eye-one, and got similar results to Toms - the closer I moved the meter, the higher the CR reading was.

I could get a consistent 2600:1 with the light meter though.

Gary.
post #228 of 2404
Tom,

Is that 2710:1 read with perfect D65 calibration?

Thank you.
Nacho.

P.D.: Now I would like to get from you the H31 Contrast Ratio
post #229 of 2404
Thread Starter 
Quote:


Originally posted by drapp1952
Tom, I know advanced adjustments of color temp, RGB contrast and brightness are available under "Image" through the standard menu, but do you have access to the service menu? Or is there nothing in the service menu an ISF service person would need?

I'm getting a H79 next week and after doing some calibration will post what I can.

Dan

Not sure of the button names but looking from the back to the front top, the four buttons in a row you hit the left two and the far right one at the same time.
In there theres, ADC for progressive scans this one puts a small set of RGB's up in the far left corner out of the way, but ADC will alter the next set of RGB's up which is in there and called Picture. There's more under DLP but these are usually set flat and the same on all the projectors.

I was told they all work off different electronics in the chain line. Me I left them alone and user the Image RGB's deleting the centered up menu for each change. Slows you down a bit though. If you try the ADC of course write down the stock numbers for ADC, DLP, Picture, Picture will be different for each signal, not sure about ADC it may stay the same, DLP will stay the same.
post #230 of 2404
Thread Starter 
Quote:


Originally posted by nachin
Tom,

Is that 2710:1 read with perfect D65 calibration?

Thank you.
Nacho.

P.D.: Now I would like to get from you the H31 Contrast Ratio

Yes I pre calibrated to 6500k and got a 2.26 gamma I believe. That figure should go higher when I take a low reading at night. Curtains were drawn but there was a decent amount of daylight in the back rooms. Plus the sensor was aiming that way toward the projector lens.
post #231 of 2404
Quote:


Originally posted by ftlee
Can you post what the lumens and contrast ratio numbers are for the high-lamp mode with the white peaking turned off.

If the CR changes much at all when you go from low lamp mode to high lamp mode it is a pretty good indication that there is something wrong with your measurements. The only thing that should change CR here at all is a different color balance from the bulb, since CR is a ratio and even if you made the bulb twice as bright the ratio between 100% and 0% should be the same.

--Darin
post #232 of 2404
I've been lurking on this thread fascinated at the large amount of effort being spent trying to get this idea of an ideal measurement of CR which to my mind has very little to do with real world conditions. I think the important task is to determine the level of whites in high key scenes with no white compression vs. the level of blacks in low key scenes with no black compression.

Here are two jpeg files to use when making this CR measurement. One is a high key TP with gray-scales ranging from 80-100% white in 5% steps. The other is a low key TP with gray-scales ranging from 0% to 20% white also in 5% steps. The key to using these charts is having a spot brightness meter to measure the center brightness of each chart, plus setting the projector controls to insure all steps are visible. It is possible that I selected too large increments for the chips, but I'd like to have your feedback on this technique.
LL
post #233 of 2404
I couldn't figure how to attach more than one file at a time:
LL
post #234 of 2404
Thread Starter 
Those are Ftcl readings from the light meter. I could get a steady 4.50 very close to the lens but no high readings till I backed up. I use blackout curtains in the main room and double curtains in the adjoining rooms so it's really not very bright in day for testing. But there's some light. I'll see if I can get the low reading down in pitch black enviroment next time.
Anyway the H79 is the brightest projector I've seen so far, thankfully the black level is still good.
post #235 of 2404
barrysb

The idea is not to invent a new measurement - but verify it against the marketed on/off measurement.

If you can find the setting that achieves the marketed measurement - then maybe it can be optically filtered to correct the greyscale - rather than losing contrast to digital correction.
post #236 of 2404
Quote:


Originally posted by barrysb
I've been lurking on this thread fascinated at the large amount of effort being spent trying to get this idea of an ideal measurement of CR which to my mind has very little to do with real world conditions.

I'm not sure why you would say on/off CR has little to do with real world conditions. You could probably go into a room or closet in your place with no visible light right now and this will give you an idea of how realistically the projector could do that. If you want to know how much CR you can see in real life, take a look at page 3 here.
Quote:



It is possible that I selected too large increments for the chips, but I'd like to have your feedback on this technique.

This just looks like a test that is in between the ANSI CR test and the on/off CR test, since there is some washout from brighter parts of the image when you put more than just an all 0% screen up. I don't think it is a bad test, but it doesn't tell you as much about how realistically the projector can do blackout scenes as the on/off CR test does. Seems a lot like the 1% ANSI CR test I proposed, although I wasn't suggesting replacing either the on/off CR test or the standard ANSI CR test with it.

If you want to know how the shadow detail is you can use a test image for that, but I don't see any need to use that same test image for doing the on/off CR test (especially since it is no longer the on/off CR test).

Given that we are talking about a super high ANSI CR projector here (a single chip DLP) that isn't super high on/off your test wouldn't actually be much different than the correct on/off CR test, but that wouldn't be true on all projectors. So, if you want to compare them with numbers you would have to get everybody to stop testing full screen 0% and start testing with a greyscale on much of the screen (and you would no longer get some information you get now).

--Darin
post #237 of 2404
Since it is what effects shadow detail, do we know what the ANSI CR is of the H79?

krasmuzik,

Do you know what the ANSI CR is for the IF7205?

Thanks,

Frank T. Lee

P.S. I am trying to decide if it is worth selling my new 7205 and upgrade to the H79...
post #238 of 2404
ftlee

SP7205 measures are in my review in my .sig

If you turn on white peaking you get marketed contrast ~2200:1 - with it off you get half marketed ~1100:1

But if you turn on white peaking you get twice marketed brightness (1800 lumens) - without it you get over marketed brightness. (1100 lumens) (spec is 880 in low power lamp)

Sounds like the H79 is maybe half the marketed brightness without white peaking and in film mode - but is higher contrast tha

Bottom line I would not do a SP7205 upgrade to a H79 without also upgrading to a higher gain screen. While it's high power rating is the same as the SP7205 low power rating - you cannot compare since Infocus gives calibrated rating - Optoma does not. (Which is why the interest in calibrated results in this thread!)

While I have measured ANSI contrast - it is not relative to the projector as I am really measuring the room. Every room is different. You would have to hide under black velvet to get a proper ANSI contrast measure.

ON/OFF contrast is independent of the room (assuming no other light source)
post #239 of 2404
I measured 2800:1 contrast ratio on the H77 at D65. I used the extech light meter that came with Smart3 calibration set.
post #240 of 2404
Quote:


Originally posted by guitarman
Those are Ftcl readings from the light meter. I could get a steady 4.50 very close to the lens but no high readings till I backed up.

Tom,

It sounds to me like your chances of getting an accurate CR with that light meter with one pass as slim to none. You may want to try the trick of measuring 100 IRE and 20 IRE at the screen and then 20 IRE and 0 (or 7.5) IRE close to the projector. I don't think that meter has enough range for one pass (since it saturates at the top when the black reading still isn't within reasonable accuracy). Then it might be interesting to compare the 2 pass measurements with what Colorfacts gives with the Trichromat. If you just gave us the 4 readings for 100 IRE, 20 IRE, 20 IRE, and 0 IRE (with the first two and the screen and the 2nd two near the projector) we could tell you what that meant for CR.

Just make sure that your 0 IRE reading with the light meter isn't too low (readings of 0.001 tell you close to nothing because your error could be bigger than your reading).

--Darin
New Posts  All Forums:Forum Nav:
  Return Home
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - $3,000+ USD MSRP › Optoma H79 review & screenshots