or Connect
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - $3,000+ USD MSRP › BenQ W 5000 little Test
New Posts  All Forums:Forum Nav:

BenQ W 5000 little Test - Page 161

post #4801 of 5147
Quote:
Originally Posted by Andreas View Post

Thx Steve for the details. In the Cine4home LED test it was mentioned Vivitek, belonging to Delta I think, used a Planar engine for their LED Beamer. That got me all thinking....and I recalled SIM also sources there. Anyway, not really important, as long as we compare DC2 with DC2 and DC3 with DC3

On the hardware up-date, my W5000 arrived today a few minuts ago at work (4 weeks and 6 days away), I saw some of my remarks where answered with "problem not found" or "problem found, but no solution". So I need to see what is really cured and what is left. In some cases the DMD mainboard is changed to a newer one by the service partner (part 5D.05Q36.0019, may not be a BENQ number.... and text says "PCBA DMD BD PRJ W5000").

Same problem here. Mine was 11 months old after the picture showed stripes on the right side of the image. Projector was fixed (received the unit yesterday) under the warranty (36 months) and the service description says:

Error code: Short Circuit (Q)
Remedy Code: Module replacement (A2)
Section Code: Systemkontroll / Logic del (SYS)

Part: 5D.05Q36.001
Description: PCBA DMD BD PRJ W5000
"Replacement DMD board"

Does anybody knows if the new replacement DMD board fixes the original problem, which might be the result of overheating?
post #4802 of 5147
Quote:
Originally Posted by SteveMo View Post

My latest results by eye with at least 100 hours on the new lamp.

Brightness 55
Contrast 50
Color 47
Tint +9
Sharpness +4
Gamma 2.2
Iris 0
BC off
DB on

Desired R
664
329
+1410

Desired B
148
58
+1134

Desired C
232
313
+1075

Desired M
281
141
+1461

Desired Y
410
490
+1384

Desired W
294
315
+1000

Warm
Red Gain 490
Green Gain 432
Blue Gain 357
Red Offset 514
Green Offset 514
Blue Offset 507

I will be ordering a spectroradiometer next.

Steve,
What are these desired settings and where do you tune them? Also the gain and offset - is it the color management settings (dont you have to set it to user1 or something to do those changes). Can you explain a bit about where to do color settings?
post #4803 of 5147
Quote:
Originally Posted by snowkirk View Post

Steve,
What are these desired settings and where do you tune them? Also the gain and offset - is it the color management settings (dont you have to set it to user1 or something to do those changes). Can you explain a bit about where to do color settings?

I'm using the ISF menu to input these settings into ISF night. The warm values are in the service menu areas color temp. These values I borrowed from Big Lebowski then set luminance, chroma (saturation), and contrast to give the most pleasing image and one that best represented what I am used to viewing. I also adjusted values based on my own personal knowledge of my screen, (slightly adjusting for green as well) against a second display, and then checked my results against a known reference (mechman's results). In a few days I will be using CalMAN and a ColorMunki to check results, then do some more work on it.
post #4804 of 5147
Quote:
Originally Posted by Sulander View Post

Does anybody knows if the new replacement DMD board fixes the original problem, which might be the result of overheating?

Good question, as you wouldnt want it to happen again, but this time out of warranty!
I hope they replace it with something thats been upgraded, as im having mine also fixed atm.
post #4805 of 5147
I was really busy for a couple of weeks and didn't have a spare minute to turn on the W5000 and watch any video. Then I went away for a week and the only TV I saw was flat panels in hotels and an old small CRT TV. Yesterday I turned on the W5000 to watch some playoff football. My first reaction to the picture was "wow, I forgot how good that looks and how sharp the picture is."

I think I'll hang on to the Benq for a while longer.
post #4806 of 5147
Hello guys. I have mounted my projector on the ceiling. And when i move the picture lower to the floor with lens shift the lower part of the screen gets a slight arc shape. See my drawing. The more lens shift i use the more arc it gets. Does anyone know why this happends? Should i send it to service? I am going crazy here. Please help
LL
post #4807 of 5147
I think that's normal distortion and it is why people try to mount the projector so that the lens shoots dead center of the screen.

Maybe someone will correct me here, or explain why it happens.
post #4808 of 5147
off topic but just to chime in quickly in case it helps clarify something for somone:

Quote:


I tired briefly to find if their was some kind of master list of specs for DVDs like I can find to BD, but no luck. So I guess it is best to try both ways and see which results work best in any given situation.
I would guess just try both ways and use what ever seems to work best.

I don't know the actual decimal figures for the frame-rates for video storage on "NTSC" flagged DVDs, but there are basically two video storage schemes.

The first is for native 24 fps film source. The frames are split into field pairs (rather than being stored as complete frames) and then laid down in sequence with flagging that identifies the "pairs". So think of it as "48 fields per second" that get stored on the disc, which would equate to 24 p if we thought of those fields as being combined back into their original pairs. This is the most pure way to store 24 fps film material on the DVD format, and some "progressive scan" DVD players actually used the meta-data flagging to identify the field-pairs for frame reconstruction. The problem is that many DVDs have the field-pair flags improperly coded, and so these trusting 480p DVD players would zip together the wrong fields on some DVDs causing pretty bad combing. For this reason, most progressive-scan DVD players actually allow the MPEG decoder to output the NTSC-compatible 480i signal with the 3-2 already applied to the 480i 60 stream, and then use 3-2 pulldown detection to reverse and reconstruct the frames and repack into a 480p 60Hz output.

The other way video can be stored on DVD is set up for native 60i video... so basically 60 fields a second. Some DVDs from film source material are mistakenly mastered in this 60 fields-per-second form masquarading as NSTC video. However, a progressive-scan player using 3-2 detection would catch that coming out of the MPEG decoder and apply proper progressive-scan reconstruction. The only waste in that case would be that more storage was used to store 60 fields a second when only 48 fields a second literally needed to be recorded onto the DVD disc to handle the 24 fps material.

I'm not sure if the DVD format (for NTSC playback) has an option for 30 frames per second progressive, but it might.
post #4809 of 5147
Quote:
Originally Posted by evgtun View Post

Hello guys. I have mounted my projector on the ceiling. And when i move the picture lower to the floor with lens shift the lower part of the screen gets a slight arc shape. See my drawing. The more lens shift i use the more arc it gets. Does anyone know why this happends? Should i send it to service? I am going crazy here. Please help

A small amount should be normal. My W5000 is ceiling mounted 16'6" from my 106" diag. screen with the lens center being 12" from the ceiling. The top of my screen is 18" down from the ceiling. My "arc shape" is less than a 1/4". How big is your's?

Jason
post #4810 of 5147
Here are results of my first calibration. I used the ColorMunki auto calibration feature and the projectors CMS. For the automatic calibration I used 2.2 gamma D65 Rec. 709. I calibrated the CMS with the automatic calibration feature on, ran the automatic feature again, and then did touch up to the magenta in the CMS. There is also a D50 gamma 1.8 gamma option to tryout.

Brightness 55
Contrast 50
Color 47
Tint +9
Sharpness +4
Gamma 2.2
Iris 0
BC off
DB on

Desired R
653
327
+1410

Desired G
274
604
+1020

Desired B
148
58
+1134

Desired C
215
339
+1075

Desired M
303
156
+1461

Desired Y
429
490
+1384


Warm
Red Gain 469
Green Gain 449
Blue Gain 421
Red Offset 514
Green Offset 514
Blue Offset 507

This was the result of my automatic calibration with the previous settings. I needed to adjust the RGB gain to get a proper measurement prior to this result which measured close to 5000K.



Here is the newer result. I noticed that the appearance of green on my desktop changed after the automatic calibration feature but I did not notice any other obvious changes.

post #4811 of 5147
Quote:
Originally Posted by SteveMo View Post

Here are results of my first calibration. I used the ColorMunki auto calibration feature and the projectors CMS. For the automatic calibration I used 2.2 gamma D65 Rec. 709. I calibrated the CMS with the automatic calibration feature on, ran the automatic feature again, and then did touch up to the magenta in the CMS. There is also a D50 gamma 1.8 gamma option to tryout.

Nice to see some serious tweaking in this thread for a change.

I assume using the ColorMunki auto calibration you do all the adjustments in the projector menu (which would be a better option)?
Did you touch ISF CCA desired W or measured values (those do affect results)?
post #4812 of 5147
Quote:
Originally Posted by Big Lebowski View Post

Nice to see some serious tweaking in this thread for a change.

I assume using the ColorMunki auto calibration you do all the adjustments in the projector menu (which would be a better option)?

I used the automatic calibration with your bc off settings for which I adjusted luminance. The first measurement above with the odd looking gamut was a greyscale result for which I measured a 80% white window. I found it easier to dial in while looking at the RGB color balance. I found that adjusting for RGB gain in the temp is very sensitive to the RGB color balance and greyscale results. I did CMS with that automatic calibration on. As I did the CMS I watched the pattern from the GetGray disk and the CIE triangle. There were moments when adjusting more or less had no effect, or they caused the pattern to become to light in saturation. I kept the X and Y values where there was the best result for both the pattern and CIE triangle. This was after CMS and before my second automatic calibration. After my second automatic calibration (shown above) I only adjusted for the magenta.



The automatic calibration measures what looks like around 20 or 30 colors and takes more than 5 minutes to complete. I'm pretty sure it's not doing any harm but have not done an a/b comparison. I still have not checked to see what difference the latest calibration has made but I will soon. The difference that I found after my first trial was very obvious.


Quote:


Did you touch ISF CCA desired W or measured values (those do affect results)?

I did not touch desired W from your settings but I left luminance at 1000. When I tried lowering or raising that I saw a change in the brightness which seemed to effect my contrast. I did not try that during measurements. I left measured values alone.
post #4813 of 5147
You might want to check Y-values for each color. From quickly browsing the charts they look pretty wrong to me. Otherwise looks very nice but I warn you beforehand that adjusting color brightness will alter everything...

Here are correct Y-values for Rec.709.
Code:
     x         y         Y 
R    0.6400    0.3300    0.2126         
G    0.3000    0.6000    0.7152         
B    0.1500    0.0600    0.0722         
Y    0.4193    0.5053    0.9278         
C    0.2246    0.3287    0.7874         
M    0.3209    0.1542    0.2848         
W    0.3127    0.3290    1.0000
It is advisable to use 75% brightness patterns.
post #4814 of 5147
Quote:
Originally Posted by Big Lebowski View Post

You might want to check Y-values for each color. From quickly browsing the charts they look pretty wrong to me. Otherwise looks very nice but I warn you beforehand that adjusting color brightness will alter everything...

Here are correct Y-values for Rec.709.
Code:
     x         y         Y 
R    0.6400    0.3300    0.2126         
G    0.3000    0.6000    0.7152         
B    0.1500    0.0600    0.0722         
Y    0.4193    0.5053    0.9278         
C    0.2246    0.3287    0.7874         
M    0.3209    0.1542    0.2848         
W    0.3127    0.3290    1.0000
It is advisable to use 75% brightness patterns.

I was not aware that Y was a standard for Rec. 709. I will check and see what I can do about Y but I'm guessing this will probably vary anyway since I am using RGB 32bit color. The color measurements I did for my GetGray disk were in the 75% brightness chapter (and say 75% above) except for 100%W.
post #4815 of 5147
Quote:
Originally Posted by SteveMo View Post

I was not aware that Y was a standard for Rec. 709.

Y (brightness) is not just for Rec.709, all standards have defined Y-values.

Quote:


I will check and see what I can do about Y but I'm guessing this will probably vary anyway since I am using RGB 32bit color.

It does not matter you are using RGB 32bit color, Y is simply brightness of color. As long as Y is wrong it does not make colors right when only x and y coordinate is correct, you need to match Y too.

Quote:


The color measurements I did for my GetGray disk were in the 75% brightness chapter (and say 75% above) except for 100%W.

Sounds right. Just make sure you then calculate color brightness from 75% white. Or simply use all 100% patterns but I prefer 75%, gives more accurate results in other levels in W5000.
post #4816 of 5147
Quote:
Originally Posted by Big Lebowski View Post

Y (brightness) is not just for Rec.709, all standards have defined Y-values.


It does not matter you are using RGB 32bit color, Y is simply brightness of color. As long as Y is wrong it does not make colors right when only x and y coordinate is correct, you need to match Y too.


Sounds right. Just make sure you then calculate color brightness from 75% white. Or simply use all 100% patterns but I prefer 75%, gives more accurate results in other levels in W5000.

So the standard for Y is not effected by chroma sampling. What is the difference I'm seeing between 16bit and 32bit?
post #4817 of 5147
Here is an example of the 16bit versus 32bit. I'm not sure what shot in Blu-ray would make as a good example, but I included some photo's of that as well.

16-bit settings




32-bit settings


post #4818 of 5147
Quote:
Originally Posted by SteveMo View Post

So the standard for Y is not effected by chroma sampling.

No, not to my knowledge anyway. Refer to this thread for more information about x, y and Y. http://www.avsforum.com/avs-vb/showthread.php?t=852536

As you can see from the examples of green color, the brightness of color (Y) makes a huge difference.

If the calibration program you use does not calculate the desired Y-values for each color, you have to calculate them yourself. Just refer to the Rec.709 table I posted yesterday. For instance you want to see desired value for green Y, you do following. Measure all colors and white using 75% brightness pattens. See your measured white Y value, say it is 15, then just multiply it with with green Y taken from the table I posted (15 x 0.7152 = 10,728) and you see what is your green brightness target. Notice that everything affects everything, so you have to measure and adjust all the colors and grey scale several, several, several times to reach target. I spent dozens of hours to reach where I am now (I'm still exhausted from the calibration sessions last spring ).

Quote:


What is the difference I'm seeing between 16bit and 32bit?

By selecting 16bit color you are reducing color depth, thus the banding in color gradiations. More information about color depth here http://en.wikipedia.org/wiki/Color_depth
post #4819 of 5147
Quote:
Originally Posted by Big Lebowski View Post

No, not to my knowledge anyway. Refer to this thread for more information about x, y and Y. http://www.avsforum.com/avs-vb/showthread.php?t=852536

As you can see from the examples of green color, the brightness of color (Y) makes a huge difference.

If the calibration program you use does not calculate the desired Y-values for each color, you have to calculate them yourself. Just refer to the Rec.709 table I posted yesterday. For instance you want to see desired value for green Y, you do following. Measure all colors and white using 75% brightness pattens. See your measured white Y value, say it is 15, then just multiply it with with green Y taken from the table I posted (15 x 0.7152 = 10,728) and you see what is your green brightness target. Notice that everything affects everything, so you have to measure and adjust all the colors and grey scale several, several, several times to reach target. I spent dozens of hours to reach where I am now (I'm still exhausted from the calibration sessions last spring ).


By selecting 16bit color you are reducing color depth, thus the banding in color gradiations. More information about color depth here http://en.wikipedia.org/wiki/Color_depth

I used 30 for greycale since I am using a lower 2.2 gamma. My meter does not recognize the color wheel so it might be hit or miss with luminance since the color wheel effects this. The RGB color balance is what is most critical as I am using RGB gamma correction. I'm pretty sure the automatic software calculates for both luminance and greyscale. It first measures 100W, then a variation of white what looks like 80, then black. After that it measures more than a few variations of each primary and secondary that start bright, and then get very dark until there is very little color to be seen left. Then it calculates and confirms the profile by measuring everything again.
post #4820 of 5147
I will try a second calibration ins ISF day sometime and see if I can't correct for luminance, or improve greyscale.
post #4821 of 5147
Sound like the program you are using is making an ICC-profile (or similar) for Windows. Basically that should work but the best solution would be that you use display's CCA controls. Using an ICC-profile reduces color depth. Also looking at data you posted here shows clearly that Y-value is all wrong for some colors.

If you want to get full potential out of your W5000 and the calibation program you use lets you do adjustments manually by using pj's ISF CCA controls, then you could make all colors right.
post #4822 of 5147
Here is my attempt at luminance.

Brightness 55
Contrast 50
Saturation 53
tint +9
Iris 0

Desired R
653
327
1860

Desired G
233
625
791

Desired Y
441
476
1692

Desired B
151
64
665

Desired C
200
335
833

Desired M
308
149
1446

Warm
Red gain 508
Blue gain 490
Green Gain 474







After making the profile





post #4823 of 5147
Now your color Y looks much better but gamma is weird looking. I had similar problem with gamma when I pushed gains too high. Try lowering them. Also I see that you have pushed CCA Y for red and yellow very high. That can cause weird looking image in certain scenes. It is better no to push them too hard.

It would be best not to use profiling, just do all the neccesary adjustments with pj's controls.
post #4824 of 5147
Quote:
Originally Posted by Big Lebowski View Post

Now your color Y looks much better but gamma is weird looking. I had similar problem with gamma when I pushed gains too high. Try lowering them. Also I see that you have pushed CCA Y for red and yellow very high. That can cause weird looking image in certain scenes. It is better no to push them too hard.

That is why brightness, contrast, and gamma are done first. I looked at 100% windows and checked the color bars when I was done.

Quote:


It would be best not to use profiling, just do all the neccesary adjustments with pj's controls.

What do mean? I did the calibration without the profile on, then created one. It can be turned off.
post #4825 of 5147
Guys, I owned a w6000 and had the same cca menu as w5000. Would like to ask if I only adjust desired value without touching measured value, it is ok? Looking at the value you guys enter, I can't make any relation link between measured and desired value. FYI, I'm using calman.
post #4826 of 5147
Quote:
Originally Posted by rjyap View Post

Guys, I owned a w6000 and had the same cca menu as w5000. Would like to ask if I only adjust desired value without touching measured value, it is ok? Looking at the value you guys enter, I can't make any relation link between measured and desired value. FYI, I'm using calman.

I did the calculations manually using the formula Big Lebowski posted. First I did luminance for green, red, yellow, cyan, magenta and then blue. Then I bumped the saturation until red luminance was closer. After that I did greyscale and checked red and blue again adjusting X and Y with some touch up to the luminance for blue because it became too high. I then did luminance again (still calculating each time) starting with yellow, then cyan, red, green, blue magenta. I looked at 100% saturation windows. I then did X Y values again and another pass at luminance again for red and yellow since I wanted to see what was happening there with greyscale. Then I did X Y values again for cyan and magenta watching the saturation and then adjusted luminance for them. After that I did luminance again in the order on the chart but I didn't do red or green. Then I measured everything again.
post #4827 of 5147
I read in Color Calibration thread that if you have CMS control, you no longer need to tune Saturation (Color) and Tint in Color Control. So I'm not sure it is necessary. We could use CCA to adjust the luminance for red until it is 21% of White without affecting other color.

If I redo, I'll use the following method.

1. Adjust Brightness/Contrast
2. Adjust greyscale.
3. Skip Color decoding (Saturation/Tint Control).
4. Jump to CCA and adjust Y first to ensure R,G,B,C,Y,M is within the targeted Y (luminance). (Under desired value).
5. Adjust x,y to make sure it fall within the 709 gamut.
6. Recheck the luminance again as adjusting x,y will cause the Y to drift. After done the Y, redo x,y again until it is as close as it get.
7. Recheck greyscale again as adjusting CCA do affect white point (W6000 does). Normally just a few notch up/down.
8. Recheck x,y,Y for each color to ensure no drift.

Any comment on above method?
post #4828 of 5147
If you adjust greyscale last that is going to effect your luminance. Notice everything effects luminance. Try and get your white point correctly aligned when doing greyscale. You might notice that as much as one click on the gains, which are actually cuts, will effect your white point. On my first graph showing the CIE chart notice the white point is at a different location than my second. I raised the red gain one click and then lowered it and it went back into the box. Your saturation is your gains. You will want to get your color that is most likely to clip out of range for clipping. On my screen this is red since that luminance is low but on yours it may be green or blue.
post #4829 of 5147
Quote:
Originally Posted by rjyap View Post

I read in Color Calibration thread that if you have CMS control, you no longer need to tune Saturation (Color) and Tint in Color Control. So I'm not sure it is necessary. We could use CCA to adjust the luminance for red until it is 21% of White without affecting other color.

If I redo, I'll use the following method.

1. Adjust Brightness/Contrast
2. Adjust greyscale.
3. Skip Color decoding (Saturation/Tint Control).
4. Jump to CCA and adjust Y first to ensure R,G,B,C,Y,M is within the targeted Y (luminance). (Under desired value).
5. Adjust x,y to make sure it fall within the 709 gamut.
6. Recheck the luminance again as adjusting x,y will cause the Y to drift. After done the Y, redo x,y again until it is as close as it get.
7. Recheck greyscale again as adjusting CCA do affect white point (W6000 does). Normally just a few notch up/down.
8. Recheck x,y,Y for each color to ensure no drift.

Any comment on above method?

1. Adjust Brightness/Contrast
2. Adjust greyscale for warm in sm.
3. Skip Color decoding (Saturation/Tint Control).
4. Measure 75% Saturation for R,G,B,C,Y,M
5. Jump to CCA and input measured 75% values in "measured values"
6. input values for rec. 709 75% saturation in "desired values"
7. Adjust x,y in desired values for R,G,B to make sure they fall on the 709 points.
6. Recheck the luminance again as adjusting x,y will cause the Y to drift. After done the Y, redo x,y again until it is as close as it get.
7. Recheck greyscale again as adjusting CCA do affect white point (W6000 does). Normally just a few notch up/down.
8. Recheck x,y,Y for each color to ensure no drift.

http://www.avsforum.com/avs-vb/showthread.php?t=1029594

http://www.avsforum.com/avs-vb/showthread.php?t=852536
post #4830 of 5147
This is the part where I don't really understand. I measured the R,G,B,C,M,Y and how I translate that into CCA?

As CCA is always in integer numbers 0 - 1000 for x,y then do I assume if I get 0.330 for x and 0.660 for y, I input as 330, 660 into the measured value? Also for Y, how can I scale this into CCA Y? Or do I need to get the 75% White Y as reference?

Example
If you look at Calman chart:

fL 75%W Red
x 0.3129 0.6521
y 0.3287 0.3395
Y 9.9959 2.5325
Tx 0.3127 0.6399
Ty 0.3290 0.3300
TY 9.9959 2.1256

Note: T is the target which should be desired value.

Should I enter measured value for White? I already get a perfect D65 for greyscale and if you look at my calman reading, it's perfect.

For Measured R, I assume I should enter
x 652
y 340
Y 253 ? What value should I input?

For Desired R
x 640
y 330
Y 213

Make sense? Thanks for your patience.
New Posts  All Forums:Forum Nav:
  Return Home
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - $3,000+ USD MSRP › BenQ W 5000 little Test