or Connect
AVS › AVS Forum › Display Devices › CRT Projectors › Gamma Correction?
New Posts  All Forums:Forum Nav:

Gamma Correction? - Page 4

post #91 of 174
Quote:
Originally Posted by arioch View Post

However - measuring low amplitudes can be very difficult, 0-20% can give very unreliable data, depending on instrument, screen gain, projector gain/contrast etc.

So you're saying that if we could measure the low amplitudes more accurately then we actually COULD adjust overall luminance without messing up a "correctly" set gray scale.......correct?

It might be a good idea to check to see that the gray scale is still in proper balance...however it likely won't change too much; if at allright?

Note: I'm sort of hammering on this because I'm trying to understand if an ISF'd system would need to re-ISF'd if the low end gamma is messed with. Is it REALLY needed?I think that could be big $$ for very little or no adjustment.
post #92 of 174
Quote:
Originally Posted by Brian Feldman View Post

Okay, I just spent the past 5 hrs with my colorimeter adjusting the G90 bias/gain and gamma settings. Here is what I have found out during the fine tuning of these adjustments.

The "gamma" circuit in the G-90 does not seem to affect the entire range as reported here. The most dramatic difference is indeed in the 20-0 IRE range; EXACTLY where it is needed!! While adjusting the gamma it absolutely throws off the bias/gain settings so it is imperitive to go back and fine tune all of these settings in order for everything to work well together and to get accurate D6500 tracking.


What I did was to adjust these settings to get as close as 6500 from 10-100 IRE which was NOT an easy task. Like I mentioned above every setting interacts with the other so even very small changes make large shifts to the grey scale tracking.

Just using the projector controls I was able to get my color temp at 6500 =- 400 degrees on the extreme ends of the scales. I then used my Lumagen HDQ and went into the 12 point correction menu and got the grey scale tracking pretty much perfect. It now sits at 6500 +- 100 degrees. What is great about the Lumagen is that you can add more shadow detail by moving around the gamma points at the 5-15 IRE settings. Since the Lumagen offers a parametric circuit, the Q of the circuit seems to be wide enough so that minor changes at a particular point carries over a bit.

I basically increased the lum (gamma) at 8 IRE and at 12 IRE by aprox 20 percent. I left the lum at the other points at their default values.

The picture quality is now truly impressive with a substantial amount of shadow detail visible. Before, I would have to reduce the brightness in order to get the "inky" blacks that I liked so much but in doing this, the shadow detail was just not there. Now, I get the fade to black AND shawdow detail in the image!!

Hey Brian,

I am using a Lumagen with my G90 in exactly the same manner you describe. I was able to lower "brightness" (black level) to 40 to get FTB for most material. The reasoning here was to bump the "shadow detail" way up and then bring it back to where it should be by lowering the "brightness" level and this is what gives me FTB while still maintaining proper detail in dark scenes. Once this was done the projector was properly calibrated. As you say one has to be careful with the amount of change to luma or things like posterization and banding will be accentuated - and this is largely dependent on the quality of the video material.

Cheers,

Grant
post #93 of 174
Quote:
Originally Posted by Deja Vu View Post

Hey Brian,

I am using a Lumagen with my G90 in exactly the same manner you describe. I was able to lower "brightness" (black level) to 40 to get FTB for most material. The reasoning here was to bump the "shadow detail" way up and then bring it back to where it should be by lowering the "brightness" level and this is what gives me FTB while still maintaining proper detail in dark scenes. Once this was done the projector was properly calibrated. As you say one has to be careful with the amount of change to luma or things like posterization and banding will be accentuated - and this is largely dependent on the quality of the video material.

Cheers,

Grant

Wow Grant, this is EXACTLY what the brightness on my G90's are set at.

Pretty damn cool.

Cliff
post #94 of 174
Sounds great Grant!! I recently received John's DVI input card for the G90 and I setup everything for this input. I left all of my custom RGB gain, bias and gamma controlls alone in the projector and did a bit more experimenting with the Lumagen.

Rather than shifting around the lower IRE settings in the Lumagen, this time after adjusting 100-0 RGB levels to get as perfect D6500 tracking, I set the 10 IRE to 7 and bumped the Lum (gamma) to 9.5 This setting on my projector is absolutely perfect and preserves excellent low level details and fades to absolute black without a hint of posterization or any banding issues.

I went back and forth in the projector settings by comparing my "custom" settings in the color temperature area to the 6500 setting and there is no comparison about the amount of low level detail that the gamma in the projector provides. For those that don't mind "tweaking" your own G90 I think that you would be impressed by the results.

I woud love to hear from the experts that setup the G90's professionally and see what they have to say about these gama settings. I feel that it is strange these adjustments are seldom mentioned.



Quote:
Originally Posted by Deja Vu View Post

Hey Brian,

I am using a Lumagen with my G90 in exactly the same manner you describe. I was able to lower "brightness" (black level) to 40 to get FTB for most material. The reasoning here was to bump the "shadow detail" way up and then bring it back to where it should be by lowering the "brightness" level and this is what gives me FTB while still maintaining proper detail in dark scenes. Once this was done the projector was properly calibrated. As you say one has to be careful with the amount of change to luma or things like posterization and banding will be accentuated - and this is largely dependent on the quality of the video material.

Cheers,

Grant
post #95 of 174
Quote:
Originally Posted by picree View Post

So you're saying that if we could measure the low amplitudes more accurately then we actually COULD adjust overall luminance without messing up a "correctly" set gray scale.......correct?

It might be a good idea to check to see that the gray scale is still in proper balance...however it likely won't change too much; if at allright?

Note: I'm sort of hammering on this because I'm trying to understand if an ISF'd system would need to re-ISF'd if the low end gamma is messed with. Is it REALLY needed?I think that could be big $$ for very little or no adjustment.

Well, if we're sure that the gamma adjustment is done equally on all the three primary colors outputs, then I guess no recalibration of the gray scale is needed.

However: I initally spoke out because I don't agree with the notion that gamma adjustments always cause artefacts, but:
Without the proper measuring equipment and some knowledge on this subject there is really no way of getting it "right", so the risk of actually getting artefacts when doing this on "freehand" is quite big.
A gamma adjustment that works for one given CRT-projector won't necessarily work for another.
If I have one that has a display gamma of 3.1 at 10-30% amplitude and I somehow adjust it to 2.5 at 10-30% and then try to implement the same adjustments to a projector that has 2.8 at 10-30% - then the second one won't be correct.
You can't just "borrow" the adjustments that someone else used.
The only way to know if Your gamma is correct is to get it ISF'd (either by a proffessional or by Yourself) - i.e. measuring it's output and adjusting until within tolerance.
post #96 of 174
What You are saying that You do with Your G90:s here is exactly what I do with my 1292 and Lumagen combo!
True fade to black and shadow detail down to 1-2% amplitude combined with D65 dE < 4 truly rocks!
What colorimeters do You guys have? I use an X-Rite/DTP-94.
post #97 of 174
I'd love to try one of the gamma devices that are the rage here to compare with what I'm doing now. Gamma correction is truly the greatest CRT tweak I know of. This is what sets CRT apart from digital - at least for now. If there was no "gamma correction" I'd own a digital projector and avoid the size, and setup issues. It always amuses me to hear guests gasp when the room goes totally black. The present crop of digital projectors are fantastic, but, and it's a big but, once you've seen total FTB in a light controlled environment you begin to understand that digital has an achilles heel or artifact and it is fade to glowing grey. It is amazing just how many FTB scenes there are.

Cheers,

Grant
post #98 of 174
A couple of weeks ago after a fresh Ken calibration on the stack I had a couple of the hotties from the sales office of my subdivision over to show them the setup.

They know NOTHING about CRT.

I popped in the NIN HD DVD and played a track that has AWESOME blacks and detail in the low end.

"Oh my God! It's so 3D!!!!" was the classic words they spoke.

Cliffy
post #99 of 174
Quote:
Originally Posted by Deja Vu View Post

I'd love to try one of the gamma devices that are the rage here to compare with what I'm doing now. Gamma correction is truly the greatest CRT tweak I know of. This is what sets CRT apart from digital - at least for now. If there was no "gamma correction" I'd own a digital projector and avoid the size, and setup issues. It always amuses me to hear guests gasp when the room goes totally black. The present crop of digital projectors are fantastic, but, and it's a big but, once you've seen total FTB in a light controlled environment you begin to understand that digital has an achilles heel or artifact and it is fade to glowing grey. It is amazing just how many FTB scenes there are.

Cheers,

Grant

I couldn't agree more! When I installed the tse box that was the very first thing I noticed...FTB...once you've seen black you will NEVER go back! I mean REALLY black. You can't see your hand in front of your face! The next thing I noticed of course was the increased richness in the colors since the underlying black levels are so much lower. It does help with the 3D effect.

I calibrated all the amps, preamps, and G2's. Right now I run about 40% on the brightness and about 50% on the contrast (calibrated w/ Avia) and even with the brightness that high I get complete blackness on FTB.

If I had a Lumagen, certainly I would adjust levels there but since I don't, the tse box is a great alternative! At a final cost of about $150 I couldn't resist.

Curt
post #100 of 174
Brian,

On my system I actually used my G90 gamma to tame mid ire level blue hump and ended up with numbers not too far off from yours.

What did you end for brightness and contrast?

What colormeter device and disk/pattern did you use for gray scale?

What source are you feeding the G90? Did you change any of the sources parameters ie. source brightness/contrast etc and if yes what parameters did you change and by how much?

I also noticed you used the Bias Fine. I never altered it from the 128 mark for all fine numbers as I did not think it would be that useful. Any comments here?

Thanks
Frank

Quote:
Originally Posted by Brian Feldman View Post

Frank, here are my settings on my G90. Color Temperature must be set to "Custom" in order to modify the gamma settings.


GAIN

R: 188
G: 235
B: 190

Bias Course

R: 124
G: 138
B: 138

Bias Fine

R: 115
G: 127
B: 132

Gamma

R: 200
G: 200
B: 70


Remember that these are the settings that I ended up with and will probably vary quite a bit from proj to proj.
post #101 of 174
Hello Frank,

On my system I have my brightness set at 36 and my contrast set at 50.

I use a Progressive Labs CA-1 colorimeter to take my measurements.

As far as my source material, I use a dishnetwork 941 HD DVR for my network programming, a Toshiba XA-20 HD DVD player and a AVS HTPC for internet and standard dvd playback.

The internal gamma adjustments made a huge difference in providing a lot more low detail and still maintaining the deep dark blacks that us crt owners crave!!








Quote:
Originally Posted by Frank D View Post

Brian,

On my system I actually used my G90 gamma to tame mid ire level blue hump and ended up with numbers not too far off from yours.

What did you end for brightness and contrast?

What colormeter device and disk/pattern did you use for gray scale?

What source are you feeding the G90? Did you change any of the sources parameters ie. source brightness/contrast etc and if yes what parameters did you change and by how much?

I also noticed you used the Bias Fine. I never altered it from the 128 mark for all fine numbers as I did not think it would be that useful. Any comments here?

Thanks
Frank
post #102 of 174
Brian,

Did you change any of the parameters (brightness/contrast etc) in the XA-20?

Your contrast is a bit lower then my number.

What screen material and gain are you using?

Do you have your foot lamb. reading at the 100 ire widow from your CA-1 colorimeter?

Thanks
Frank

Quote:
Originally Posted by Brian Feldman View Post

Hello Frank,

On my system I have my brightness set at 36 and my contrast set at 50.

I use a Progressive Labs CA-1 colorimeter to take my measurements.

As far as my source material, I use a dishnetwork 941 HD DVR for my network programming, a Toshiba XA-20 HD DVD player and a AVS HTPC for internet and standard dvd playback.

The internal gamma adjustments made a huge difference in providing a lot more low detail and still maintaining the deep dark blacks that us crt owners crave!!
post #103 of 174
Earlier in this thread I uploaded a simple Excel file for gamma adjustments. I also, recalled that initially when we tried adjusting gamma the results were overstated (someone in this thread mentioned that the values I put in the file above were too agressive...confirming our initial results). I couldn't find my original spreadsheet that might help explain what we finally did to correct our results (I remembered there was some trick to translating the tse curve but I couldn't remember how we finally did it).

Well, I found the spreadsheet and graph we used...

We realized that if we were going to use the graph provided by tse we had to make sure it was scaled properly in order to get the right results. Also notice that the graph uses millivolts such that 0-100 IRE is 0-700 mV (or -100-600 Vout). It's not a 1:1 chart. That throws the perspective off.

Attached are the final figures we used in a Lumagen HDP 11-point gamma curve to adjust gamma for a Sony G90. I'm very interested in how these numbers match up with what others may be using. I would also like to know if someone might like to try these settings and report back on the results. The numbers we used are in blue. The numbers in red are the values you should change if you want to tweak the results (note your original settings so you can go back!).

I also included the tse .jpg curve we used. On it I marked in red what we did with the original gamma curve to arrive at a scaling factor. Using the crude graph we picked a point that was at about 10 IRE (12.5 actually) and measured the amount of correction in ABSOLUTE terms (key...namely absolute millivolts). Thusly, the amount of correction to apply was about 20% of what we first charted (0.5 boxes/2.5 boxes...taking into account the slight offset in tse's chart).

So, 10 IRE only gets bumped to about 12...etc. This method may be unconventional but the results looked pretty good. I'm curious how close it comes to a professional setup...Can anyone comment (even in general terms) on the accuracy here...or is this a bunch of hogwash?

Curt

What we used:

LUMA....New
0........0
10......12
20.....22.2
30......32
40.....41.7
50.....51.4
60.....61.2
70.....70.9
80.....80.6
90.....90.4
100.....100

 

Curve used to calculate gamma shift.zip 30.9619140625k . file
post #104 of 174
All of the parameters in the XA-20 are set at 0 or at their default levels.

My screen is the Wilsonart "designer white" laminate and is aprox 90" wide with a 1:85 A/R. There are literally hundreds of threads on this material and I believe the gain of this screen is somewhere around 1.20 or so.

With a 100 IRE window, I just tested this and I am getting 14.2 Ft.lmb according to the CA-1.



Quote:
Originally Posted by Frank D View Post

Brian,

Did you change any of the parameters (brightness/contrast etc) in the XA-20?

Your contrast is a bit lower then my number.

What screen material and gain are you using?

Do you have your foot lamb. reading at the 100 ire widow from your CA-1 colorimeter?

Thanks
Frank
post #105 of 174
Your numbers in the Lumagen look good though they are significantly different from what I ended up with.

I was able to boost the gamma in the G90 projector so I just wanted a little more boost on the lower end of the IRE. What is really cool about the Lumagen is that once you go into the 11 point adjustment window, you can have it display what the IRE value is present while watching source material. What I noticed is that the low level information that I want boosted was around 5-8 IRE. What I did was to replace the 10 IRE with 7 and boost the Lum (gamma) value from 7 to 9.5. I left all the Lum values from 20-100 at their default settings which btw should be the same numerical value as the actual IRE number. For example 20 IRE should equal 20 Lum and so on.

The extra boost at 7 IRE proivdes me with just enough additional shadow detail but not too much as to make everything look to grey. I still get a near perfect fade to inky black and good overall low level detail.

A source that I use to test the shadow detail and fade to black is the movie Shawshank Redemption. In the scene where the prisoners are brought into a room to meet the warden, when the warden first comes into view I pay attention to the wardens jacket. If the gamma is set correctly you should just be able to see the color and texture of the jacket. If the gamma is set too low you will not see the jacket well and if it is set too high there will be excessive background noise and the fade to black will be poor.



Quote:
Originally Posted by picree View Post

Earlier in this thread I uploaded a simple Excel file for gamma adjustments. I also, recalled that initially when we tried adjusting gamma the results were overstated (someone in this thread mentioned that the values I put in the file above were too agressive...confirming our initial results). I couldn't find my original spreadsheet that might help explain what we finally did to correct our results (I remembered there was some trick to translating the tse curve but I couldn't remember how we finally did it).

Well, I found the spreadsheet and graph we used...

We realized that if we were going to use the graph provided by tse we had to make sure it was scaled properly in order to get the right results. Also notice that the graph uses millivolts such that 0-100 IRE is 0-700 mV (or -100-600 Vout). It's not a 1:1 chart. That throws the perspective off.

Attached are the final figures we used in a Lumagen HDP 11-point gamma curve to adjust gamma for a Sony G90. I'm very interested in how these numbers match up with what others may be using. I would also like to know if someone might like to try these settings and report back on the results. The numbers we used are in blue. The numbers in red are the values you should change if you want to tweak the results (note your original settings so you can go back!).

I also included the tse .jpg curve we used. On it I marked in red what we did with the original gamma curve to arrive at a scaling factor. Using the crude graph we picked a point that was at about 10 IRE (12.5 actually) and measured the amount of correction in ABSOLUTE terms (key...namely absolute millivolts). Thusly, the amount of correction to apply was about 20% of what we first charted (0.5 boxes/2.5 boxes...taking into account the slight offset in tse's chart).

So, 10 IRE only gets bumped to about 12...etc. This method may be unconventional but the results looked pretty good. I'm curious how close it comes to a professional setup...Can anyone comment (even in general terms) on the accuracy here...or is this a bunch of hogwash?

Curt

What we used:

LUMA....New
0........0
10......12
20.....22.2
30......32
40.....41.7
50.....51.4
60.....61.2
70.....70.9
80.....80.6
90.....90.4
100.....100
post #106 of 174
Quote:
Originally Posted by Brian Feldman View Post

What I did was to replace the 10 IRE with 7 and boost the Lum (gamma) value from 7 to 9.5. I left all the Lum values from 20-100 at their default settings which btw should be the same numerical value as the actual IRE number. For example 20 IRE should equal 20 Lum and so on.

The extra boost at 7 IRE proivdes me with just enough additional shadow detail but not too much as to make everything look to grey. I still get a near perfect fade to inky black and good overall low level detail.

A source that I use to test the shadow detail and fade to black is the movie Shawshank Redemption. In the scene where the prisoners are brought into a room to meet the warden, when the warden first comes into view I pay attention to the wardens jacket. If the gamma is set correctly you should just be able to see the color and texture of the jacket. If the gamma is set too low you will not see the jacket well and if it is set too high there will be excessive background noise and the fade to black will be poor.

Thanks for the reply. When we set this up we didn't know (at least I didn't) that you could put any Lum value we wanted into the Lumagen. Thought it was fixed at 10, 20, .....in any case we ended up with about the same magnitude shift as what you got (up 2 units in the <</u>10 range).

I've read that it is important to make sure any shift in gamma settings are smooth over the entire IRE range. Maybe...maybe not? Probably hard to discern in a movie but you may want to feather the gamma increase out instead of just leaving 20-100 IRE at their defaults.

I'll check out that scene if I can find that disc.
post #107 of 174
So no one has modified their card yet?

I received my new card today. Works great!

We watched World Trade Center on HDDVD. Plenty FTB and dark scenes to show of the gamma circuit. Really looks good.

I didn't notice any banding in this movie. In some dark scenes, I can see the artifacts in some dark scenes elevated a bit, but this was very dependant on specific scenes, and I was specifically looking for this. I have not yet decided whether or not I'll reduce the gamma bump yet. Need to view some more material to see if I think it is worth the trouble.

So Moome,

What resistors to change out on V1.32 card to reduce gamma bump?
post #108 of 174
I have experimented with changing the values in the Lumagen and you definately want to be gentle with the changes. Especially in the lum (gamma) settings.

I have found that if you change the lum value too much or group together several of the IRE values too close, then there will be signifcant banding and other artifacts introduced into the image.

The best results that I was able to obtain was to simply substitute the 10 IRE with 7 and raise the Lum value to 9.5 I left the 20-100 IRE values at their default lum values. This along with the gamma adjustments that I made in the G90 has netted me the best results. The gamma is not too aggressive where it washes out the image but there is just enough to prevent any black crush and provide subtle details to very dark scenes. The fade to black in an image is superb.

Watching Star Wars the other night was awesome. When they display the star field with nothing else in the image the effect is like you are literally floating in space. No grey appearance or washed out look, simply inky black background with pinpoints of stars.





Quote:
Originally Posted by picree View Post

Thanks for the reply. When we set this up we didn't know (at least I didn't) that you could put any Lum value we wanted into the Lumagen. Thought it was fixed at 10, 20, .....in any case we ended up with about the same magnitude shift as what you got (up 2 units in the <</u>10 range).

I've read that it is important to make sure any shift in gamma settings are smooth over the entire IRE range. Maybe...maybe not? Probably hard to discern in a movie but you may want to feather the gamma increase out instead of just leaving 20-100 IRE at their defaults.

I'll check out that scene if I can find that disc.

post #109 of 174
Quote:
Originally Posted by Brian Feldman View Post

I have experimented with changing the values in the Lumagen and you definately want to be gentle with the changes. Especially in the lum (gamma) settings.

.......

The best results that I was able to obtain was to simply substitute the 10 IRE with 7 and raise the Lum value to 9.5 I left the 20-100 IRE values at their default lum values. This along with the gamma adjustments that I made in the G90 has netted me the best results. The gamma is not too aggressive where it washes out the image but there is just enough to prevent any black crush and provide subtle details to very dark scenes. The fade to black in an image is superb.

Watching Star Wars the other night was awesome. When they display the star field with nothing else in the image the effect is like you are literally floating in space. No grey appearance or washed out look, simply inky black background with pinpoints of stars.

Your description makes me smile I'll mention this to my neighbor for his G90. That may be why he was inclined to set the Lumagen back to flat line (the tse curve we replicated may have been too aggressive for his liking). He may be more inclined to just make the small tweak that you made.

Side comments on the tse box/gamma curve. I use a tse box on my BG808 for gamma boost. Some say it's too aggressive for a Barco. I also have an external moome transcoder. Some have said variously that the moome crushes black's or in other cases washes out the colors In any event I did notice that the gamma results one gets from a tse and/or moome depends upon how they are connected (interplay of all the little amps I suppose). For me the moome completely washed out the image when I directly connected the it into a tse box...horrid. So I switched things up. The best combo I have come up with so far (to get inky blacks and vivid colors) is:

In the stack:
Oppo DVD==>DVI 2 foot cable out
DVI in to Moome==> RGB 1.5 foot cable out
RGB in to Extron==> 25 foot RGBHV snake out

At the projector:
RGBHV in to tse box==> 1.5 foot RGBHV cable out
RGBHV in to Barco

Complicated and messy I know (purists probably hate me) but I think the results are really sharp with great color reproduction...inky blacks too. I've compared it to my neighbors G90 as a reference. True, his is more vivid, richer looking, razor sharp, with a jaw dropping gray scale! But I'm not too far off from that. Still impressive on HD.

Thoughts
1) Oppo DVI out-using DVI reduced the static and noise in my video signal by a factor of at least 10.
2) Moome-simply transcodes to RGB but can't drive a strong signal
3) Extron-drives the signal up to the Barco to make sure its 700 mV (scoped and it is)
4) tse box-for gamma boost.
5) Getting the Barco amps (scoped) and G2's just right was also really important.

Thats what I've found out by messin' around. FWIW
post #110 of 174
Quote:
Originally Posted by picree View Post

Your description makes me smile I'll mention this to my neighbor for his G90. That may be why he was inclined to set the Lumagen back to flat line (the tse curve we replicated may have been too aggressive for his liking). He may be more inclined to just make the small tweak that you made.

Side comments on the tse box/gamma curve. I use a tse box on my BG808 for gamma boost. Some say it's too aggressive for a Barco. I also have an external moome transcoder. Some have said variously that the moome crushes black's or in other cases washes out the colors In any event I did notice that the gamma results one gets from a tse and/or moome depends upon how they are connected (interplay of all the little amps I suppose). For me the moome completely washed out the image when I directly connected the it into a tse box...horrid. So I switched things up. The best combo I have come up with so far (to get inky blacks and vivid colors) is:

In the stack:
Oppo DVD==>DVI 2 foot cable out
DVI in to Moome==> RGB 1.5 foot cable out
RGB in to Extron==> 25 foot RGBHV snake out

At the projector:
RGBHV in to tse box==> 1.5 foot RGBHV cable out
RGBHV in to Barco

Complicated and messy I know (purists probably hate me) but I think the results are really sharp with great color reproduction...inky blacks too. I've compared it to my neighbors G90 as a reference. True, his is more vivid, richer looking, razor sharp, with a jaw dropping gray scale! But I'm not too far off from that. Still impressive on HD.

Thoughts
1) Oppo DVI out-using DVI reduced the static and noise in my video signal by a factor of at least 10.
2) Moome-simply transcodes to RGB but can't drive a strong signal
3) Extron-drives the signal up to the Barco to make sure its 700 mV (scoped and it is)
4) tse box-for gamma boost.
5) Getting the Barco amps (scoped) and G2's just right was also really important.

Thats what I've found out by messin' around. FWIW

yes! the old version dvi box can not drive too long cable. new box has solve this problem!
post #111 of 174
moome-you have a new external transcoder???
post #112 of 174
The Lumagen gives you great flexibility to increase/reduce the amount of gamma as well as correct any push your projector might have so that you get near perfect D6500 tracking.

I have had the G90 up and running now close to a month and am very impressed by the PQ and overall quality of the image. I still prefer the NEC XG series menu structure along with its near infinite point adjustments and much more logical memory inputs but the G90 provides a much brighter/sharper image IMO.

My XG-135LC is certainly an excellent projector and I don't think that I will ever get myself to sell it.

I recorded a Soundstage concert last week. Here is a screen shot using a Canon 20d downsized to 1024 X 768 res being projected on a 48x92" Wilsonart laminate "screen" using the G90.




Image from the opening scene of Star Wars. Inky black background.





Quote:
Originally Posted by picree View Post

Your description makes me smile I'll mention this to my neighbor for his G90. That may be why he was inclined to set the Lumagen back to flat line (the tse curve we replicated may have been too aggressive for his liking). He may be more inclined to just make the small tweak that you made.

Side comments on the tse box/gamma curve. I use a tse box on my BG808 for gamma boost. Some say it's too aggressive for a Barco. I also have an external moome transcoder. Some have said variously that the moome crushes black's or in other cases washes out the colors In any event I did notice that the gamma results one gets from a tse and/or moome depends upon how they are connected (interplay of all the little amps I suppose). For me the moome completely washed out the image when I directly connected the it into a tse box...horrid. So I switched things up. The best combo I have come up with so far (to get inky blacks and vivid colors) is:

In the stack:
Oppo DVD==>DVI 2 foot cable out
DVI in to Moome==> RGB 1.5 foot cable out
RGB in to Extron==> 25 foot RGBHV snake out

At the projector:
RGBHV in to tse box==> 1.5 foot RGBHV cable out
RGBHV in to Barco

Complicated and messy I know (purists probably hate me) but I think the results are really sharp with great color reproduction...inky blacks too. I've compared it to my neighbors G90 as a reference. True, his is more vivid, richer looking, razor sharp, with a jaw dropping gray scale! But I'm not too far off from that. Still impressive on HD.

Thoughts
1) Oppo DVI out-using DVI reduced the static and noise in my video signal by a factor of at least 10.
2) Moome-simply transcodes to RGB but can't drive a strong signal
3) Extron-drives the signal up to the Barco to make sure its 700 mV (scoped and it is)
4) tse box-for gamma boost.
5) Getting the Barco amps (scoped) and G2's just right was also really important.

Thats what I've found out by messin' around. FWIW
post #113 of 174
WOW Brian! Black is black! Nice contrast with her hair and flesh tones! Good punchy image! I'm trying to get to this point but need some suggestions.

I was helping my friend adjust the gamma on his G90/Lumagen setup last night. The problem: The Black window pattern on Avia comes up looking like almost 10 IRE. We can't get it to FTB. As a result his overall imaging just doesn't have the punch we know it should. He's using a Marantz DVD player that basically has no video controls what-so-ever. All corrections are being done in the Lumagen HDP.

Without an input the G90 has good G2 levels. The input on the Lumagen is set at 0 IRE (not 7.5) and the output is set to PC (according to the Lumagen rep PC=CRT and Video is for LCD's). PC is the darkest setting. The rep said we had to ISF the Lumagen/Marantz first with a reference monitor (not really an option).

To restate, the trouble is that, an all black scene (Avia) looks like 10 IRE! This is with a standard IRE gamma curve in the Lumagen (no correction...no boost).

Any suggestions? Is there any sort of input configuration settings on the G90 we should check?

Curt
post #114 of 174
Hello Curt,

From what you are saying it sounds like either the bias or gain settings on the G90 may not be correct. What is your brightness/contrast levels set to on the G90? What is the brightness and contrast settings on the Lumagen??

I would first set everything in the Lumagen back to the default values and make sure that the lum values equal the IRE numbers.

Regarding the Marantz dvd player you mentioned that there are no adjustments but often these players will have a setting to pass "blacker than black" or 0 IRE setting. I would make sure that this unit does not have such an adjustment.

Going back to the G90, you might want to go into the service menu and select "custom" as the color temp. You can then load in the values that I have on my 90 and see if this improves anything for you. Also, check the "gamma" settings on his unit but to do that you have to go into the "expert" menu.

This is one of the things I hate about the G90. Once you are in the service menu you should be able to access everything. To have to keep switching between "expert" and "service' is just plain foolish. At any rate that is what you have to do to access these controls. To get to the gamma settings, go to the appropriate menu and then hit the bias and gain controls at the same time. This will bring up the gamma window. I have posted my settings for this as well so it should give you a good base line in which to go by.

Let me know how it goes..



Quote:
Originally Posted by picree View Post

WOW Brian! Black is black! Nice contrast with her hair and flesh tones! Good punchy image! I'm trying to get to this point but need some suggestions.

I was helping my friend adjust the gamma on his G90/Lumagen setup last night. The problem: The Black window pattern on Avia comes up looking like almost 10 IRE. We can't get it to FTB. As a result his overall imaging just doesn't have the punch we know it should. He's using a Marantz DVD player that basically has no video controls what-so-ever. All corrections are being done in the Lumagen HDP.

Without an input the G90 has good G2 levels. The input on the Lumagen is set at 0 IRE (not 7.5) and the output is set to PC (according to the Lumagen rep PC=CRT and Video is for LCD's). PC is the darkest setting. The rep said we had to ISF the Lumagen/Marantz first with a reference monitor (not really an option).

To restate, the trouble is that, an all black scene (Avia) looks like 10 IRE! This is with a standard IRE gamma curve in the Lumagen (no correction...no boost).

Any suggestions? Is there any sort of input configuration settings on the G90 we should check?

Curt
post #115 of 174
Also, AVIA doesn't use standard levels - right? You should probably use Digital Video Essentials instead, or the built-in patterns of the Lumagen.
post #116 of 174
The problem with Avia is that it has a 1-bit chroma error causing a magenta push. At high IREs this is insignificant, but at low IREs it can make a pretty big difference.
post #117 of 174
Gary...arioch-please...we're talking about a very gross error in the setup somewhere; not minor quirks with Avia. The Avia scene in question is inky black on my Barco.

With no input the G90 looks like it has the proper G2 levels. Fine. But for any FTB scene (Avia, whatever) the best he can do is Fade To about 5 IRE. He spoke with a Sony G90 tech today who said that the bias settings need to be properly calibrated. We never really did this so perhaps this is where the problem lies.

How did we do it? We set bias similar to how we set it in a Barco...we never touched green bias (since a Barco doesn't have one we weren't sure how to approach it...AND we were afraid of shifting things all over the map, we wanted to maintain some standard reference point so we left green alone). We simply adjusted R & B bias around green to get the 30 IRE gray-scale right. According to the tech we need to drive all three bias settings down while looking at a 0 IRE input until we get FTB (analogous to the G2 setting in a Barco. We're taking a look at this now.

Anyone else done this? How do you adjust bias in a G90? R & B? or R & B & G?

Curt
post #118 of 174
The tech's advice seems sound to me. Display IRE 0 (with BRIGHTNESS at your desired setting) and lower the BIAS (probably all 3 roughly equally) until you get FTB. Then go back to 30 IRE and re-jigger R & B for D6500.
post #119 of 174
Thanks for the confirmation Gary. Didn't mean to jump you. It's frustrating. We're getting close and I didn't want to loose focus in the thread. True what you say about Avia though. I think we simply missed a step or two in the original setup.....we've punted back to the rough set up stage so Avia will hopefully do fine at this point.

Curt
post #120 of 174
Quote:
Originally Posted by picree View Post

Thanks for the confirmation Gary. Didn't mean to jump you. It's frustrating. We're getting close and I didn't want to loose focus in the thread. True what you say about Avia though. I think we simply missed a step or two in the original setup.....we've punted back to the rough set up stage so Avia will hopefully do fine at this point.

Curt

Hi Curt,

What are your G90 numbers for Bias red, green and blue before (and after too if you have made the adjustment) this adjustment (for both course and fine)? Also what are your gain numbers for red, blue and green?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: CRT Projectors
AVS › AVS Forum › Display Devices › CRT Projectors › Gamma Correction?