or Connect
AVS › AVS Forum › Display Devices › Plasma Flat Panel Displays › LG Plasma Panel tweaks for better blacks.
New Posts  All Forums:Forum Nav:

LG Plasma Panel tweaks for better blacks. - Page 11

post #301 of 1390
Quote:
Originally Posted by rob80b View Post

Hi austindub

Having started this thread I felt harvro01's approach was a good summation of what we've accomplished so far and a good place to start if someone really wants to jump in so I decided to quote it in the first post until we really nail down the technical procedure.
xrox has given us some good technical insight but I still feel we've got a bit of a way to go yet before I'm really comfortable with the procedure.

No doubt... I did feel like I was just pulling hand grenade pins and hoping they wouldn't go off. Not to mention thinking I would get shocked to death if I slipped off of the pots. I love that I would risk death for lower MLLs. Haha.
post #302 of 1390
Thread Starter 
Quote:
Originally Posted by austindub View Post

No doubt... I did feel like I was just pulling hand grenade pins and hoping they wouldn't go off. Not to mention thinking I would get shocked to death if I slipped off of the pots. I love that I would risk death for lower MLLs. Haha.

I'm comfortable around the electronics, although one should always be cautious. I was actually thinking around the terms of when I had my first analogue synth and being confronted with 30 or more knobs which one dialed in to get a specific tone or sound effect. Each dial affected the other, the procedure is somewhat relevant here and I think once we get a handle on the concept the procedure should be more or less straight forward.
My understanding so far is (but do not quote me yet): Step_up dictate the positive and negative excursion as to the horizontal axis + and -, Step_down dictates the width excursion as to the verticall axis of the Ramp wave form, VZ the duration for the trigger (reset) of the Ramp waveform, YZB controls the duration of the sustain period, VS the amplitude (strength) of the trigger for the reset. (obviously over simplified but if anyone is up to it they may want to put LG document https://data.epo.org/publication-ser...iDocId=4997940 into layman's terms.
"[0007] In such a sub-field driving method, the reset interval RP is set for each sub-field to initialize the discharge cells in the same state. Due to the reset interval RP, however, a spurious light-emission that does contribute to the brightness is generated at the rising and falling edges of the reset voltage pulse Pp every sub-field SF1 to SFn. A brightness of a black level rises from such a spurious emission to lower the contrast. In order to overcome this contrast deterioration, a scheme of including one reset interval per frame or a reset interval having a lower frequency than the prior art, that is, a full writing period FWP as shown in Fig. 4 has been disclosed in Japanese Laid-open Patent Gazette No. Pyung 5-313598."

Once I understand the exact relation this has on the panel discharge and how to control a stable black level without messing anything else up it should be a piece of cake, maybe.
There will always be slight deviations as all panels are not created equal but the approach would be similar, we don't what to dial in a thunder clap when a clarinet is called upon.
post #303 of 1390
Quote:
Originally Posted by rob80b View Post

translate document https://data.epo.org/publication-ser...iDocId=4997940 into layman's terms.
"[0007] In such a sub-field driving method, the reset interval RP is set for each sub-field to initialize the discharge cells in the same state. Due to the reset interval RP, however, a spurious light-emission that does contribute to the brightness is generated at the rising and falling edges of the reset voltage pulse Pp every sub-field SF1 to SFn. A brightness of a black level rises from such a spurious emission to lower the contrast. In order to overcome this contrast deterioration, a scheme of including one reset interval per frame or a reset interval having a lower frequency than the prior art, that is, a full writing period FWP as shown in Fig. 4 has been disclosed in Japanese Laid-open Patent Gazette No. Pyung 5-313598."

Translated that quote reads the reset period sets all cells into the same wall charge state but in doing so generates unwanted light emission that we see as black level. To reduce the black level we can only generate reset once or twice per frame instead of 10 times per frame.

As for FWP I would have to see figure 4. There are two types of addressing.

Write address - All cells in off state after reset and wall charges are selectively added to put a pattern of cells into the on state

Erase address - All cells in on state after reset and wall charges are selectively removed to put a pattern of cells into the "off" state
post #304 of 1390
Thread Starter 
Quote:
Originally Posted by xrox View Post

Translated that quote reads the reset period sets all cells into the same wall charge state but in doing so generates unwanted light emission that we see as black level. To reduce the black level we can only generate reset once or twice per frame instead of 10 times per frame.

As for FWP I would have to see figure 4. There are two types of addressing.

Write address - All cells in off state after reset and wall charges are selectively added to put a pattern of cells into the on state

Erase address - All cells in on state after reset and wall charges are selectively removed to put a pattern of cells into the "off" state

Thanks xrox

I wasn't even sure if I pulled the right paragraph to quote, all an interesting read though.
post #305 of 1390
Hello,
Is this tweak reducing the green dancing dots on the darker colors? or the black crush? or just making deeper black?
post #306 of 1390
Thread Starter 
Quote:
Originally Posted by *MaDJiK* View Post

Hello,
Is this tweak reducing the green dancing dots on the darker colors? or the black crush? or just making deeper black?

In some cases all of the above, on my PG25 I was able to eliminate the "green dancing dots" with a slight adjustment of the pots, on the other hand you could make it worse as they all interact with on another.
In your case it would be a matter of trial and error or you could try reducing the brightness, the green dancing pixels are inherent for all makes and dependent on the source.
Comparing some films I've seen it more pronounced on my brother's Panasonic ST30 than my PX950 for example.
If it's not apparent on everything I would probably leave it as is, but if it is a major problem you could try adjusting the pots.
post #307 of 1390
Thread Starter 
Which brings us to another topic.
One thing I've been curious about is the black levels displayed from different sources, dvd, Blu-ray (lowest black), Apple TV- Netflix, Cable SD, HD broadcast etc, if we take in the perceived black when viewing letter box bars for example or even the overall black I'm quite sure the MLL does vary.
Blu-Ray is usually quite stellar but some of the same films on DVD or Netflix can be mediocre including the black bars if any, SD cable curiously enough usually looks fine.
As xrox mentioned To reduce the black level we can only generate reset once or twice per frame instead of 10 times per frame I wonder if the different source signals some how dictates how the programming and electronics of the panel interact with the discharge and reset, just food for thought.
post #308 of 1390
Quote:
Originally Posted by rob80b View Post

Comparing some films I've seen it more pronounced on my brother's Panasonic ST30 than my PX950 for example.

Really? Panasonic have dancing dots too?
OK, thanks for your answer. You're right the green dots are not always disturbing.


I also notice some sources have different black levels, and even more: on some HDMI sources the TV setting "Black level: high or low" has the opposite effect.
I guess it depends on the settings and encoding of the source. I'm not an expert but I know there is for example sRGB 16-235 (the default on DVD if I remember well) and sRGB 0-255 (BluRay, still if I remember well) which render a more contrasted image with better blacks.
post #309 of 1390
all plasma has green dancing dots, its how plasma makes dark colors. LG seems more prone than most though. If you crank the brightness over 50, you'll see green dancing dots on the black bars also. But at 50 it crushes details, no amount of pots tweak would make the green dots go away at 59 brightness where its should be.
post #310 of 1390
Just noticed some strange pixelization on a football game tonight, especially during fast motion pans. This is the first time I have noticed any kind of errors after this mod. Not sure what I need to back off or change at this point. Also wondering if it is due to the source? It's coming off of an HD antenna. Regardless, the blacks are blacker and even if I can't correct it I think the tradeoff is worth it!
post #311 of 1390
Thread Starter 
Quote:
Originally Posted by austindub View Post

Just noticed some strange pixelization on a football game tonight, especially during fast motion pans. This is the first time I have noticed any kind of errors after this mod. Not sure what I need to back off or change at this point. Also wondering if it is due to the source? It's coming off of an HD antenna. Regardless, the blacks are blacker and even if I can't correct it I think the tradeoff is worth it!

Hi Austin

If your ok with a few pixel misfires and everything else appears hunky-dory I'd leave it until this tweaking evolves. I'm sure in time we'll eventually understand exactly what effects each dial and how each interacts with the the other.
Bear in mind though we've only have the pots that LG has implemented to work with the voltage controlled waveforms and to go further may be a firmware to control the microprocessors.
post #312 of 1390
Upon closer inspection, I have red "sparkles" in any white areas. I didn't notice this on a DVD (Finding Nemo) I used when originally making the changes. I am guessing I have to back off on the Set_Down poti?

I agree though, I might wait until there is a bit more research done instead of pulling the back panel off a hundred times to make little tweaks.
post #313 of 1390
I notice that after been tweaking the tv for a couple of weeks that the VZB POT is responsible of giving more punch to the colors, you will get more color pop with the VZB POT also will lower some mll.

post #314 of 1390
Quote:
Originally Posted by austindub View Post

Just noticed some strange pixelization on a football game tonight, especially during fast motion pans. This is the first time I have noticed any kind of errors after this mod. Not sure what I need to back off or change at this point. Also wondering if it is due to the source? It's coming off of an HD antenna. Regardless, the blacks are blacker and even if I can't correct it I think the tradeoff is worth it!

I had the same issue; so I cut the brightness from 53 to 52. It worked for my pixelation.
post #315 of 1390
Quote:
Originally Posted by losservatore View Post

I notice that after been tweaking the tv for a couple of weeks that the VZB POT is responsible of giving more punch to the colors, you will get more color pop with the VZB POT also will lower some mll.


Indeed. Thanks to Harvro, I re-opened back, just for that (VZB), and man was it worth it! Mine was only able to turn a hair clockwise and maxed out quickly. Same on yours? What Model?
post #316 of 1390
Quote:
Originally Posted by Turrican4D View Post

@derod68
Ever tried to set the Daynmac Contrast to "Off"?

Quote:
Originally Posted by rob80b View Post


...with contrast at 95 and Gamma high, both are pushing the black levels.



You and Turrican were so right. I thought dynamic contrast on low was a must but not so much needed now after tweaks. Turrican suggested I turn off and I did. Re-did Wizard, backed off some Contrast from 95 to 92, Gamma still high and kept the pop I waned. Now I can see lapels clearly on black blazers .
post #317 of 1390
Quote:
Originally Posted by derod68 View Post

Indeed. Thanks to Harvro, I re-opened back, just for that (VZB), and man was it worth it! Mine was only able to turn a hair clockwise and maxed out quickly. Same on yours? What Model?

I have plenty of room for the VZB, use a white screen when you use the VZB, go clockwise very slow and look in the background for jumping pixels just watch very close because the dancing pixels are hard to see not sure is those are sub pixels jumping...

MODEL:LG 60PK550
post #318 of 1390
Quote:
Originally Posted by losservatore View Post


I have plenty of room for the VZB, use a white screen when you use the VZB, go clockwise...

MODEL:LG 60PK550

Thanks losservatore. I have a 60PZ950 and only budged clockwise. Still did somethin though
Wonder if anyone has had more room for the VZB or which/how they turned on set I have with outcome? ?
post #319 of 1390
Thread Starter 
Quote:
Originally Posted by derod68 View Post

Thanks losservatore. I have a 60PZ950 and only budged clockwise. Still did somethin though
Wonder if anyone has had more room for the VZB or which/how they turned on set I have with outcome? ?

Just a hunch, but you may try reducing (or increasing) the VSC a touch, which "sets the amplitude of the complex waveform"
post #320 of 1390
This is helpful, though I will not touch my new TV just incase
post #321 of 1390
Quote:
Originally Posted by austindub View Post

just noticed some strange pixelization on a football game tonight, especially during fast motion pans. This is the first time i have noticed any kind of errors after this mod. Not sure what i need to back off or change at this point. Also wondering if it is due to the source? It's coming off of an hd antenna. Regardless, the blacks are blacker and even if i can't correct it i think the tradeoff is worth it!


To remove any pixel delay if the set_up is too clockwise you have to go counterclockwise little by little till you dont see the pixel delay. I recommend that you use a blue and black 1080p picture from the link below, put the pictures in to a usb flash drive and play the black first and quick change it to the blue switch between both colors really fast, use the set_up pot until the delay is eliminated.
andhttp://www.pbase.com/jackcnd/1080p_tests
post #322 of 1390
Hi,

thanks a lot for this post, it definitely helped me to bring new life to my PK250 (spanish name for the PK350, I think).

I had a 0,14cd/m2 black level before the mod, and now it is a much more convincing 0,04-0,05 cd/m2.

I followed more or less the instructions of havro, though I didn't notice any changes with the modifications of some of the pots. For my set, Set_Up is the main responsible of black level changes, in fact I couldn't tell any of the other pots made a diference in this regard. VA and Vy seem to give more room to Set_Up going clockwise.

I didn't notice that "pop" or dullness in the image produced by VZB, it didn't introduce any artifacts either, so I left it almost totally clockwise.

Set_dn is a pain in the ass, I moved it and the festival of pixel missfires began, I think i finally returned it to its original position.

I found very useful the AVS HD disc for doing these tests:

[I cannot post URLs, it seems, but I guess you know where to find it ]

the main screen with the "AVSHD" logo was perfect to see problems appear. In the lower left section (over a white background), Vy too much counterclockwise makes magenta pixels appear (set_dn also helps), and in the dark grey box that contains the submenus, red pixels begin to sparkle all over it (mainly in the bottom of the box) if you set Set_up too much clockwise. Also the grey field of the last submenu (the icon of the submenu) flickers if Set_up is too high.

The ISM method "color erasing" (don't know the exact name of the option in the english menus) also was very useful to detect pixel missfires.

Another problem that Set_up too much clockwise produces is "black bars bleeding", I mean, in a film with black bars, if a near to black image is next to the black bar, the absolute black of the bar partially "invades" the image. I noticed this in "The incredible Hulk", 1 exact hour of film I think, Norton's nape overlaps the upper black bar and this effect appears.

In a first tweak I got a 0.018cd/m2 measure, and that was amazing, but then I began to notice all these problems, finally I left it in a safer 0.045 cd/m2 level. Not so great, but much better than my original black level.

It is possible that with a better understanding of what each pot does I could get better results, but I don't really know what to do with Set_dn and VS, for example, or even VA ... clockwise or counterclockwise? (not much of a difference for me, really).

Anyway, great post, great contributions, and I will follow reading this to see if some "standard method" comes out!

Thanks all ! (and sorry for the english if it is a bit confuse or something)
post #323 of 1390
Quote:
Originally Posted by rob80b View Post

Which brings us to another topic.
One thing I’ve been curious about is the black levels displayed from different sources, dvd, Blu-ray (lowest black), Apple TV- Netflix, Cable SD, HD broadcast etc, if we take in the perceived black when viewing letter box bars for example or even the overall black I’m quite sure the MLL does vary.
Blu-Ray is usually quite stellar but some of the “same” films on DVD or Netflix can be mediocre including the black bars if any, SD cable curiously enough usually looks fine.
As xrox mentioned “To reduce the black level we can only generate reset once or twice per frame instead of 10 times per frame” I wonder if the different source signals some how dictates how the programming and electronics of the panel interact with the discharge and reset, just food for thought.

Just fyi, on my 60PK550 the MLL differs quite a bit between 60fps and 24fps inputs. I discussed this quite a bit in another thread several months ago. It's been a while since I last measured it but I recall that the 24fps input had a significantly higher MLL. This is probably characteristic of most of these displays I imagine. Based on this, I would assume that choice of input frame rate is an important thing to consider when tweaking.

The best explanation I heard was that the MLL level changes due to differences in how the signals are applied to the subfields for the different refresh rates (60 for 1808p60 vs 72 for 1080p24).
post #324 of 1390
Quote:
Originally Posted by rob80b View Post

Which brings us to another topic. I wonder if the different source signals some how dictates how the programming and electronics of the panel interact with the discharge and reset, just food for thought.

Depends on the Plasma. A Pioneer model should have a stable black level regardless of source. A Panasonic model will change the MLL depending on the APL of the signal. The higher the APL the higher the MLL. They do this by increasing the number of reset periods.
post #325 of 1390
Quote:
Originally Posted by rpauls View Post

The best explanation I heard was that the MLL level changes due to differences in how the signals are applied to the subfields for the different refresh rates (60 for 1808p60 vs 72 for 1080p24).

I would speculate that moving to 72Hz inherently increases the number of reset periods.
post #326 of 1390
Quote:
Originally Posted by losservatore View Post

To remove any pixel delay if the set_up is too clockwise you have to go counterclockwise little by little till you dont see the pixel delay. I recommend that you use a blue and black 1080p picture from the link below, put the pictures in to a usb flash drive and play the black first and quick change it to the blue switch between both colors really fast, use the set_up pot until the delay is eliminated.
andhttp://www.pbase.com/jackcnd/1080p_tests

That's what I did and it works very well (used blu ray player instead of flash drive) .

I think if you want to play it safe (till this process gets ironed out), then just tweek the Vy slightly counter clockwise then turn the SET_UP clockwise till you get pixel misfires then back off till there are none. Your blacks may not be as low as they can be set using the other pot combinations but it's still a world of difference as to what it was before. It has been 3 weeks since I did mine (PK550) and I have experienced no ill effects, just a great picture!
post #327 of 1390
i noticed that using VZB, you can no longer turn VY with out misfires.
post #328 of 1390
Thread Starter 
Quote:
Originally Posted by xrox View Post

I would speculate that moving to 72Hz inherently increases the number of reset periods.

Hi xrox

I believe you are right, if I understand it correctly going through LGs white paper, a higher res signal increases the "discharge sustaining interval" which also improves the contrast ratio which would explain why Netflix films vary so much as the streaming resolution will vary accordingly on the signal to eliminate wait times or interruption in the data.
post #329 of 1390
Quote:
Originally Posted by rob80b View Post

Hi xrox

I believe you are right, if I understand it correctly going through LGs white paper, a higher res signal increases the "discharge sustaining interval" which also improves the contrast ratio which would explain why Netflix films vary so much as the streaming resolution will vary accordingly on the signal to eliminate wait times or interruption in the data.

We are talking about refresh rate, not resolution. When the LG displays an input that is in 24p (true film mode) it shows each frame three times (24 * 3 = 72Hz). I believe this means each pixel is excited 72 times per second. By comparison, when viewing a 30 frame per second input (most DVDs, cablebox, netflix), each frame is displayed two times (30 * 2 = 60 Hz). In my experience, the MLL is brighter in 72Hz mode, which is unfortunate, since this is the more desirable mode for most movies and is the mode used by bluray.
post #330 of 1390
Thread Starter 
Quote:
Originally Posted by rpauls View Post

We are talking about refresh rate, not resolution. When the LG displays an input that is in 24p (true film mode) it shows each frame three times (24 * 3 = 72Hz). I believe this means each pixel is excited 72 times per second. By comparison, when viewing a 30 frame per second input (most DVDs, cablebox, netflix), each frame is displayed two times (30 * 2 = 60 Hz). In my experience, the MLL is brighter in 72Hz mode, which is unfortunate, since this is the more desirable mode for most movies and is the mode used by bluray.

Hi rpuals

Yes that I'm aware of that, by increasing the brightness it also makes black look darker in most regular lit scenes therefore increasing the contrast ratio but adversely makes dark scenes look grayer, unless we lower the MLL.

But my original inquiry was from reading LGs paper, it looks like resolution is also a contributing factor, not too sure how this applies to the up-scaling algorithms so my theory could be incorrect.
"[0008] In the PDP adopting the sub-field driving method, the brightness is determined by the display interval, that is, the discharge sustaining interval. Since a relatively long time is wasted due to the address interval allocated equally for each sub-field SF1 to SFn, however, a time allocated for the discharge sustaining interval determining the brightness lacks. For instance, when 480 lines are scanned by a scanning voltage pulse with a width of 3.m in the address interval of each sub-field, a time of about 1.44ms is required. Accordingly, since a time of about 12ms (i.e., 1.44ms×8) is allocated for the total address interval when 16.7ms is allocated for on frame display interval consisting of 8 sub-fields so as to display a 8-bit image data, a time of about 4ms is allocated for the discharge sustaining interval except for the reset interval. As a result, the conventional PDP has a problem in that the brightness is low due to a relative lack of the discharge sustaining interval determining the brightness. Furthermore, when it is intended to implement a screen with a high resolution, a discharge sustaining interval becomes more lack due to an increase in the address interval according to an increase in the scanning lines to make the display itself impossible."
https://data.epo.org/publication-ser...iDocId=4997940
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Plasma Flat Panel Displays
AVS › AVS Forum › Display Devices › Plasma Flat Panel Displays › LG Plasma Panel tweaks for better blacks.