THE SONY SERVICE CODES - Articles, Comments, Discoveries - Page 32 - AVS Forum
Forum Jump: 
 2Likes
Reply
 
Thread Tools
post #931 of 2973 Old 02-07-2006, 06:52 PM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,482
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 176
Apologies for jumping in the middle of a discussion, but just thought I'd throw in my 2c re some of the 2170P-3 edge controls, since I've been experimenting with these the last few days. This is still sort of a work in progress, so editing / retractions / updates (without notice) are likely, if not certain.

Some of the controls seem to operate with finer granularity depending on which SYSM setting is used, so I've broken them down into two categories based on that.

All of these controls seem to have their strongest effect in a horizontal direction and little or no effect in a vertical direction, so their behavior is usually easiest to see on vertical edges. And when I refer to "vertical edges" or "horizontal detail" below, I'm really talking about the same thing.

All of these observations were made using patterns/video content at 480p (input at 33.75khz) via DVI on a 34XBR800.

NOT (OR DOESN'T APPEAR TO BE) SYSM-SENSITIVE
  • SHF0 seems to change the "frequency" of sharpening in the image, creating a subtle "loosening" or "tightening" effect on horizontal details. 0=loose. 1=tight. It's easiest to see the effect with Sharpness raised on a pattern like this with SYSM=1. Though you can also sense the tightening/loosening effect in video content at nominal Sharpness as well. I don't think this control "scales" with SYSM, but it does effect the character of edge-distortions (ie ringing/overshoot) with each SYSM setting.
  • PROV is also easiest to initially see with Sharpness elevated. And seems to shift the "emphasis" to either the left or right of the edge in small increments. Like SHF0, the effect of this control is pretty subtle at nominal Sharpness settings, but it can be perceived as sort of pulling your eye towards either the left or right side of the screen. So it may have some value in achieving better "equilibrium" in the picture.

    PROV may need to be tweaked differently depending on how SYSM and other controls are configured. Although it's easier to see the behavior of this control at higher Sharpness settings, I seem to get more reliable results adjusting it at nominal Sharpness, based on the way it seems to pull my eye toward either the left or right of the screen.
  • LTLV functions differently depending on how LTMD is configured.

    If LTMD=0, then LTLV seems to smooth out details that are similar in brightness/color, while sharpening those that are aren't. So you may be able to think of it as sort of a combination low-con noise-reducer/hi-con edge-enhancer. It particularly seems to reduce "noise" (and maybe also some clarity) in darker areas of the image, ie shadow detail.

    If LTMD=1, then LTLV seems to have a more generalized sharpening effect, basically reducing softness in the picture across the board, including shadow detail. This could be useful for enhancing shadow detail on 480p sources with SYSM=1. Whether it would be beneficial with the other SYSM settings, I don't really know yet. Preliminarily it looks like it might perhaps be a little heavy-handed with SYSM=2.

    To see any effect from the two LTMD settings, LTLV has to be higher than 0 (ie between 1 and 3). LTLV=3 has the strongest effect. And the behavior of these controls is probably seen most easily at nominal Sharpness probably with SYSM=1. You can probably get a better feel for the difference between the two LTMD settings if you set LTLV to 3, and then flip back and forth between LTMD=0 and LTMD=1 while watching something with a pretty good level of detail like AOTC.

SYSM-SENSITIVE
  • F1LV seems to better define vertical edges, by boosting brightness on the lighter side(s) of the edge. Its effect can be seen fairly easily on patterns of fine dark detail or text against a lighter background, with Sharpness at nominal settings. This control is sensitive to how SYSM is set (ie it seems to operate at higher frequency/smaller granularity with SYSM=2/3), and it's easiest to see it's effect with SYSM=1.

    Although this does increase "noise" in the picture (especially at SYSM=1), it may have some "defining" value in terms enhancing shadow details, especially at the higher granulity with SYSM=2/3.
  • CTLV seems to control the saturation of color on fine horizontal detail where two different colors meet. You can see the effect fairly well on the attached CTLV patterns with SYSM=1 at nominal Sharpness. With SYSM=2/3, CTLV seems to operate at a higher frequency/smaller granularity (watch the thinnest lines in the CTLV1.JPG pattern with SYSM=2/3).

LL
LL

ADU
ADU is offline  
Sponsored Links
Advertisement
 
post #932 of 2973 Old 02-07-2006, 08:13 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by ADU View Post

CTLV seems to control the saturation of color on fine horizontal detail where two different colors meet.

I very much appreciate your additional insights on these "enhancement" parameters. Some of them are very hard to describe, and it's good to hear another view.

I'll bet that these parameters similarly named but starting with "L" or "C" are the same filters acting on Luminance and Color information, respectively. Whatever Sony's mysterious "LTI" is, there seems to be an equivalent "CTI," controlled by LTLV and CTLV. When I experimented with them, I didn't like what they did to light/dark boundaries, and I did notice a slight softening of fine textures. But I never thought to check for shadow vs highlight effects, as you described. I will now have a look, too -- although I am skeptical that I will ever find a use for them.

In the MID5 parameters, there are groups of filters that enhance color sharpness (MHCR, MHCE, MHCO) just as the ones for luminance do (MHYR, MHYE, MHYO) -- except here the luminance is referenced by a "Y." There is a downside to sharpening color, as color boundaries, especially between certain color pairs, are already screwed up by the color encoding during compression or by the NTSC broadcast standard. I once noticed that enhancing sharpness for color alone was not increasing the attractiveness of the picture but drawing attention to these artifacts. So I have chosen to leave color alone for now, as evidenced by all of the zeroes in the MID5 color parameters that I have recommended. The eye is known to be remarkably uncritical of color resolution, so maybe it's a lost cause.

Live and learn, though. Maybe there is a role for these color-signal sharpening features. I just haven't found it yet.

BTW, where did you get those lovely color patterns? Are there more?

KenTech
"We all get smart slowly."
KenTech is offline  
post #933 of 2973 Old 02-07-2006, 08:24 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by loadams View Post

I've got the older, bigger brother 40" gorilla. Now with 9 of 10 dvd rentals being in 16x9, why would I want to continue using 480p (component) when the set upconverts those dvds to 960i anyway? Pros.....cons ?

What I can't predict for you is how your set handles these different video standards visually. On my 36XS955, for example, the set is more competent dealing with a 480i input in producing a marvelous picture, whether interlaced 960 or the CineMotion equivalent. If I feed 480p via component inputs, there is a limit on the detail somewhere -- the picture has lost some important texture information, and it can't be recovered. So I never use it. (HDMI is better, but that's another discussion.) Admittedly I can't tell if this is a shortcoming in my Panasonic S97 player's Faroudja/Genesis chip or a characteristic of the TV. I just know there's nothing superior to the component inputs for 480i, and so I would sacrifice 480p in a heartbeat!

So, that said, I would dump 480p, assuming you're happy with how it handles the 480i. Then you can twiddle your 2151 chip's settings to make 1080i similar in color rendition.

KenTech
"We all get smart slowly."
KenTech is offline  
post #934 of 2973 Old 02-07-2006, 08:35 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by SurfingMatt27 View Post

You want a sharp picture, turn all the MIDE5 settings to 0, your PQ will be ultra sharp with no enhancements neccessary.Just look at text and tell me how perfect it is.

This certainly should be, IMO, one of the columns available for your MID5 table. On mine, it's #63. But to say all-zeroes is all you need in all cases is ignoring what MID5 can do for you in many cases. All of the settings from #5/MHYR thru #18/MVCE are sharpness *enhancers* that have no visible downside. That is, they don't exaggerate some detail while squashing finer detail. The use of MHYR=3, MHYO=0, and then choosing a MHYE from 1-7 really restores fine detail and texture nearly lost in the video-processing chain. Let your eyes be your guide! Changing MYHO to 1 coarsens the effect, which is useful for SD video. Do what looks right to your eyes. There are no "best" settings!

KenTech
"We all get smart slowly."
KenTech is offline  
post #935 of 2973 Old 02-07-2006, 08:52 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by justsc View Post

'm just thinking there may well be more "sharpness" to be enjoyed by "bringing it out" rather than "adding it in."

So, I'm focusing on finding ways to "expose" levels of fidelity that are already there but I haven't seen - yet.

Well, this is the philosophy behind most of the "enhancements" discussed here. What Matt said, that you can't recover what isn't there, is true, indeed, and oversharpening is ugly! But sometimes detail and texture are really there and attenuated, like rolled-off highs in a sound system. A little boost restores the fidelity -- as long as you can tolerate a bit more noise, because you'll be boosting that, too.

The second reason to add sharpness to a soft-ish picture is to fool the eye when the picture is viewed at an appropriate distance -- the basic technique with broadcast SD, such as analog cable. Get too close, and you can see what crap the picture is; but get back a ways, and the picture integrates in your eye/mind and doesn't look so bad. If you add a little sharpening to the mix, that illusion can be improved, even if there is some outlining when viewed close. (Solution: Don't view close!)

As is obvious from reading Matt, viewing preferences vary a lot from person to person, and there is no "right" choice for everyone on what makes a great picture. Further, many of these judgements can't be made in a snap -- one has to live with a setup for a while to see if it's good.

KenTech
"We all get smart slowly."
KenTech is offline  
post #936 of 2973 Old 02-07-2006, 09:23 PM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,482
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 176
Quote:
Originally Posted by KenTech View Post

I very much appreciate your additional insights on these "enhancement" parameters. Some of them are very hard to describe, and it's good to hear another view.

Agreed. See some of my already add'l edits above re the "tightening" effect of SHF0, and the "pulling" effect of PROV.
Quote:


There is a downside to sharpening color, as color boundaries, especially between certain color pairs, are already screwed up by the color encoding during compression or by the NTSC broadcast standard. I once noticed that enhancing sharpness for color alone was not increasing the attractiveness of the picture but drawing attention to these artifacts. So I have chosen to leave color alone for now, as evidenced by all of the zeroes in the MID5 color parameters that I have recommended. The eye is known to be remarkably uncritical of color resolution, so maybe it's a lost cause.

Live and learn, though. Maybe there is a role for these color-signal sharpening features. I just haven't found it yet.

Raising CTLV seems to reduce the color saturation at boundaries, softening, rather than sharpening color detail. So it might perhaps be useful for reducing certain color artifacts, or taking some of the bite off high-contrast color edges in some instances.
Quote:


BTW, where did you get those lovely color patterns?

Made 'em myself in Photoshop!

ADU
ADU is offline  
post #937 of 2973 Old 02-07-2006, 11:15 PM
AVS Special Member
 
SurfingMatt27's Avatar
 
Join Date: Dec 2005
Location: Massachusetts
Posts: 2,789
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 12
Quote:
Originally Posted by KenTech View Post

Well, this is the philosophy behind most of the "enhancements" discussed here. What Matt said, that you can't recover what isn't there, is true, indeed, and oversharpening is ugly! But sometimes detail and texture are really there and attenuated, like rolled-off highs in a sound system. A little boost restores the fidelity -- as long as you can tolerate a bit more noise, because you'll be boosting that, too.

The second reason to add sharpness to a soft-ish picture is to fool the eye when the picture is viewed at an appropriate distance -- the basic technique with broadcast SD, such as analog cable. Get too close, and you can see what crap the picture is; but get back a ways, and the picture integrates in your eye/mind and doesn't look so bad. If you add a little sharpening to the mix, that illusion can be improved, even if there is some outlining when viewed close. (Solution: Don't view close!)

As is obvious from reading Matt, viewing preferences vary a lot from person to person, and there is no "right" choice for everyone on what makes a great picture. Further, many of these judgements can't be made in a snap -- one has to live with a setup for a while to see if it's good.

Exactly Ken

I feel your pain..it seems me and you are in the same boat always trying settings and seeing if you can live with them and test it with real material then constantly im changing parameters left and right.

Well after all this tinkering i realize you are never going to get a perfect picture but at least it's watchable,Even ISF is'nt guaranteed perfect given the limits of CRT technology.

I'm in the market for a replacement anyways, not that i don't like my sony it's just that me and my brothers are looking for a bigger display for gaming and movies and the closest we have found to match crt in PQ is DLP, so that's what i'm going to be looking into, especially the new 1080p OLED DLP's coming soon or the HP DLP tv.

peace,

Matt~
SurfingMatt27 is offline  
post #938 of 2973 Old 02-08-2006, 07:41 AM
AVS Addicted Member
 
Q of BanditZ's Avatar
 
Join Date: Dec 2003
Location: blurayoasis.com
Posts: 15,242
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by SurfingMatt27 View Post

Exactly Ken

I feel your pain..it seems me and you are in the same boat always trying settings and seeing if you can live with them and test it with real material then constantly im changing parameters left and right.

Well after all this tinkering i realize you are never going to get a perfect picture but at least it's watchable,Even ISF is'nt guaranteed perfect given the limits of CRT technology.

I'm in the market for a replacement anyways, not that i don't like my sony it's just that me and my brothers are looking for a bigger display for gaming and movies and the closest we have found to match crt in PQ is DLP, so that's what i'm going to be looking into, especially the new 1080p OLED DLP's coming soon or the HP DLP tv.

peace,

Matt~


You got a spare kidney you wanna part ways with?

Great ISF Job by Chad B.
Q of BanditZ is offline  
post #939 of 2973 Old 02-08-2006, 08:48 AM
AVS Special Member
 
justsc's Avatar
 
Join Date: Jun 2005
Location: San Jose, CA
Posts: 2,689
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by SurfingMatt27 View Post

...I'm in the market for a replacement anyways...

...the closest we have found to match crt in PQ is DLP, so that's what i'm going to be looking into, especially the new 1080p OLED DLP's coming soon or the HP DLP tv.

peace,

Matt~

Matt!

A 1080p OLED? Are you sure that's what you really mean? The largest yet shown was about 20" and they couldn't even estimate a price because there's still many bugs to work out. OLED and DLP are not technologies used together, it's either - or.

We are researching FOLED display technology for the Heads Down Displays in our simulation cockpits and the prices are astounding. For a 13" display you're looking at about $40K, and that's not 1080p.

We're probably about 2-3 years out from SED at a decent price point, and about 5-10 years out for OLED at any kind of size and decent price point.

I recommend you look at HP's 1080p DLP - it's very, very nice.

Cheers!
justsc is offline  
post #940 of 2973 Old 02-08-2006, 08:49 AM
AVS Special Member
 
SurfingMatt27's Avatar
 
Join Date: Dec 2005
Location: Massachusetts
Posts: 2,789
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 12
LOL yeah, my left nut

But seriously though, i'm not going to make thepurchase to about late march oe april so it gives me plenty of time to save up.The sony's been great these past 2 years, i just wanted something bigger.I was planning on buying a bigger tv in the first place but did'nt have enough cash at the time when i purchased my sony.

Any thoughts on specific DLP models i may be interested in please feel free to PM me.Or better yet i'll just make a topic thread in the RP display thread.

That way we can get back on topic of Ken's thread.

Matt~
SurfingMatt27 is offline  
post #941 of 2973 Old 02-08-2006, 09:00 AM
AVS Special Member
 
SurfingMatt27's Avatar
 
Join Date: Dec 2005
Location: Massachusetts
Posts: 2,789
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 12
Quote:
Originally Posted by justsc View Post

Matt!

A 1080p OLED? Are you sure that's what you really mean? The largest yet shown was about 20" and they couldn't even estimate a price because there's still many bugs to work out. OLED and DLP are not technologies used together, it's either - or.

We are researching FOLED display technology for the Heads Down Displays in our simulation cockpits and the prices are astounding. For a 13" display you're looking at about $40K, and that's not 1080p.

We're probably about 2-3 years out from SED at a decent price point, and about 5-10 years out for OLED at any kind of size and decent price point.

I recommend you look at HP's 1080p DLP - it's very, very nice.

Cheers!

Thanks for the advice and yeash your probably right, maybe i meant to say LED DLP are'nt these supposed to be coming out soon later in april?Oh and thanks for the suggestion of the HP i've been hearing a lot of good things about it and will look into that as well as any other advice would be greatly appreciated.

Well i'll be making another thread now in the RP display forum, feel free to help me out anytime since i'm new to DLP and would like some good suggestions.

I'll be selling my sony too by the way for cheap, anyone interested please PM me, since i really have no need for it anymore once i get my new replacement soon.My father wants me to get rid of it any ways since theres really no room to accomodate it now because of the bulk and weight and we have enough tv's as it is in our house about 8 by the way ....So i can't really keep it.

Peace,

Matt~
SurfingMatt27 is offline  
post #942 of 2973 Old 02-08-2006, 09:39 AM
AVS Special Member
 
justsc's Avatar
 
Join Date: Jun 2005
Location: San Jose, CA
Posts: 2,689
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by SurfingMatt27 View Post

...Oh and thanks for the suggestion of the HP i've been hearing a lot of good things about it and will look into that as well as any other advice would be greatly appreciated.

...i'm new to DLP and would like some good suggestions...

Matt,

The newest 1080p sets have alot to offer. I recently had the pleasure of calibrating a Sharp Aquos 45" 1080p LCD set - what a fantastic picture! In my search for that set I looked very closely at Sony's 50" 1080p SXRD set - awesome as well.

HP is getting excellent reviews of it's two new 1080p DLP sets, and they stand apart as some of the only ones that can actually accept as well as display 1080p.

Here's a piece from CNET:

The MD6580n also offers extremely impressive all-around performance, and although it didn't score quite as well during testing as Sony's SXRD-based KDS-R60XBR1, it came close enough to make it our favorite DLP-based HDTV yet.

You can also find good reviews here:

http://www.ultimateavmag.com/rearprojectiontvs/106hp/

http://www.hdblog.net/2006/01/23/rev...n-ultimate-av/

Best of luck in your search!
justsc is offline  
post #943 of 2973 Old 02-08-2006, 10:18 AM
AVS Special Member
 
SurfingMatt27's Avatar
 
Join Date: Dec 2005
Location: Massachusetts
Posts: 2,789
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 12
Thanks steve

Anybody you know looking for a sony 34" for cheap?If anybody's intertested in my sony 34hs420 just e-mail me or PM me.

By the way is'nt there a forum section for selling stuff?

Matt~
SurfingMatt27 is offline  
post #944 of 2973 Old 02-08-2006, 12:06 PM
AVS Special Member
 
Nitewatchman's Avatar
 
Join Date: Dec 2001
Location: Middletown, Ohio
Posts: 6,292
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 18
It might be interesting to hear from someone who has worked with the relevant 2170P-3 and MID5 settings on both a HS420 and one of the SFP sets.

Perhaps we are mostly dealing with issues relating to what any given person "likes", and as Ken said -- there is no "right" choice of settings for everyone ... Perhaps such issues as differences in output from DVD players also are involved.

Nevertheless, I wonder if it is possible we might also be getting something a little different on this depending upon if we are using SFP tube or no, or maybe there are some other differences between the different models which may be involved. Also -- apologies if this has been covered -- when, for instance we are talking about, say 480p or 480i from DVD via V5/V6 component inputs and SYSM=3 and MID5 column's at all Zero's -- are we also using the same, other relevant 2170P-3 settings? I think ADU's excellent info on these demonstrates how slight differences in how they are set up can be important as well.

-----------------------------

Anyway - FWIW - With ADU's excellent info on the relevant P2170-3 settings printed out and brought along -- Played around with these in a "seat of the pants" manner a little last night With KD34XBR960 and 480p DVD(Zenith DVD2381) via component input. Difficult to say how much of this, if any is "specific" to my DVD player.

I was working with these in my "experimental" pic mode(just happens to be "vivid"), to easily make comparisons to the settings I normally use, which are set up for "Standard" pic mode. Along with test patterns from AVIA, "The Phantom Menace" happened to be the first DVD I grabbed - and although not the best(It seems to have quite a bit of EE "built in" in spots for instance), various scenes from it, as well as its "THX Optimizer" along with Text+graphics in my DVD Players "setup" menu is what I ended up looking at.

Started with a MID5 "all zeros" column and SYSM=1 to more easily see what was going on. Then tried SYSM=2 or 3, making slight adjustments as necessary. I also played with changing MID5 columns set up with different settings along with making small changes to the "vivid" mode settings and trying different MID5 settings/etc.

Again, Just doing "seat of pants" stuff here - FWIW (not taking the time as is required to investigate in detail, or spending a lot of time "living with" small changes) - and just given a couple of hours playing with it :

#1) I thought it was interesting what one can do w/o use of MID5 column when one knows a little more of what is going on with these 2170P-3 settings.

#2). I do think I was probably able to come across the best results I've managed to date for 480p DVD with SYSM=3, utilizing the following applicable 2170P-3 settings - In brackets are my current settings I use with "standard" and "movie" pic modes (Note I have customized my Pic modes per KenTech's "customizing Picture modes" article earlier in thread - update : also note I included the VM "
shaping" settings, as well as I was using "VM" "medium" in user menu, with P2170-3 #19 VMM = 6 (set default was 8) :

#0 SYSM - 3 [2]
#2 VMCR - 0 [0]
#3 VMLM - 3 [3]
#4 VMFO - 2 [2]
#5 VMDL - 8 [8]
#6 SHOF - 0 [0]
#7 SHFO - 1 [1]
#8 PROV - 3 [3] (Note : I first ended up with PROV=2 for SYSM=1)
#9 F1LV - 1 [0]
#10 LTLV - 2 [0]
#11 LTMD - 1 [1]
#12 CTLV - 1 [0]
#16 MIDE - 60 [60, 48 or 49 - haven't decided yet]

And following MID5 column (Settings not noted are all set at "zero")

#1 POP - 60
#6 MHYL - 3
#7 MHYE - 3
#10 MHCL - 3
#11 MHCE - 3

For reference, here's My MID5 Column #48 (again - Settings not noted are all set at "zero")

#0 POP - 48
#6 MHYL - 3
#7 MHYE - 3
#10 MHCL - 3
#11 MHCE - 3
#14 MVYL - 3
#15 MVYE - 1

And MID Column # 49 (Settings not noted are all set at "zero"):

#0 POP - 49
#6 MHYL - 3
#7 MHYE - 3
#10 MHCL - 3
#11 MHCE - 3
#14 MVYL - 3


With previous experiments (which didn't work out very well) trying to use SYSM=3 for 480p DVD --- which mostly involved emphasis on "tweaking" MID5 column settings with(except for SYSM=3) the same 2170P-3 settings as shown in brackets above -- I was never able to bring out the "details" without adding too much EE or "ringing".

Last night, with SYSM=3, I found If I increased MHYE slightly to 4 or 5 for the experimental "vivid" mode settings noted above, I was just starting to approach some slight, noticable "ringing" occuring as evidenced in text in the DVD player's "setup" menu, and MHYE=2 may have been best.

I think it was (with LTMD=1) LTLV at a higher value which "helped" the most with this, concerning an overall increase in "sharpness" using SYSM=3. However, although I can see where it would be beneficial in some cases, one issue I had with raising LTLV is that I didn't really *need* to increase "shadow detail".

#3). Using SYSM=2 instead of 3, I noticed I was also able to get somewhat "close" to what I'm getting from my current "standard mode" settings for 480p DVD. Except with the use of "all zero's" in the MID5 Column. "Ringing" didn't seem to be an issue(which is especially noticable when it occurs in text for my DVD player's "setup" menu), nor was visable EE being an issue. -- but, I couldn't get "close enough" when it comes to "bringing out" the details.

#4). Using SYSM=2 and bringing the MID5 settings back into the equation -- I still couldn't seem to find anything better than my current "standard" mode settings, although again, this is "seat of the pants" stuff -- I tried to "back off" on MHYE/MHCE/etc, and also tried to "back off" several of the 2170P-3 settings while adjusting MID5 settings as necessary to try to find a better "compromise". I couldn't find one, even if the only MID5 settings that weren't "0" were MHYL=3, MHYE=1 -- For example, that resulted in "ringing" in the text on the DVD player setup menu as compared to my "standard" mode settings. I did however end up spending quite a bit of time looking closely at a CTLV setting of "1", instead of "0", and, if anything, it looks like it might be possible a different CTLV setting may be something I might want to look at more.

#5). While I didn't spend much time with test patterns for anything of importance(and can only use MS with the patterns ADU posted, as I don't have a DVD recorder), perhaps of some interest - some of the most apparent "changes" I noticed using various test patterns involved the "sharpness" bars in THX optimizer when changing values for "SHFO, F1LV or LTMD/LTLV", and the boundries/edges between color bars (such as from AVIA) and adjusting "CTLV". The effects of increasing LTLV with LTMD=1 were also noticable on THX optimizer "black level"("brightness") adjustment screen.

Update: Oh - for all above experiments, I left user menu Sharpness slider at 31 - 30~31 is also what I have it for normal viewing with "standard or movie" pic modes.

Jeff
Nitewatchman is offline  
post #945 of 2973 Old 02-08-2006, 12:37 PM
Member
 
loadams's Avatar
 
Join Date: Apr 2003
Location: Mount Pleasant,SC
Posts: 158
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Thanks Ken for your reply, will probably go in that direction.

And yes, after watching 480p for a number of years, watching 480i w/ tweaks is very pleasing !!! Great stuff as always guys.
loadams is offline  
post #946 of 2973 Old 02-08-2006, 02:07 PM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,482
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 176
A couple more thoughts on the 2170P-3 edge controls...

SYSM
So far SYSM=3 seems to be giving a bit of a biased look to edges on my TV that's a bit uncomfortable to my eyes. It's possible that tinkering with other controls may be able to make this a bit more manageable, but so far I haven't spent much time on that, and have been focusing more attention on SYSM=1 and SYSM=2. Edge distortion is lowest with SYSM=1 at nominal Sharpness on my TV. SYSM=2 seems a little "noisier" looking, but I can see alot of potential in the higher granularity and generally better definition it seems to confer especially to lighter details and shadow information in the picture versus SYSM=1.

Per something Napoleon D said in another thread, rather than simply randomly flipping back and forth between the different SYSM settings while working with the other controls, I think there's probably something to be said for focusing attention on one SYSM setting at time (once you've got a pretty good grasp of the effect each 2170P-3 edge control has). And trying to break down the best configuration of edge controls to go with that particular SYSM setting. Then move on and repeat the process with the next SYSM setting, working in sort of hierachical fashion to narrow down the best configurations or range of potential adjustments for each (with a given input/signal).

However, I would not make the mistake of assuming that just because a particular 2170P-3 parameter doesn't "help" with one SYSM setting, that that control should necessarily be thrown out of the mix for the other SYSM configurations. Because as mentioned above some controls seem to work with finer granularity and more subtle effect with SYSM=2/3 than they do with SYSM=1. I'm thinking particularly of F1LV here, which seems to have a much finer and potentially more beneficial "defining" effect with SYSM=2/3 than with SYSM=1.

SHF0
In addition to a loosening (SHF0=0) and tightening (SHF0=1) effect, I think I'd also describe SHF0 as having sort of a deepening (SHF0=0) versus shallowing (SHF0=1) effect on the picture. The higher "frequency" of the detail with SHF0=1 seems to "flatten" the picture a bit. Although one generally thinks of depth as being a good thing, I wouldn't necessarily describe the shallowing effect of SHF0=1 as "bad" on my TV. Although it seems to add a bit more noise/grain/texture, whatever you want to call it, the shallow picture may perhaps be a little easier on my eyes in some ways. The greater depth, clarity, looseness of SHF0=0 has some positives as well though. So far, it's been kind of a tough call. Another way I'd describe the difference is SHF0=0 seems more "TV-like", and SHF0=1 seems more "monitor-like", if that means anything.

Since there only two choices here, one might want to try to work out separate configurations for each, and then decide which is preferred.

PROV
No noteworthy additions on this yet.

F1LV
Although this generates fairly coarse looking "noise" with SYSM=1, as mentioned above it seems to have some greater potential with SYSM=2/3, and may have some value in elevating shadow information so it's a bit less strenuous on the eyes.

LTMD=0/LTLV=1-3
Still tinkering with this. Haven't found much use for it yet, but might have some potential in small amounts with certain configurations. And might be useful in cleaning up especially noisy signals. The effect LTMD=0 has at high LTLV settings reminds me alot of the "Vivid" mode on Panasonic plasmas, the way it smooths details similar in color/intensity, and over-defines high-contrast edges.

LTMD=1/LTLV=1-3
Although this seemed to have potential for better-defining shadow detail with SYSM=1, my enthusiasm for it has diminished a bit after watching it for awhile. The effect of this control is pretty heavy-handed (especially with the greater "granularity" that's inherent with SYSM=2/3), and after playing a bit more with some of the other controls it seems as if there may be other ways of working out shadow issues with possibly a bit more finesse. If more "definition" or "edge" is all you seek though, look no further than LTMD=1.

CTLV
0=highest color saturation at edges. 3=the least, and generally softest-looking color detail. Seems fairly straight forward.

ADU
ADU is offline  
post #947 of 2973 Old 02-08-2006, 02:31 PM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,482
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 176
Quote:
Originally Posted by Nitewatchman View Post

#1) I thought it was interesting what one can do w/o use of MID5 column when one knows a little more of what is going on with these 2170P-3 settings.

Good point Jeff. Since the signal I use bypasses the digital MID controls, the 2170P-3 analog controls are basically the only ones I have for tweaking edge detail on my TV (aside from Clear Edge VM, which I've mostly ignored so far, and basic Sharpness.)
Quote:


Nevertheless, I wonder if it is possible we might also be getting something a little different on this depending upon if we are using SFP tube or no, or maybe there are some other differences between the different models which may be involved.

Quite possibly, if not probably. Using the MID controls & VM could certainly put a different spin on what settings work best as well.

Even if the "best settings" for these edge controls follows no particular pattern from one TV/user to another (or even one input/signal to another), I think you may be able to get a better feeling for the behavior of these controls and possibly where they might benefit you by playing around with them.
Quote:


I think it was (with LTMD=1) LTLV at a higher value which "helped" the most with this, concerning an overall increase in "sharpness" using SYSM=3. However, although I can see where it would be beneficial in some cases, one issue I had with raising LTLV is that I didn't really *need* to increase "shadow detail".

Another good point. See some of my related and somewhat similar impressions above.
Quote:


for all above experiments, I left user menu Sharpness slider at 31 - 30~31 is also what I have it for normal viewing with "standard or movie" pic modes.

This is similar to what I refer to as "nominal" sharpness in my earlier post. Nominal sharpness to me is the setting which produces the least edge distortion, but without introducing unwanted "softening" effects either. On my TV it's around (but a few ticks below) the middle of the slider (and I've reconfigured 2170P-4/USHP to make that setting my Pro mode default.) Lately I've been using this pattern to set that adjustment, looking, as I say, for minimum sharpening or softening distortions on the vertical lines.

Since there are other Sharpness offsets in the SM (such as 2170P-3/SHOF), the adjustment of this setting could vary a bit for different inputs/signals.

Tks for the kudos btw. Just trying to expand a bit more on Ken's already detailed info on these controls earlier in the thread.

ADU
ADU is offline  
post #948 of 2973 Old 02-08-2006, 03:47 PM
AVS Special Member
 
Nitewatchman's Avatar
 
Join Date: Dec 2001
Location: Middletown, Ohio
Posts: 6,292
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 18
Quote:
Originally Posted by KenTech View Post

If I feed 480p via component inputs, there is a limit on the detail somewhere -- the picture has lost some important texture information, and it can't be recovered.

Very interesting. Using component input -- While I do find I need to do "more" with the image processing settings for 480p DVD that is the case anywhere else -- I haven't really noticed a limit on detail. Maybe I should try to look into this in more detail.

Problem is, I've never been happy with the 480i from this particular player(Zenith DVD2381), including when using it with other sets, but I've allways been very happy with 480p from it.

Oh BTW Ken -- I'd responded to your PM reply and also had sent you another PM yesterday/etc with a few updates on various matters. But, the way the email notifications seem to "work"(or not work), sometimes(or maybe I managed not to click on the read recepit "OK" button for your message), I'm not sure whether you've seen those yet or not, if not hopefully they are in your PM inbox.

Jeff
Nitewatchman is offline  
post #949 of 2973 Old 02-08-2006, 03:52 PM
AVS Special Member
 
Nitewatchman's Avatar
 
Join Date: Dec 2001
Location: Middletown, Ohio
Posts: 6,292
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 18
Quote:
Originally Posted by ADU View Post

So so far, it's been kin dof tough call. Another way I'd describe the difference is SHF0=0 seems more "TV-like", and SHF0=1 seems more "monitor-like", if that means anything.

What excellent descriptions! I have never been able to put into words my thoughts about SHF0, now I have some words I can use. There was a period over about 3 months that I changed nothing else except SHF0 for various inputs. For those 3 months, I was using SHFO=0, but finally decided I liked that "flat", "tightened" response and "monitor-like" aspect of SHF0=1 better, and I've been using it ever since.


Quote:
Originally Posted by ADU View Post

Per something Napoleon D said in another thread, rather than simply randomly flipping back and forth between the different SYSM settings while working with the other controls, I think there's probably something to be said for focusing attention on one SYSM setting at time (once you got pretty good grasp of the effect each 2170P-3 edge control has). And trying to break down the best configuration of edge controls to go with that particular SYSM setting. Then move on and repeat the process with the next SYSM setting, working in sort of hierachical fashion to narrow down the best configurations or range of potential adjustments for each (with a given input/signal).

However, I would not make the mistake of assuming that just because one particular 2170P-3 parameter does"help" with one SYSM setting, that that control should necessarily be thrown out of the mix for the other SYSM configurations. Because as mentioned above some controls seem to work with finer granularity and more subtle effect with SYSM=2/3 than they do with SYSM=1. I'm thinking particularly of F1LV here, which seems to have a much finer and potential more beneficial "defining" effect with SYSM=2/3 than with SYSM=1.

I agree. I think it's probably best to spend quite a LOT of time "equalizing" other controls with the Use of a single SYSM setting and finding what is best. In my case, concerning 480p DVD specifically - over the past ~7 months, that's involved long periods using either SYSM=2 or SYSM-3 for 480p with various other settings, and generally using Ken Tech's finding as a "base" to work from and do "fine tuning" from there.

Last night -- Given my quick "seat of pants" experiments I explained in last post on this -- I should note some of my "experiments" and what I found were not only based on that more "seat of the pants" experimenting session, but also from previous occasions when I'd also played around with those settings more extensively with "SYSM=2" or "SYSM=3", but over much longer periods of time. The info ADU has posted did however, greatly help me not only "put 2+2 toghether" concerning what those controls were actually doing so to speak, but also helped me to find better results with SYSM=3 and 480p DVD than I had been able to come across in previous experiments.

Quote:
Originally Posted by ADU View Post

Although the "best settings" for these may follow not particular pattern of consistency from one TV/user to another (or even one input/signal to another), I think you may be able to get a better feeling for the behavior of these controls and possibly where they might benefit you by playing around with them.

Absolutely, very well put. In my case, it has turned out to be the case that Ken Tech's findings(especially the most recent) have worked out very, very well for me, and other than fine tuning the MID5 Columns, I haven't really been able to "add" any signifcant improvments to using the settings he has come up with - and, it hasn't really been necessary, either.

The only real exception to that was for 480p DVD "back" when I was using SYSM=3 for everything. Prior to the info in his Post #707, I also wished I'd found the time to experiment more with SYSM=2 for other inputs besides DVD, as well as a couple of other things in 3dcomb and 2101-3 pertaining to "SD" via DRC from S-video or internal "RF" tuner. I also "used" to occasionally use a pic mode set up to use a MID5 Column with all "zero's" for all inputs(given use of SYSM=3, and 2170P-3 settings very similar to those in brackets in my earlier post).

Besides, it's "fun" to play around with, when I have the time. Perhaps Especially when you already have a good set of "settings" set up in one Pic mode that are working well for you and can just play around with trying to make improvments in another, "experimental" pic mode you have set up. Ok, sure, I'm a little crazy to enjoy such leisure activities ...


Quote:
Originally Posted by ADU View Post

(and I've reconfigured 2170P-4/USHP to make that setting my Pro mode default.)

Just a FYI -- Even though it's shown in the XBR960 servicecode listings, there is no 2170P-4 UPIC~UTMP, on my KD34XBR960. I have 2170P-4/#6 SHUO, then next is 2170P-4/#7 RYR. "GAMM is control #11", for instance, and so on.

UPIC~UTMP are However present in either the "QM" or "QT"(I'll have to look to be sure which one) "undocumented" sections of SM -- I do recall looking at UPIC at one point to see if I could see what it does, and it seems to behave quite "differently" than is the case with your XBR800. If I recall correctly, it has seperate values per pic mode, and instead of actually "doing anything", it merely contains info on the Set defaults for the "picture" slider for the various picture modes -- On the XBR960, there are "little dots" on the sliders which correspond to where the slider was for the set defaults for the diffierent pic modes, and perhaps that may be what UPIC~UTMP are used for on my set -- to place those little "dots".

Jeff
Nitewatchman is offline  
post #950 of 2973 Old 02-08-2006, 04:31 PM
AVS Special Member
 
justsc's Avatar
 
Join Date: Jun 2005
Location: San Jose, CA
Posts: 2,689
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by ADU View Post

A couple more thoughts on the 2170P-3 edge controls...

SYSM
So far SYSM=3 seems to be giving a bit of a biased look to edges on my TV that's a bit uncomfortable to my eyes. It's possible that tinkering with other controls may be able to make this a bit more manageable, but so far I haven't spend much time trying to do that, and have been focusing more of my attention on SYSM=1 and SYSM=2. Edge distortion is lowest with SYSM=1 at nominal Sharpness. SYSM=2 seems a little "noisier" looking, but I can see alot of potential in the higher granularity and generally better definition it seems to confer especially to lighter details and shadow information in the picture versus SYSM=1....

ADU,

This confuses me a bit. And maybe I've been misusing the terminology. I have been equating edge distortion with noisiness or harshness. I found on my set that SYSM=3 was the smoothest, least harsh setting. I find now that SYSM=2 or 1 seems too sharp or too harsh.

When I see you indicating that edge distortion is lowest with SYSM=1 at nominal sharpness, I sense that I've been misunderstanding the terminology at the least.

Would you be so kind as to help me here? I need to be in sync with how y'all are describing what you're seeing or I'll be bouncing all over the place.
justsc is offline  
post #951 of 2973 Old 02-09-2006, 11:18 AM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by Nitewatchman View Post

FWIW - With ADU's excellent info on the relevant P2170-3 settings printed out and brought along -- Played around with these in a "seat of the pants" manner a little last night With KD34XBR960 and 480p DVD(Zenith DVD2381) via component input. Difficult to say how much of this, if any is "specific" to my DVD player.

Quote:
Originally Posted by ADU View Post

A couple more thoughts on the 2170P-3 edge controls...

These entire posts were interesting to me because they describe some of what I have observed in different words. I think we're verging on discussing "flavor," here! In particular . . .
Quote:
Originally Posted by ADU View Post

So so far, it's been kin dof tough call. Another way I'd describe the difference is SHF0=0 seems more "TV-like", and SHF0=1 seems more "monitor-like", if that means anything.

Yep, this works for me, too. I am trying to avoid the "TV" look and going for something more film-like, more photographic. One of the signatures of the TV look, I think, is the smoothing of fine texture along with oversharpening of medium-fine detail. I generally screen these parameters for their effect on fine texture, and if it is attenuated for some settings, then those settings are verboten on my set. SHF0 is one of those, and so I have abandoned it, leaving it at 1.
Quote:


#14 MVYL - 3
#15 MVYE - 1

The trouble with vertical sharpening is that its scope of action (distance) is always a minimum of one scan line, and on 480i/p material, that's not too fine! My DVD player has a little built-in, so I don't dare add more. Further, if one is trying to rescue fairly soft video, a bit of vertical sharpening "balances out" whatever horizintal sharpening you may be adding. Many good video cameras and both HD and SD broadcasts show an irreducable vertical sharpening, too. But I generally don't like its effect on good source video. What you've listed is a very small amount, however, and, believe it or not, there is still a just-detectable amount still in effect with MVYE=0 if MVYL=3. MVYL has to be at 0 to shut it off completely.

KenTech
"We all get smart slowly."
KenTech is offline  
post #952 of 2973 Old 02-09-2006, 11:52 AM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by ADU View Post

So far SYSM=3 seems to be giving a bit of a biased look to edges on my TV that's a bit uncomfortable to my eyes.

I have come to think of it in audio tone-control terms: SYSM=3 raises high frequencies, but as a *plateau,* much like the standard trebel control on a stereo. When cranked up, there is a response-curve ramp-up at some frequency, and then it levels off for higher frequencies. Thus, very fine detail is not boosted proportionately to fine detail but equal to it. SYSM=2, however, seems to me more like a tilting ramp upwards on the high-frequency response curve. When cranked above "neutral", medium detail is enhanced a bit, fine detail a lot more, and the finest textures most of all. So it has a very different "look" from SYSM=3. At a given SYSM=3 Sharpness boost setting, one could add those finest textures, I guess, by having a corresponding MID5 setting for MHYE. Trouble is, MHYE is *fixed* and SYSM is variable. So it would appear balanced only at one Sharpness setting -- not a very versatile use of Sharpness.

SYSM=0/1 is scary to me! Its scope is quite large (distance of action), and it creates weird double-edge ghosts when cranked down. I have a feeling it was intended for heroic image-rescue, but I can't match it up with anything I use as a source.
Quote:


F1LV: Although this generates fairly coarse looking "noise" with SYSM=1, as mentioned above it seems to have some greater potential with SYSM=2/3, and may have some value in terms of "elevating" shadow information so it's a bit less strenuous on the eyes.

'Scuse a little nitpicking, here: none of these really "generate" noise; I think you meant "accentuate" for what noise that's already present. I'm one of those lucky enough to have impeccable analog-cable service, and so I see very little noise as I try to optimize these SD broadcasts. My greatest source of noise is from grainy DVDs, and they abound! E.g. The Matrix trilogy is *very* sharp, but the film grain is clearly visible on smooth textures. In-player 3DNR is very useful, here.
Quote:


CTLV: 0=highest color saturation at edges. 3=the least, and generally softest-looking color detail. Seems fairly straight forward.

Very clear description. Motivates me to experiment, too. Gotta keep an open mind!

KenTech
"We all get smart slowly."
KenTech is offline  
post #953 of 2973 Old 02-09-2006, 12:02 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by Nitewatchman View Post

Very interesting. Using component input -- While I do find I need to do "more" with the image processing settings for 480p DVD that is the case anywhere else -- I haven't really noticed a limit on detail. Maybe I should try to look into this in more detail.

Problem is, I've never been happy with the 480i from this particular player(Zenith DVD2381), including when using it with other sets, but I've allways been very happy with 480p from it.

What is particularly frustrating is that, absent a few $$$ pieces of test equipment, none of us can really tell whether a particular artifact or characteristis is caused by the source (say, DVD) or the TV. My Panasonic S97 is well-known to have stunning component output, and I have read reviews of other players who performs far less well on component than HDMI. Component performance seems to vary a lot!

Another example: I don't like the *very* soft look to my player's upscaling to 1080i over HDMI -- on my TV, anyway. I have a friend who just bought the Sony D50XBR1 SXRD 1080i RP unit. He bought the same player I have. His component desplay is good, but is forced to upcscale to 1080p. *His* 1080i over HDMI is great from this player! Uh . . . sort of implicates the 1080i/HDMI capabilities of our beloved TVs, eh? Go figure! 1080i digital input on a (purported) 1080i-native TV. Wonder where the fault lies in the processing chain.

KenTech
"We all get smart slowly."
KenTech is offline  
post #954 of 2973 Old 02-09-2006, 12:08 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by ADU View Post

Lately I've been using this pattern to set that adjustment, looking, as I say, for minimum sharpening or softening distortions on the vertical lines.

It's good to double-check with edges that are not part of 1-2-pixel lines, too, since a 1-pixel line can cause an abnormal ringing that almost never appears in real program material. The "Sharpness" pattern on the AVIA disk is particularly revealing, with its crossed 1, 2, 3, and 4-pixel lines, plus white and black blocks that are devoid of antialiasing. (The circle and text are antialiased.)

Of further help on the same disk are the various sub-menus. On the big colorful main menu screen, almost everything *is* antialiased (softened) slightly, but the sub-menus are hard-edged graphics. Getting those images free of artifacts -- or even the main menu -- is quite an acid test!

KenTech
"We all get smart slowly."
KenTech is offline  
post #955 of 2973 Old 02-09-2006, 12:28 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by justsc View Post

This confuses me a bit. And maybe I've been misusing the terminology. I have been equating edge distortion with noisiness or harshness. I found on my set that SYSM=3 was the smoothest, least harsh setting. I find now that SYSM=2 or 1 seems too sharp or too harsh.

We're all reduced to metaphor and simile. Next thing you know, we'll all be referring to "airy" detail and "muddy" lower-upper-mid-bass. Er, I mean -mid-detail. (Sorry. I have worked in the high-end audio industry. Hope nobody holds that against me!)

I make all of my comments about edges from close observations, ignoring the picture -- and generally using known test patterns, the AVIA disk being especially valuable. Beyond that, there are certain frames on Monsters, Inc and other good DVDs that are revealing of texture and diagonal/curving fine lines (Sully's hair!). I see ringing and overshoot mostly from this close-up vantage point.

But when I sit down to watch a known-good DVD at a reasonable distance, paying attention to the overall fidelity of the image, I see characteristics that are best described as "smooth" or "harsh" or "grainy" or, well, *ugly* or *beautiful.* The eye/brain combination is integrating everything and "locking" onto the image. The 11pm newscasts here all all different from each other, but one is *very* fine. I can look at the faces of these (perfectly-lighted) newsreaders, and they might jump out as nearly 3D and authentic, or have lots of texture and detail but are *wrong* somehow. That's how I've come to reject certain of the filters: They do something that theoretically *should* help (according to test-pattern observations) but something's not right when viewing real video.

So, for me, ringing, edge-distortion, overshoot -- those are all thing I can see and identify only up close. What effect they have on real video I describe as harsh, unauthentic, finely-detailed, balanced, "TV" look, etc. Occasionally something that looks like (e.g.) overshoot on a test pattern really helps real-world viewing. I think that's the "compensation" factor coming into play. Describing it here is kinda tough.

KenTech
"We all get smart slowly."
KenTech is offline  
post #956 of 2973 Old 02-09-2006, 02:51 PM
AVS Special Member
 
SurfingMatt27's Avatar
 
Join Date: Dec 2005
Location: Massachusetts
Posts: 2,789
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 12
I've been tinkering with those MIDE-5 tables last year,and aftyer al the tweaking i come to the conclusion that simply turining them all offto 0 is the best solution to a better picture quality.

No matter how many you have on or little as possible, you r still adding stuff to the picture that just is'nt there.You can't expect a stock honda to beat a porsche lol.

Maybe it's me, but i find the less enhancements the better the PQ looks and more cleaner iand sharper it looks.With edge enhancements on it agve me a sharp picture but it was unnaturaly sharp kind of like a cookie cutter sharpness appearance.

I would rather have my tv do no processing and just feed the signal to the screen,any EE will be from the source material not my tv set.

The PQ looks razor sharp without them EE enhancemnts,so they really are'nt neccessary it just makes the PQ unnatural.With EE off the PQ is naturally sharp they way film should be.

just my 2 cents.
SurfingMatt27 is offline  
post #957 of 2973 Old 02-09-2006, 02:55 PM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,482
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 176
Quote:
Originally Posted by Nitewatchman View Post

Just a FYI -- Even though it's shown in the XBR960 servicecode listings, there is no 2170P-4 UPIC~UTMP, on my KD34XBR960. I have 2170P-4/#6 SHUO, then next is 2170P-4/#7 RYR. "GAMM is control #11", for instance, and so on.

Interesting.
Quote:


UPIC~UTMP are However present in either the "QM" or "QT"(I'll have to look to be sure which one) "undocumented" sections of SM -- I do recall looking at UPIC at one point to see if I could see what it does, and it seems to behave quite "differently" than is the case with your XBR800. If I recall correctly, it has seperate values per pic mode, and instead of actually "doing anything", it merely contains info on the Set defaults for the "picture" slider for the various picture modes -- On the XBR960, there are "little dots" on the sliders which correspond to where the slider was for the set defaults for the diffierent pic modes, and perhaps that may be what UPIC~UTMP are used for on my set -- to place those little "dots".

Actually that sounds pretty much the same as on the XBR800. On the XBR800, 2170P-4/UPIC through UTMP configure the default positions of the sliders and temp settings for each picture mode in the User Menu. For slider controls, the numbers correspond to ticks, with 0 representing the first tick, and 31 representing 32 ticks. If, for example, you wanted the Brightness slider to default to say 27 ticks instead of the usual 32 in Pro mode, you could change that by making UBRT=26. Same for Sharpness, Contrast, etc.

So why use this approach to reconfigure Sharpness? Well, because the 2170P-3/SHOF sharpness offset is already at it's lowest setting for the DVI signal/input I'm using on my XBR800, and Sharpness is still not "optimal" w/o further reducing the slider in the User menu.

Perhaps I should have mentioned these adjustments earlier when you were having some of your Brightness dilemmas. But I figured you were probably already aware of this, and wanting to find a way to get Brightness in the SM to work with the existing slider configs in the User Menu.

ADU
ADU is offline  
post #958 of 2973 Old 02-09-2006, 03:55 PM
AVS Special Member
 
justsc's Avatar
 
Join Date: Jun 2005
Location: San Jose, CA
Posts: 2,689
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by SurfingMatt27 View Post

I've been tinkering with those MIDE-5 tables last year,and aftyer al the tweaking i come to the conclusion that simply turining them all offto 0 is the best solution to a better picture quality.

No matter how many you have on or little as possible, you r still adding stuff to the picture that just is'nt there.You can't expect a stock honda to beat a porsche lol.

Maybe it's me, but i find the less enhancements the better the PQ looks and more cleaner iand sharper it looks.With edge enhancements on it agve me a sharp picture but it was unnaturaly sharp kind of like a cookie cutter sharpness appearance.

I would rather have my tv do no processing and just feed the signal to the screen,any EE will be from the source material not my tv set.

The PQ looks razor sharp without them EE enhancemnts,so they really are'nt neccessary it just makes the PQ unnatural.With EE off the PQ is naturally sharp they way film should be.

just my 2 cents.

Matt,

I really do understand where you're coming from.

I'm just wondering - don't you think there's a chance that there is some detail that may be there, but not yet visible? Kinda like when it's a dark scene and by turning up brightness you find that there is detail you didn't even know was there? So maybe you bump up your brightness setting just a little so you can see what the director intended for you to see.

Now that's not adding anything to the signal, it's just "bringing out" or "exposing" the detail.

My interest is in exposing as much of the "intended" detail as I can. And I see there may be more than one way of doing that. And it's important to understand that our tv sets, as they've been manufactured and set-up at the factory, may not come to us in such a condition that we can view all this intended wonderful detail - that's why we calibrate right?

The first task is to explore the settings to discover those that "expose" those details without adding things in.

The second comes from having a full understanding of what the tv's s/w and h/w is doing to the signals. We might come to learn that there's circuits or code that actually "dial back" or attenuate details within the signal so that something must be done to "dial" those details back in. But maybe there's no suitable adjustment that does this, and the signal has to be massaged so that, in the viewers perception, from the appropriate viewing distance, this once attenuated "detail" is now exposed. And it could very well be that the only way to see this detail is by "adding in" something, possibly a little edge enhancement, so we can "perceive" what would have been there.

Either approach is simply to bring out what we were supposed to see. This is why I don't exclude the possibility of some edge enhancement.
justsc is offline  
post #959 of 2973 Old 02-09-2006, 04:32 PM
AVS Special Member
 
Nitewatchman's Avatar
 
Join Date: Dec 2001
Location: Middletown, Ohio
Posts: 6,292
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 18
Quote:
Originally Posted by ADU View Post

Perhaps I should have mentioned these adjustments earlier when you were having some of your Brightness dilemmas. But I figured you were probably already aware of this, and wanting to find a way to get Brightness in the SM to work with the existing slider configs in the User Menu.

1st off, please stay with me here until the next section and attached screenshot ....

No, there was no need for you to mention those, but thanks. I'm afraid you are still misunderstanding the issue I had with getting the "picture" slider centered, instead of using it at "24" which worked just perfectly, all I was trying to do was to get the picture slider to read "31"(32 ticks) and achieve the same results as was the case with "24"(25 ticks). I didn't have any problem "issues" whatsoever with brightness, or "contrast" or balancing those(or the "sharpness slider either for that matter) among inputs/etc, just wanted to center the Picture slider at "31".

I perhaps used the wrong terminology --- In those posts I was talking about "brightness" as it relates to "white balance" and "contrast", NOT brightness as it refers to the "brightness" slider, or "black levels", the latter I took care of long ago with 2170P-3 UBOF and 2170P-1 SBRT.

It's all water under the bridge, now though, as I've already explained I was able to figure it out using a lower value for 2170P4 SPIO (and, long ago I had already used 2170P-4/SPIC to balance contrast among inputs, although I touched those up a bit anyway when I was working on this). Sorry, I probably did not do a good job explaining that, but I can't think of a better way to say it than I already have(along the lines of what Ken said, sometimes it is awfully difficult for me to talk about these issues and know that my comments are "understandable" by others), and I'm afraid going into that again and repeating it again won't be useful.

Quote:
Originally Posted by ADU View Post

Interesting.Actually that sounds pretty much the same as on the XBR800. On the XBR800, 2170P-4/UPIC through UTMP configure the default positions of the sliders and temp settings for each picture mode in the User Menu. For slider controls, the numbers correspond to ticks, with 0 representing the first tick, and 31 representing 32 ticks. If, for example, you wanted the Brightness slider to default to say 27 ticks instead of the usual 32 in Pro mode, you could change that by making UBRT=26. Same for Sharpness, Contrast, etc.

So why use this approach to reconfigure Sharpness? Well, because the 2170P-3/SHOF sharpness offset is already at it's lowest setting for the DVI signal/input I'm using on my XBR800, and Sharpness is still not "optimal" w/o further reducing the slider in the User menu.
.

Interesting. I don't think it's quite working the same, at least with "mode memory" turned on on 34XBR960 in as "useful" a manner.

Fom what I recall from checking this, On KD34XBR960, QM UPIC~UTMP do nothing to actually "configure" the sliders(update: except to allow you to hit "reset" button to return to a certian set of default values as Ken mentions below/End update) , other than if you use the little "dots" shown in the slider menu as a reference to return the values for the sliders back to factory defaults for any given pic mode, or if you change the values for UPIC~UTMP, your "new" values would correspond to new posistions for these "little dots", which otherwise mark the factory defaults for the sliders for the different pic modes.

They only SHOW you [update again : or let you use the reset button to return to a certian set of "defaults" - either the factory, or values you've set up ; end update], in the "video" user menu what those factory defaults were, with little "dots". They don't effect what happens to the picture, you do that by changing the slider yourself, which you can do on KD34XBR960 in a Input specific manner(in other words you can use different slider values for different inputs(or pic modes for that matter with "mode memory/On) by turning "mode memory" on in user menu, or you can use the same slider "settings" for all inputs by turning "mode memory" to : off.

:update :

Again, as noted in my reply to Ken farther below - sorry ADU for my "misunderstanding" of what you were talking about on this -- Entirely my fault, but, rather than editing them except for these updates as noted, I think I'll go ahead and leave my comments as they were in this post, if nothing other than a demonstration of how easy it can be to "misread", or "misinterpet" things - again, in this case, entirely MY fault.

Because of the availability "mode memory: On" function and different slider settings(or other input/scan rate/etc. offset controls available in SM - such as "UBOF") for all the different sources I use, I've never really thought of a reason to make use of the reset button, or to define a certian set of "defaults" with QM UPIC~UTMP settings for the User mode "sliders" ... Although, thinking about it more, it could perhaps be useful for me in order to more quickly "switch" to a "default" set of slider settings when "checking" things, or, perhaps to set the sliders quickly back to my "normal" mid range settings after adjusting to preference to compensate for "differences in programming material."

:end update

Perhaps a picture is worth 1,000 words. So, attached is screenshot "sliderdots.jpg"(pardon the quality of the shot and the reflection from a lamp in the room) -- I've adjusted the sliders(this is NOT what I use them at, LOL) so you can see all the "little white dots" which UPIC~UTMP correspond to for "factory Mode" for each slider. For instance -- UBRT for this Pic mode = 32, and that's right where the little white dot is for the "brightness" slider that shows the "brightness" slider factory default for this pic mode. The "brightness" slider would JUST "cover up" the little white dot if I had the slider set at "31" as I normally do. If I change QM/UBRT to "45" for this pic mode, the little white dot will move up to correspond with the "44" value for the slider itself, but, it doesn't effect the actual value of the slider itself in any way shape or form -- nothing else will change except the posistion of that little white dot. Perhaps it may behave differently however if I wasn't using "mode memory" set to "on" - I didn't try that.
LL

Jeff
Nitewatchman is offline  
post #960 of 2973 Old 02-09-2006, 06:38 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by SurfingMatt27 View Post

No matter how many you have on or little as possible, you r still adding stuff to the picture that just is'nt there.

Actually you are increrasing the level of stuff that *is* there, but if you don't like the look, then by all means leave them all at zero. You have to please only yourself in this matter. But I wouldn't advise that this is somehow a "correct" setting.
Quote:


Maybe it's me, but i find the less enhancements the better the PQ looks and more cleaner iand sharper it looks.With edge enhancements on it agve me a sharp picture but it was unnaturaly sharp kind of like a cookie cutter sharpness appearance.

Then you really had too much of something added in, and maybe it came from *before* the MID processing. If the source is 480i, it could be the 2103 chip, which adds a horrible "TV-like" edginess if left untamed.

KenTech
"We all get smart slowly."
KenTech is offline  
Reply Direct View (single tube) CRT Displays

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off