LCD Buyer's Guide

Page 17 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

xtknight

Elite Member
Oct 15, 2004
12,974
0
71
ddekany: Does your 970P have pixel dither patterns or is it actually moving noise? I can see moving noise on my VP930b in one color if I set the gamma to just the right setting. And a rather ugly liney effect as well. I've never known what it was.

This is how the VP930b looks to some extent (worse case scenario: brightness=100, contrast=0, gamma=nonlinear curve): http://xtknight.atothosting.com/vp930b-dotcrawl1.png

The first is noisy and the second has alternating slightly red-shaded lines.

Notice the noise. The red literally moves around in there.
 

ddekany

Junior Member
Jan 31, 2006
18
0
0
Well, I'm not involved in the TFT manufacuring business, so I'm not sure if 8 bit is really a hard rule for a PVA or rather just a habit because PVA-s are more expensive than TN-s. All I can say that 970P (this series at least) surely does 2x2 dithering, and it does it with all monitor settings, no mater what. And it doesn't look like TN because of the viewing angles. BTW, 193P+ did 2x2 dithering too according to some other photos.

(Back to the PVA is always 8 bit thing... I'm ignorant at the low level details of TFT manufacturing, nor do I know anything about the internals of PVA vs TN... but AFAIK, the TFT panel itself (the glass sandswitch with the liquid crystal and voltage keeper condensators and gate transistors and color filter and polariser layers) is an analog unit. So the whole "how many bits" question start to have meaning only when you attach the source driver IC-s + timing controller IC to the panel and whatever other stuff like that. These are all inside the monitor, but still, this last step is a separate step. It can be done by other companies sometimes and like... Now, which part of the whole is the "PVA"? Maybe just the TFT panel without the source drivers and like? Because then a PVA panel can be 6 bit as well, you see. Anyway, again, while I'm surely know what is dithering and what is not, I'm ignorant in these low level manufacturing details... just thinking loudly.)
 

ddekany

Junior Member
Jan 31, 2006
18
0
0
The pixel dither patterns don't move at all; they are exactly like if you do it with some drawing program. BTW this is why I find them annoying; I can't say if the dots are really there on the image, or just the monitor shows them. I have tried a lot of monitor settings, frequencies, 3 totally different computers, DVI, D-SUB... and again, have seen the photos of some other people. It always do 2x2 dithering on the same extent. Because, I bet it has only 64 degrees per subpixel, so it had to. (Or, it could use FRC for the trick as well... but it doesn't do it seems.)
 

xtknight

Elite Member
Oct 15, 2004
12,974
0
71
Another simulation of the lines I'm seeing in one gray transition: http://xtknight.atothosting.com/vp930b-dotcrawl2.png

Yeah maybe PVA isn't always 8-bit. All I've seen have been reported as 8-bit though.

You seem to have a better idea than I of the internals so unfortunately I couldn't help you there. As far as I know, all LCDs are technically based off twisted nematic crystals though (I could be wrong of course ). Perhaps the crystals on the PVA panels can be adjusted with more precision to let just enough light in, whereas on the TNs it's more rough and can't control the incoming light as easy.

There is also the factor of gamma. Eizo has an article on it here, but apparently with 14-bit precision gamma the "color seepage" in the lower grayscale spectrum can be eliminated. Instead of color seepage like I saw on my TN, I now see color seepage AND noise on my P-MVA. http://www.eizo.com/support/wp/pdf/wp_04-006A.pdf

The noise on mine is actually moving, and it's rather distracting if you had to look at that color for a while. Red literally just pops out of nowhere within a 10x10 square about once every half a second. It doesn't happen in just grayscale either. Talk about it tomorrow I guess but thanks again for making us aware of it. It would be nice if you could stay around on these forums.
 

ddekany

Junior Member
Jan 31, 2006
18
0
0
I don't really get what those 2 images meant to be... The noise that you refer to usually looks like when you have weak reception with an old TV, and is mostly visible in the dark colors? Then, don't you happen to use D-SUB instead of DVI?

Regarding the gamma and seepage... the thing is that even if, let's say, a panel has 8 bits (256 grades) per subpixel, and doesn't apply tricks like FRC, then those 8 bits will not be enough to correctly display 16,7M, execept at a single contrast + brightnes + gama + color-temperature setting. So it is not just gamma, almost all monitor settings do mater. If you have different settings from the said ideal one, then there will be some banding, seepage, etc. Think about it as mapping the values from 0-255 (input coming from DVI) to numers between 0-255 (value that drivers the brightness of the subpixel). If you don't do an same-to-same mapping here (i.e. 0 to 0, 1 to 1, 2 to 2, ... N to N), then you necessarily will have some repetations (0 to 0, 1 to 1, 2 to 1, 3 to 2) or jumps (0 to 0, 1 to 1, 2 to 3, 3 to 4) that will cause banding, or you may reach white to early (251 to 253, 252 to 254, 253 to 255, 254 to 255, 255 to 255). Also if this mapping is different for R and G and B subpixels (for example because of color temperature settings), then you will have some incorrect color (seepage). But with tricks like FRC and dithering you can have more grades (more precisely, dithering, at least in the sense as 970P does it, doesn't add any extra grades if we look at an individual pixel). Said EIZO document seems to say that his monitor is real 8 bit, plus it uses FRC to seemingly achieve even more grades. Now, Samsung 970P seems to have subpixels with 64 grades (6 bit), and then it applies dithering to apparently achieve 256 grades... (or rather just 253, that only gives 16.2M colors, not 16.7). So, if you look at a gray gradient (not a real-world image like a photo or a game) 970P has obvious banding and seepage, except if you set contrast to 50, brightness to 50 or more (because over 50 it just gradually increases the streangth of the backlighting), gamma to 0, and turn off color temperature correction. Even then, because of the dithering, the gradient will have an awkward texture (checkboard-like patterns). But of course, you hardly ever watch gradients in real usage... Still, I see saying 16.7M for this monitor as a marketing trick.
 

xtknight

Elite Member
Oct 15, 2004
12,974
0
71
ddekany: I'm using DVI, native resolution, and 32-bit color. I think the lines and noise are the VP930b's method of dithering. It's moving noise, believe it or not. It's possible I got a defective panel maybe but since you caught the 970P using dithering, I think my VP930b is doing the same. Mine only occurs in very rare instances (just one or two colors on a whole 256-level grayscale) so it's not unthinkable that they'd sell such a thing. I had briefly used my 6-bit TN on my new video card, and had not noticed any noise. It's got to be something with the monitor. I have modified the Guide to take into account and credit your discoveries on the 970P. If you don't mind I've also added a link to your page about the 970P.
 

imported_ST

Senior member
Oct 10, 2004
733
0
0
...in about a month, the proliferation of th next gen larger 1080p LCD TV's will come (42" / 47"), with introductions from Sceptre, Westinghouse, Vizio, etc. I'll be trading mine up to the 42" and will have a full review then.
 

ddekany

Junior Member
Jan 31, 2006
18
0
0
Originally posted by: xtknight
ddekany: I'm using DVI, native resolution, and 32-bit color. I think the lines and noise are the VP930b's method of dithering. It's moving noise, believe it or not. It's possible I got a defective panel maybe but since you caught the 970P using dithering, I think my VP930b is doing the same. Mine only occurs in very rare instances (just one or two colors on a whole 256-level grayscale) so it's not unthinkable that they'd sell such a thing.

There is a method that uses "moving" patterns, called FRC. This is what most 6 bit TFT-s use, as far as I know. Maybe it depends on the viewever's eye and brain, but I preceive FRC as if many faint flickering diagonal lines are crawking in certain colors. But I doubt anybody sees it as random noise, so after all, what you see is maybe not FRC.

Note that if you use dithering or FRC or whatever method to get near the 16.7M color with 6-bit, you will have to use the chosen trick for about the 75% of the grays, not only for 2 of them. So you should rather say that you perceive the strange effect in 2 colors. (However, if you have a 8-bit monitor, then maybe it uses the trick only for a very few colors.)

I had briefly used my 6-bit TN on my new video card, and had not noticed any noise. It's got to be something with the monitor. I have modified the Guide to take into account and credit your discoveries on the 970P. If you don't mind I've also added a link to your page about the 970P.

That's fine.
 

xtknight

Elite Member
Oct 15, 2004
12,974
0
71
ddekany: Looking VERY closely, I can see an extremely faint series of blue lines move horizontally across a grayscale as I adjust the contrast up and down. This must be dithering/FRC also on my P-MVA. And you're probably right, I can only perceive it on two colors. Actually I can see some pixel movement on other colors too I think but it's so hard to see I don't know if it's there or not. I've sent AU Optronics an e-mail and am awaiting a response on this. (E-mailing ViewSonic is like talking to a wall.)
 

ddekany

Junior Member
Jan 31, 2006
18
0
0
As I said in the FAQ on my page, don't try to call dithering phenomena that are not that... dithering is the standing checboard-like pattern as the figure clearly shown there. If you see faint blue colorisation in some grays that's probably because your LCD uses some color correction (otherwise the image would be too yellow), and the increase jumps in the birghtness of the R and G and B subpixels hapens not exactly at once. This is normal for TFT-s. As far as I know this is called color seepage. 970P does it too, unless I set color temperature to "off" with MagicTune, in which case the image is bit too yellow. It has nothing to do with dithering.
 

Luddite

Senior member
Nov 24, 2003
232
3
81
Here is a new review for the Samsung 970P from CNET. They rated it as 7.4 ("very good") with great image quality:

"The 970P's grayscales were smooth, consistent, and free of color-tracking errors, save for tiny flushes of pink. The 970P's colors were bright and consistent and changed hues uniformly. DVD-playback performance was slightly below average; we saw overly red flesh tones, areas of overexposure, and lots of noise. Gaming performance was much better, with crisp backgrounds and vibrant colors."




 

xtknight

Elite Member
Oct 15, 2004
12,974
0
71
I should have mentioned they were alternating lines. Adjacent pixels within the same color value and some have lines and some don't. Additionally, fuzziness is noticeable with certain colors much more than any others.

Maybe this will help explain what I'm talking about. In this image, the top colors look almost exactly like the bottom colors on my screen: http://xtknight.atothosting.com/vp930b-reprod.png
 

ddekany

Junior Member
Jan 31, 2006
18
0
0
I see. I don't know what's that exactly (you may take a photo of that)... However, note that even 8 bit panels may use some kind of dithering or similar trick, because 8 bit is not enough to display 16,7M in reality, since the contrast/brightness/gamma is adjustable and also some color corrections may be needed. (In the case of 970P, however, the extent of the effect is far too much for 8 bits... It's what one could expect from a 6 bits, like the 75% of grays are dithered, whatever monitor settings you chose.)
 

xtknight

Elite Member
Oct 15, 2004
12,974
0
71
Originally posted by: SoylentG
So what's 8-bit mean? I'm a bit lost on that...

8 bits per subpixel. 3 subpixels = 3 color components (R, G, B). (2^8) ^ 3 = 16777216 unique colors. How it applies to LCD panels? It means the crystals can twist to more angles to let different amounts of light in. The end result? Generally, more perceived colors. But see ddekany's post above about this.

6 bit panels (and some 8 bit apparently) need to dither (blend pixels to simulate actual perceived color by humans), since natively using the same math above and 6-bit, they can only display 262,144 colors.

"32-bit" under Windows XP is 24-bit (8 bits per subpixel) and an 8-bit alpha channel (for ease of programming). The alpha channel is handled by the graphics card and everything is converted from ARGB into RGB space again, as far as I understand.
 

xtknight

Elite Member
Oct 15, 2004
12,974
0
71
I guess IPS/VA doesn't always mean 8-bit after all. Look at this. 262K color IPS panel. http://www.idtech.co.jp/en/products/index.html

It says in the PDF, "RGB 6-bit data driver". The panel and crystals themselves may be capable of it, but beware, they may cut corners in the controlling electronics! I believe we can still be sure TN panels are still always 6-bit.
 

Mice007

Junior Member
Feb 4, 2006
2
0
0
Hi

i've just joined to this forum. The Guide is really good!

Also i have to agree that some _not_ TN panels can't reproduce true colors. My previous monitor was a Samsung 193P (from the first series with 800:1 contrast ratio).

I've discovered issues about color reproduction in the first days compared to my old iiyama CRT but it was acceptable. I've working on a monitor diagnostics software and i saw strange gaps/patterns on the gradient tests (single color goes black from '255' through the screen) but i've found this caused by color management on LCD-s (MagicTune or Adobe Gamma for example). Also (thanks to ddekany's website) i've found that the patterns on this gradients are not color management results but the 6bit dithering! I'll make a new test screen about this issue for sure.

Also i've just bought a Dell 2405 FPW and it seems this samsung panel is true 8 bit stuff.
 

Mice007

Junior Member
Feb 4, 2006
2
0
0
8 bit means the monitor is able to display 16.7m colors "ideally" or let say.... with the default parameters.... but in the case you color manage your monitor you'll lose several shades from the true color palette.
 

Michael

Elite member
Nov 19, 1999
5,435
234
106
xtknight,

I bought a VP930b last week. I noticed that your main guide now lists it as 6-bit instead of 8-bit even though everything I've found for the panel says 16.7M colours instead of 16.2M colours. Dithering doesn't equal 6-bit. You have to dither (FRC, whatever) an 8-bit panel too just to handle all the colour settings. I haven't been running it much as I just have it set up next to my old 19" LCD (going with my old computer for my daughter). I did fool around with having both monitors going at the same time and it is cool to have 2 but I'm not sure is it is worth the extra $$$.

What made you decide that the monitor is definately 6-bit? Did you find something that showed that the driver ICs were only capable of 6-bit? From the spec sheet I pulled up, it clearly states 8-bit RGB. Spec Sheet

On another topic, I'm the opposite of the "extra sensitive" you describe yourself as being. I barely notice the ghosting on my old monitor. My problem is that my colour sense is not good enough that I feel comfortable setting the monitor properly. I can get the grey scales OK (and I don't see the dithering that ddekany noted in his 970P examples on the VP930b). I've been thinking about getting a Gretag Macbeth Eye One unit. I figure that I have 4 monitors in my house plus a work laptop and the cost per screen goes down (suggested price is $250 US) and I wouldn't have to worry about "tweaking" by sight alone to get the colours right. Do you (or anyone else) have any experience with Gretag Macbeth?

thanks,

Michael
 

xtknight

Elite Member
Oct 15, 2004
12,974
0
71
Michael: I did notice that spec sheet too. I might have a slightly different panel (EG01 is not the one quoted by Tom's Hardware but it's the one that best matches). But what I'm seeing on my VP930b is definitely some form of simulation. Either that or it's the pixel "swarming" or noise they see on S-PVA/P-MVA panels, I'm not sure. But alternating lines kind of makes you wonder. I couldn't find the spec sheet for the Samsung panel, but if that said 8-bit, I'd know not to trust even that. Well, it may be true, but Samsung/ViewSonic's monitor division might not add 8-bit electronics to it. One would think it's easier just to do the whole spectrum than to do less and have to implement a dithering algorithm, but I guess not? Can you look at the picture I posted above and set your VP930b to the settings listed in the picture, and see if the colors look the same on the top and bottom (they should not)?

Originally posted by: Michael
Dithering doesn't equal 6-bit. You have to dither (FRC, whatever) an 8-bit panel too just to handle all the colour settings.

262K=6-bit
16.2M=6-bit+dithering/FRC
16.7M=true 8-bit

As far as I know. (Of course you can't trust the manufacturer's listed spec on that.) But I was pretty sure 8-bit reached 16.7M (24-bit RGB color) to start with.

No, I don't have any experience wtih any of the calibration tools (not the 'spyders' you stick on your monitor). But I've fussed around with my monitor using my eyes since the beginning of time. I still find that SRGB mode on the VP930b and gamma=1.0 (default) in your display control panel is the best combo for entertainment. For text, I just set what's easy on my eyes and don't worry about the colors. I just switch between user and SRGB depending on what I'm doing.

Also, darXoul says:
The Polish site mva.pl tested the VP930 some time ago, and they claimed the 8-bit color depth was achieved through dithering. Another confirmation then.
 

Michael

Elite member
Nov 19, 1999
5,435
234
106
xtknight,

From the panel finding link in your first post:

"ViewSonic VP930 has a 19 inch 8 ms (g2g) P-MVA (AUO M190EG01 V0) panel."

The spec sheet is the same number and it says 8-bit.

I tried looking at the picture you posted and I don't see the same thing you're seeing. I haven't had a huge amount of time to fool with it and I admit that I'm in the majority that doesn't see the edge cases that the very sensitive do see. I did take a magnifying glass out to "zoom" the pixels and still no sign of what you're seeing.

I would write Viewsonic again and point out this thread to them and ask to be sure.

Again - there's a difference between dithering to get to 16.7M colours (I think it can only be 16.2 but who knows what the marketing people say) and being 8-bit and dithering or using FRC to get the full range of colours you need for all the warm/cool and other settings. I don't think you should state that it is 6 bit based on you seeing some line move on the one monitor you own, but I am very curious what the real answer is.

Michael



 

xtknight

Elite Member
Oct 15, 2004
12,974
0
71
Michael: I have put a question mark in that space for now then. What's the manufacturing date of your VP930b? Check with software called softMCCS (changing settings could be dangerous unless you know what you're doing). Mine is: 2005 ISO week 46. Mine happens at both 60 Hz vertical and 75 Hz. It's important to note that this only happens at about three certain contrast levels. Increase your contrast from 0-100 as you carefully eye the top pictures and see if you see any lines appear. If not, I guess I'm just one unlucky soul. Or my video card's 2D electronics are defective (but I never noticed this on my CRT or 17" TN LCD). I guess I could go and try Linux and see if I see the same thing. I really doubt it's Windows though (as bad as it is ). It's not bothersome enough for me to do any in-depth scientific investigations of it. I have only noticed it once on my desktop picture at certain settings, but other than that I have never, ever noticed it in practical usage.

It's common for manufacturers to swap panels within the LCD. Dell does this frequently, and ViewSonic has used two different AUO panels for the VP191b (VP191b-1 and VP191b-2).

I have thought about thoroughly researching the panels in the recommended LCDs and adding response time (max.) to the specs in the list. What do you think about that? I'm not sure if it would mislead people when they see a 50 ms. on a VA+overdrive LCD (reaches that high in rare circumstances). But I want to try something different.
 

Michael

Elite member
Nov 19, 1999
5,435
234
106
xtknight,

Based on what the box says, my unit was made in September 2005. I went to the Polish site you mentioned and did a little online translating of the review:

"In specification served full, bit palette of color is attainable due to employment of unit 8 FRC, dynamic realizing < realize > dithering. There is observable from close distance mainly and it is not obliged to prevent in normal use in (to) majority employment excessively. At viewing of movie, but in smallest degree of photo, however, it is possible to observe certain image graininess, caused work exactly FRC. It at displaying microexemplar one of test image confirm also ? exert (show) intensive tremor."

Another translation of the first sentence:

"Quotation on packing list full capacity bitowa pallet bouncers is not obtainable thank employment unit FRC , realising dynamic dithering"

The difference between the two is that one says that it is achieved using FRC and the other says the specifications are not achieved and FRC is used. I'm guessing that the monitor uses FRC which does not mean it isn't 8-bit. However, in order to represent the full range of colours across all of the settings, even the 8-bit monitor has to use FRC. If there's a native Polish speaker who can translate Polish review of VP930b , please help out.

The European site for Viewsonic lists 16.7M colours, the USA site actually is silent on how many colours it has.

If the unit is using FRC, then the edge cases you're seeing will happen. Most of us would never notice them, though.

Michael

 

imported_Mant

Member
Mar 3, 2005
103
0
0
Hey xtknight, excellent guide. Its tough to find a good guide with some up-to-date recommendations, Im glad you've kept this thread current!
Im in the market for a LCD because my 17"CRT is about to head south (screen is starting to flicker at startup). I'm not sure if I should wait till it actually dies though, because I saw an article here at DailyTech that said prices are about to fall as the manufacturers get new factories up and running.

Anyway, my question is about that BenQ FP91G+ 19" that you just added to the list. I am a gamer and I haven't heard anything about that monitor yet, but the Egg has it (after rebate) for $230 right now! I'm pretty cheap, but I'd be glad to shell out 230 for a good 19in LCD. So how do you think that monitor compares to the Samsung 940B for gaming, which is $100 more? And have you seen a review somewhere you can link me to, this reply is partially to see if anyone that has this monitor to post their experiences woth it too.

Thanks, keep up the good work.
 

xtknight

Elite Member
Oct 15, 2004
12,974
0
71
Due to the panel inside the FP91G+ I think it may be even a bit faster than the 940B, but I have no confirmation of that. I'm sure either will be perfectly desirable for gaming. I tend to like Samsung's monitor interface (though I have never seen BenQ's). Colors might be a tad better on the Samsung, but I am just guessing at this point.

You might find a review on one of these sites. Unfortunately two are in foreign languages (Polish and Danish) but you can see the color calibration graphs, etc. I'll tell you if I see any when I can (probably few hours).

http://www.mva.pl/
http://www.flatpanels.dk/
http://www.behardware.com/

Edit: I'll get an answer for you on that by tomorrow afternoon.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |