Challenges in building a 10 bit display (or higher)

Everything about displays and monitors. 120Hz, 144Hz, 240Hz, 4K, 1440p, input lag, display shopping, monitor purchase decisions, compare, versus, debate, and more. Questions? Just ask!
Post Reply
spacediver
Posts: 505
Joined: 18 Dec 2013, 23:51

Challenges in building a 10 bit display (or higher)

Post by spacediver » 21 Jan 2014, 18:06

This is something I've been wondering about for a while.

The majority of (non CRT) displays (whether for computing or video) are limited to 8 bit. 10 bit or higher seems limited to premium or reference displays.

I am fairly ignorant when it comes to LCD panel technology, but my understanding is that the primary variable that determines how much light is displayed is the degree to which the molecules rotate, which is in turn a function of the applied voltage.

So, assuming you could quantize the voltage finely enough for 10 bit processing, in principle, you could have a 10 bit display.

My guess is that there are noise limitations in the way the molecules react to voltage. Perhaps they don't react cleanly enough to respond with 10 bit precision.

Or perhaps quantizing the voltage in the driving circuitry is the issue.

Does anyone have any insights into the challenges in creating a 10 bit LCD panel?

lol37
Posts: 59
Joined: 24 Dec 2013, 15:29

Re: Challenges in building a 10 bit display (or higher)

Post by lol37 » 22 Jan 2014, 03:13

Hello there,
what LCD technology are you being refer to ?
IPS and (at a least extent VA) are relatively easy to make 10-bits precision
TN can achieve 6-bit precision and up to 8-bit with dithering or some other hack, but by their principe, TN has NOT to be that so precise, they are fast by conception.
by the way why are you considering such a mod ? it is highly physic but also electronic related
over 8 bit are somewhat useless on a sRGB space, there's no posterization effect on that gamut space, and nearly all wide-gamut monitor have at least 10 bit processor + pixel capable..

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Challenges in building a 10 bit display (or higher)

Post by Chief Blur Buster » 22 Jan 2014, 03:54

cirthix might have something interesting to say about this matter.
As might a few others.

I think that TN panels are somewhat limited by the noise factor.
The faster pixel response may be playing a factor too.

Traditionally, the way LCD works is that voltage is sent to LCD pixels, and the electronics has already moved on to refreshing the next row of LCD pixels. But the pixels keeps moving on their own momentum long after you've refreshed the pixel, and the pixel gets closer and closer to its final value. Kind of like curling or archery. Metaphorically, the drive voltage is the curler or the archer imparting aim/momentum on the object that needs to get closer to its destination. LCD pixels are kind of like that. You pulse a voltage and hope the pixel coasts correctly to its final color value, long after you've pulsed that pixel and moved onto the next LCD pixel. Hopefully you aimed correctly (sent the correct voltage).

There are more subtle, advanced driving behaviors involved in different LCDs that may use voltage duration (spend longer time sending voltage into some pixels, and sending less time sending voltage to other pixels), chasing-scan algorithms (a voltage impulse to get the pixel accelerated, and a few scan lines behind in the scanning pass (a millisecond later), a voltage impulse to get the pixel decelerated). I suspect that many good 3D-panel-ready response-time acceleration algorithms use an accelerate(microsecond-league) followed by a decelerate voltage(microsecond-league). The panel manufacturers have gotten pretty creative and seem to be doing some crazy things to get LCD pixels to move faster, to its final color, than we previously thought possible.

There are transistors controlling each pixel, which helps keep a sustained voltage on the pixel long after it's refreshed, keeping it stable, and are higher contrast and better picture quality than panels that don't use transistors behind each pixel (e.g. passive matrix LCDs which aren't used anymore in computer monitors; even modern TN monitors are now all active matrix)

You can imagine there are voltage inaccuracies that build up because you're sending voltage for a tiny fraction of a millisecond (possibly less than 1 microsecond if pixel-at-a-time driven, and several tens of microsecond if row-of-pixels-at-a-time driven, more if multiple-row driven). The faster you try to move pixels, the more inaccuracies can occur (noise), limiting color depth. Ultra-accurate voltages for ultra-accurate time periods, to shoot LCD pixels to its correct value. That said, I believe TN quality can still improve with more bits, if the driving accuracy is available (which would increase cost, as TN is often the cheapest panel technology available, so traditionally one don't really want to spend funds pimping-out a Yugo with fancy addons, versus pimping-out a Ferrari)

I imagine there's nothing stopping TN from gaining direct-driven 10-bit accuracy eventually, but might come at a compromise (e.g. expense, response speed, little noticeable benefit due to noise, etc).
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
nimbulan
Posts: 323
Joined: 29 Dec 2013, 23:32
Location: Oregon

Re: Challenges in building a 10 bit display (or higher)

Post by nimbulan » 22 Jan 2014, 04:07

Chief Blur Buster wrote:That said, I believe TN quality can still improve with more bits, if the driving accuracy is available (which would increase cost, as TN is often the cheapest panel technology available, so traditionally one don't really want to spend funds pimping-out a Yugo with fancy addons, versus pimping-out a Ferrari)

I imagine there's nothing stopping TN from gaining direct-driven 10-bit accuracy eventually, but might come at a compromise (e.g. expense, response speed, little noticeable benefit due to noise, etc).
I have no doubt that this is true, but if cost is the main concern, why are $300+ gaming monitors still 6 bit? It is a market segment that would greatly benefit from improved color due to persistent color banding in water, smoke, fog, sky, and other gradients in video games on 6 bit panels. Even if improved color processing increases the response time, it would still be faster than any VA or IPS panel.

lol37
Posts: 59
Joined: 24 Dec 2013, 15:29

Re: Challenges in building a 10 bit display (or higher)

Post by lol37 » 22 Jan 2014, 10:28

nimbulan wrote:
Chief Blur Buster wrote:That said, I believe TN quality can still improve with more bits, if the driving accuracy is available (which would increase cost, as TN is often the cheapest panel technology available, so traditionally one don't really want to spend funds pimping-out a Yugo with fancy addons, versus pimping-out a Ferrari)

I imagine there's nothing stopping TN from gaining direct-driven 10-bit accuracy eventually, but might come at a compromise (e.g. expense, response speed, little noticeable benefit due to noise, etc).
I have no doubt that this is true, but if cost is the main concern, why are $300+ gaming monitors still 6 bit? It is a market segment that would greatly benefit from improved color due to persistent color banding in water, smoke, fog, sky, and other gradients in video games on 6 bit panels. Even if improved color processing increases the response time, it would still be faster than any VA or IPS panel.
well you are wrong : the eizo FORIS FG2421 can push up to 16,7 milions of colors and it is one of the most expensive gaming monitor
but yeah it is a VA panel
http://www.eizo.com/global/products/for ... html#tab02

spacediver
Posts: 505
Joined: 18 Dec 2013, 23:51

Re: Challenges in building a 10 bit display (or higher)

Post by spacediver » 22 Jan 2014, 13:18

Mark, thanks for the fantastic insights. That whole idea about momentum never occurred to me. It certainly illustrates some of the deep challenges in creating quality LCD panels.

The Vpixx we have is 12 bits, and I believe is a TN panel with excellent response time. I'll probably be going to VSS this may, and I'll definitely ask Peter how he achieved this.

lol37 I'm not interested in making a mod, I'm interested in learning about the engineering challenges involved in designing and manufacturing these displays.

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Challenges in building a 10 bit display (or higher)

Post by Chief Blur Buster » 22 Jan 2014, 13:45

spacediver wrote:The Vpixx we have is 12 bits, and I believe is a TN panel with excellent response time. I'll probably be going to VSS this may, and I'll definitely ask Peter how he achieved this.
For those not aware, this is the Viewpixx Scientific LCD, which comes in both IPS and TN versions.
A 120Hz, 1920x1080/1920x1200 LCD with 12-bit RGB, scanning backlight, full gamut RGB LED, and costs five figures, filling a special niche. Vpixx also sells a 500Hz capable projector.

Yes, this is clearly an example of a 12-bit TN LCD, which shows it's not impossible.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Challenges in building a 10 bit display (or higher)

Post by Chief Blur Buster » 22 Jan 2014, 13:47

nimbulan wrote:I have no doubt that this is true, but if cost is the main concern, why are $300+ gaming monitors still 6 bit?
$300 is cheap.
I remember paying $1249 for my Samsung Syncmaster 17GLSi CRT computer monitor around the year 1995. And early flat panel LCD computer monitors used to cost >$1000 too, for something that were massive ghost monsters (>33ms).

Eventually, we will get our 8-bit or even 10-bit TN LCDs. Monitors such as the ASUS ROG PG278 (first officially LightBoost/ULMB/3DVision 1440p) is an 8-bit TN LCD, and it better'd be, considering it's a $799 1440p TN LCD reportedly to be much better quality. DisplayLag also confirms it's an 8-bit TN.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
nimbulan
Posts: 323
Joined: 29 Dec 2013, 23:32
Location: Oregon

Re: Challenges in building a 10 bit display (or higher)

Post by nimbulan » 22 Jan 2014, 14:22

Chief Blur Buster wrote:Eventually, we will get our 8-bit or even 10-bit TN LCDs. Monitors such as the ASUS ROG PG278 (first officially LightBoost/ULMB/3DVision 1440p) is an 8-bit TN LCD, and it better'd be, considering it's a $799 1440p TN LCD reportedly to be much better quality. DisplayLag also confirms it's an 8-bit TN.
Now that is good news. Hopefully that becomes the norm for gaming monitors in the next couple years.

Post Reply