Page 1 of 1

thoughts on native 8bit TN quality?

Posted: 27 May 2014, 16:50
by Edmond
The ROG Swift is the only monitor that so far is known to use an native 8bit TN panel, right? its better than every other TN out there, right? in viewing angles and what not.

Gsync needs fast response times, so i guess TN only. Until Oled.
Im sure you could put gsync on an IPS, but it will probably introduce new artifacts or smth.

Do you think anyone would ever consider making a 21:9 monitor out of that 8bit TN? Im just trying to grasp if a 120hz 21:9 is something that could be considered.

Re: thoughts on native 8bit TN quality?

Posted: 27 May 2014, 17:55
by RealNC
G-Sync does not depend on fast response times. It's totally irrelevant, in fact :-)

Even the slowest IPS panel could be driven by G-Sync.

Re: thoughts on native 8bit TN quality?

Posted: 27 May 2014, 18:02
by Chief Blur Buster
Edmond wrote:The ROG Swift is the only monitor that so far is known to use an native 8bit TN panel, right?
Not necessarily -- there's the Viewpixx Scientific LCD (vpixx.com), which comes in both IPS and TN versions, and it's a 10-bit TN. But as far as I know, this is probably the first time consumer TN has reached 8-bit levels at high-def resolutions.
Edmond wrote:Gsync needs fast response times, so i guess TN only.
Not necessarily -- GSYNC could work on any LCD that can be adapted to refresh at a variable rate, even heoretically a slow 33ms LCD. Overlord Computer is wanting to bring out an IPS GSYNC monitor at some point in the future (Essentially a GSYNC version of Tempest X270OC). No ETA yet, there are a lot of monitors that will come out first.

Remember, response time has nothing to do with refresh rate.
Response time can be slower than a refresh. e.g. 33ms at 60Hz. The refreshes simply ghosts into each other.
Response time can be faster than a refresh. e.g. 1ms at 120Hz.

Response is equivalent to inertia -- pixels coasting from an old grey value to a new grey value, GtG.
Refresh is equivalent to momentum -- injecting voltage into pixels to get them coasting to new grey value.

Multiple refresh cycles can occur even long before the pixel has coasted to its new value (the slower the pixel is, the more ghosting/streaking artifacts when pixels are still not arrived to its final color value). LCD scanout goes one row of pixels at a time, and long after that scanout has been done, GtG is still occuring by itself due to sheer momentum within the physical liquid crystals sandwiched between glass (crystals realigning to new positions).

However, motion-blur-reducing strobe backlights (e.g. LightBoost) do require fast response times, since good-quality strobing requires practically finishing current pixel response/transitions for the current refresh cycle, before the next refresh cycle begins.
Edmond wrote:Im sure you could put gsync on an IPS, but it will probably introduce new artifacts or smth.
Not if you don't overclock. For example, GSYNC 60Hz should be relatively artifact free on 60Hz IPS panels. Also, slower response also tends to mask LCD inversion artifacts that can become noticeable on TN GSYNC monitors at low refresh rates.
Edmond wrote:Do you think anyone would ever consider making a 21:9 monitor out of that 8bit TN? Im just trying to grasp if a 120hz 21:9 is something that could be considered.
At the moment, I think it likely would be easier to add artifact-free 60Hz GSYNC to a 60Hz 21:9 monitor, than to drive a 21:9 panel reliably artifact-free at 120Hz.

In the nearer term, to get GSYNC-style motion fluidity with 21:9 may be easier at the moment to buy an overkill of GPU's (e.g. SLI/Titans) and/or adjust detail levels to ensure full 60fps@60Hz VSYNC ON operation, and attempting to input lag via an in-game-engine framerate cap such as the one found in Source Engine, like fps_max 59 (1 microstutter per second) or fps_max 59.8 (1 microstutter every 5 seconds). But this can be more finicky than just simply having a GSYNC monitor.

Re: thoughts on native 8bit TN quality?

Posted: 28 May 2014, 02:53
by glenster
"The ROG Swift is the only monitor that so far is known to use an native 8bit TN
panel, right?"

I just saw this: an Asus PB287Q 3840x2160 10 bit TN review:
http://www.tomshardware.com/reviews/asu ... ,3832.html

Re: thoughts on native 8bit TN quality?

Posted: 28 May 2014, 21:35
by spacediver
Chief Blur Buster wrote: Not necessarily -- there's the Viewpixx Scientific LCD (vpixx.com), which comes in both IPS and TN versions, and it's a 10-bit TN.
I spoke with Peter April about this last week at the VSS conference. He did indeed confirm your view that the challenge in increasing bit depth on these panels is the viscosity and volatility of these molecules. This is why their IPS panels can go to 12 bits.

Re: thoughts on native 8bit TN quality?

Posted: 28 May 2014, 23:47
by flood
for me, the main problem with TN displays is the vertical viewing angles.

Re: thoughts on native 8bit TN quality?

Posted: 29 May 2014, 03:19
by Edmond
flood wrote:for me, the main problem with TN displays is the vertical viewing angles.
Really hope Sony`s 4k OLED 27" is gona spark some oled competition in monitors.
I see OLED the only way we are going to get viewing angles, color and speed at the same time. IN OUR LIFETIME.