Manufacturers: We Need User-Programmable Overdrive Lookup Tables

Advanced display talk, display hackers, advanced game programmers, scientists, display researchers, display manufacturers, vision researchers & Advanced Display Articles on Blur Busters. The masters on Blur Busters.
Post Reply
User avatar
Chief Blur Buster
Site Admin
Posts: 11653
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Manufacturers: We Need User-Programmable Overdrive Lookup Tables

Post by Chief Blur Buster » 29 Apr 2020, 01:48

We Need Customizable Overdrive Lookup Tables

Attention: Monitor Manufacturers / Monitor Firmware Vendors / Scaler Vendors / TCON Vendors / Etc.

We can better and tune improve motion quality on gaming monitors with better access to custom overdrive.

We know some manufacturers already lurk the Area 51 forum. As you know, we wrote the Pixel Response FAQ: GtG versus MPRT.

After writing the Technical Overdrive Rabbit Hole, I have been thinking.

We need monitor to support user-rewriteable overdrive lookup tables of the full size. Preferably full 256x256, not skimpy 9x9 or 17x17 overdrive lookup tables. We can accept that OD LUTs can be transmitted over DDC commands (VCP codes), or via USB cable (to monitor's USB port) to be saved into a series "user-defined OD LUT slots". Conventional DDC VCP commands can then be used to switch between monitor's standard overdrive, and these custom OD LUTs.

We find this is very important to have full access to the full 256x256 array because improves quality at colder temperatures and warmer temperatures. Most scientists and engineers generate an OD LUT at room temperature, but the limitations of interpolated OD LUTs start to emerge in unusual situations.

Customers of monitors in the artic in winter, and tropics in summer, sometimes complain about overdrive problems for temperatures far away from the room temperature – since LCD is sensitive to temperature changes. And different refresh rates often requires different OD LUTs, as does strobed modes.

The formula is A(B) = C whereas:

A = original subpixel greyscale value
B = destination subpixel greyscale value
C = intentional overshoot greyscale value to accelerate pixel transition from A to B

The OD LUT is executed 3 times per pixel every refresh cycle, for each subpixel channel (red, green, blue), as if they were 3 completely independent pixels. This LUT math runs every pixel every refresh cycle.

The TCON/scaler is already executing these maths, we just want user-defined OD LUTs for monitor firmwares. Often the OD LUT is permanently saved into the monitor in the factory, and not modifiable. We've even seen manufacturers recycle the wrong OD LUTs on a different panel, damaging the quality of overdrive.

Blur Busters wants to totally change the unmodifiable overdrive lookup table, because Blur Busters can generate better OD LUTs than most manufacturers (We offer this service; services.blurbusters.com).

For a 256x256 LUT, that’s 65,536 values (minus the 256 no-change values), consuming 64 kilobytes of TCON/scaler memory. We’d like to see multiple user-defineable OD LUT slots that can be recalled by a single DDC command or attached to a specific profile (e.g. specific strobed refresh rate).

Overdrive Gain (all 100 levels) should also still function as a multiplier factor on these custom LUTs, with a centered OD gain executing the LUT unmodified. The Overdrive Gain needs to be 100% accessible in user menus and/or via DDC command, given that overdrive artifacts can appear at colder and warmer temperatures -- not all homes are the same temperature as a display engineering laboratory. Offering only 2 or 4 or 10 levels of overdrive is not enough, we need access to all 100 levels.

It is now easy to adjust Overdrive Gain, users can just view www.testufo.com/ghosting while adjusting an Overdrive gain slider (similiar to a Brightness slider or Contrast slider).

Optional / Harder Bonus: For strobed modes, I’d love to see programmable OD Gain zones along the Y dimension of the panel, since that helps reduce strobe crosstalk at the very edges of the panel where the time differential between scanout and strobing is tighter – an OD Gain gradient (low at top edge, high setting at bottom edge) along the vertical dimension of the screen, perhaps configurable using 3 settings (Top OD Gain, Center OD Gain, Bottom OD Gain) with a linear interpolated gradient along the vertical dimension (pixel row number). This can improve strobing near max Hz when there’s not enough headroom to cram GtG in VBI.

We already have equipment that can generate custom overdrive lookup tables, including those specifically optimized for strobe-backlight modes and 3D-glasses mode.
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!

Post Reply