Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

frisbfreek

macrumors member
Nov 11, 2020
33
12
So what’s the current status?

M1 Pro:
Can do 4K144 via USB-C to DisplayPort
Can’t do 4k120 via USB-C to HDMI

Is that right?

Also, not sure if this addresses any of the issues identified in this thread, but have people tried BetterDummy?

 

Lazylex

macrumors newbie
Oct 28, 2021
18
23
Arizona
So what’s the current status?

M1 Pro:
Can do 4K144 via USB-C to DisplayPort
Can’t do 4k120 via USB-C to HDMI

Is that right?

Also, not sure if this addresses any of the issues identified in this thread, but have people tried BetterDummy?

Correct, HDMI is limited, even with a DP 1.4 -> HDMI 2.1 adapter.

Unfortunately no luck on BetterDummy either -- OS is still limiting to 60hz, https://github.com/waydabber/BetterDummy/issues/3
 

frisbfreek

macrumors member
Nov 11, 2020
33
12
Correct, HDMI is limited, even with a DP 1.4 -> HDMI 2.1 adapter.

Unfortunately no luck on BetterDummy either -- OS is still limiting to 60hz, https://github.com/waydabber/BetterDummy/issues/3
I see 8K resolutions (and beyond) @ 60hz in BetterDummy. Does that suggest that the M1 Pro is capable of outputting 8K? I might need to get a USB-C to HDMI 2.1 adapter and go to Best Buy to try it out on an 8K TV!
 

Artagra

macrumors regular
Sep 6, 2007
122
42
I see 8K resolutions (and beyond) @ 60hz in BetterDummy. Does that suggest that the M1 Pro is capable of outputting 8K? I might need to get a USB-C to HDMI 2.1 adapter and go to Best Buy to try it out on an 8K TV!
It can do 8k, only at 30hz. If we solve the 4k120hz issue that would also be a good sign for 8k60.
 
  • Like
Reactions: frisbfreek

joevt

macrumors 604
Jun 21, 2012
6,963
4,257
I see 8K resolutions (and beyond) @ 60hz in BetterDummy. Does that suggest that the M1 Pro is capable of outputting 8K? I might need to get a USB-C to HDMI 2.1 adapter and go to Best Buy to try it out on an 8K TV!
It means the M1 Pro can scale 8K down to whatever the display expects.
Actually outputting 8K60 is a different matter.
 

kasakka

macrumors 68020
Oct 25, 2008
2,389
1,073
Yes I was talking about ASBL. Not really using high enough OLED Light for ABL to cause any noticeable difference. ASBL was definitely noticeable though in desktop use.
Yeah I found ASBL annoying as well. ASBL = automatic static brightness limiter, dims the desktop because the TV thinks it is seeing static content.

I disabled it on mine from the service menu using an old phone with an IR blaster and the LG service menu Android app. Over a year of use later and no burn in. Might void your warranty though.
 

kasakka

macrumors 68020
Oct 25, 2008
2,389
1,073
It means the M1 Pro can scale 8K down to whatever the display expects.
Actually outputting 8K60 is a different matter.
Exactly. With a 4K screen connected and set to less than native scaling, what MacOS does is it first renders the image at double the scaled resolution and then downscales it from there to the native resolution of the display. This apparently results in better quality.

For actual max resolution Apple is always confusing and vague. They state that you could have 6K @ 60 Hz with 10-bit color but does that translate to 8K @ 60 Hz with 8-bit color or not?
 
  • Like
Reactions: frisbfreek

Artagra

macrumors regular
Sep 6, 2007
122
42
It can do 8k, only at 30hz. If we solve the 4k120hz issue that would also be a good sign for 8k60.
I was probably mistaken here - it might be able to do 8k 30, but no guarantees. Would need to test.

With the new adapters (if we could get them working), it should in theory support 8k60 but some OEMs of them have been stating they only support 8k30, as they haven't been seeing 8k60 consistently in windows.
 
  • Like
Reactions: frisbfreek

Artagra

macrumors regular
Sep 6, 2007
122
42
Comparing the two EDIDs, I've excerpted the sections that seem to be material. The first 4k120 mention for mine (HDMI2.1 not working) does not have a specific Pixel Clock, and where it does have a pixel clock it's 594mhz. The first 4k120 mention for the DP1.4 working display does have a pixel clock, and it's 1328.25mhz.

People who know more about EDID, could this be related to the problem? Would an EDID override (currently not possible on M1 macs) solve this problem? Could an EDID override dongle allow it, or firmware from an adapter vendor that includes a higher pixel clock (or removes the lower pixel clock)?


My EDID (DP 1.4 to HDMI 2.1 adapter into LG CX - HDMI 2.1 display - only working at 4k60, should support 4k120):

-----------------------------------------------------
------------ EXTENSION EDID BLOCK 1 ---------------
-----------------------------------------------------

CTA-EXT: CTA 861 Series Extension:
----------------------------------
Revision:...........................3
First DTD block at offset...........84
Display Supports:
Underscan on PC modes
Basic audio
YCbCr 4:4:4
YCbCr 4:2:2

Data Block #1
Video Type: standard CTA Timings
VIC # 0: ( 97) 3840 x 2160 @ 60Hz - 16:9
VIC # 1: ( 96) 3840 x 2160 @ 50Hz - 16:9
VIC # 2: (118) 3840 x 2160 @ 120Hz - 16:9
VIC # 3: (117) 3840 x 2160 @ 100Hz - 16:9
VIC # 4: (102) 4096 x 2160 @ 60Hz - 256:135
VIC # 5: (101) 4096 x 2160 @ 50Hz - 256:135
VIC # 6: (219) 4096 x 2160 @ 120Hz - 256:135
VIC # 7: (218) 4096 x 2160 @ 100Hz - 256:135
VIC # 8: ( 16) 1920 x 1080 @ 60Hz - 16:9
VIC # 9: ( 31) 1920 x 1080 @ 50Hz - 16:9
VIC #10: ( 4) 1280 x 720 @ 60Hz - 16:9
VIC #11: ( 19) 1280 x 720 @ 50Hz - 16:9
VIC #12: ( 5) 1920 x 1080i @ 60Hz - 16:9
VIC #13: ( 20) 1920 x 1080i @ 50Hz - 16:9
VIC #14: ( 3) 720 x 480 @ 60Hz - 16:9
VIC #15: ( 2) 720 x 480 @ 60Hz - 4:3
VIC #16: ( 18) 720 x 576 @ 50Hz - 16:9
VIC #17: ( 32) 1920 x 1080 @ 24Hz - 16:9
VIC #18: ( 33) 1920 x 1080 @ 25Hz - 16:9
VIC #19: ( 34) 1920 x 1080 @ 30Hz - 16:9
VIC #20: ( 21) 1440 x 576i @ 50Hz - 4:3
VIC #21: ( 1) 640 x 480 @ 60Hz - 4:3
VIC #22: ( 93) 3840 x 2160 @ 24Hz - 16:9
VIC #23: ( 94) 3840 x 2160 @ 25Hz - 16:9
VIC #24: ( 95) 3840 x 2160 @ 30Hz - 16:9
VIC #25: ( 98) 4096 x 2160 @ 24Hz - 256:135
VIC #26: ( 99) 4096 x 2160 @ 25Hz - 256:135
VIC #27: (100) 4096 x 2160 @ 30Hz - 256:135
VIC #28: ( 63) 1920 x 1080 @ 120Hz - 16:9
VIC #29: ( 64) 1920 x 1080 @ 100Hz - 16:9


LazyLex EDID (Thanks again! USB C DP-Alt Mode to DP1.4 Monitor - working at 4k120)

-----------------------------------------------------
------------ EXTENSION EDID BLOCK 2 ---------------
-----------------------------------------------------

DID-EXT: Display ID Extension:
------------------------------
DisplayID version........1.2
Display Product Type: Standalone display device

Detailed Timings Block type 1
Detailed Timing #1 in Block
Mode = 3840 x 2160 @ 143.999Hz
Pixel Clock............. 1328.25 MHz Not Interlaced

Horizontal Vertical
Active.................. 3840 pixels 2160 lines
Front Porch............. 48 pixels 3 lines
Sync Width.............. 32 pixels 5 lines
Back Porch.............. 80 pixels 138 lines
Blanking................ 160 pixels 146 lines
Total................... 4000 pixels 2306 lines
Scan Rate............... 332.062 kHz 143.999 Hz

Is default Mode
3D stereo: no
Sync: Digital separate with
* Negative vertical polarity
* Positive horizontal polarity

Detailed Timing #2 in Block
Mode = 3840 x 2160 @ 94.981Hz
Pixel Clock............. 858.25 MHz Not Interlaced

Horizontal Vertical
Active.................. 3840 pixels 2160 lines
Front Porch............. 48 pixels 3 lines
Sync Width.............. 32 pixels 5 lines
Back Porch.............. 80 pixels 91 lines
Blanking................ 160 pixels 99 lines
Total................... 4000 pixels 2259 lines
Scan Rate............... 214.562 kHz 94.981 Hz

Is not default Mode
3D stereo: no
Sync: Digital separate with
* Negative vertical polarity
* Positive horizontal polarity
[/ISPOILER]
 
Last edited:

Lazylex

macrumors newbie
Oct 28, 2021
18
23
Arizona
Here is a LG C1 connected via USB-C to HDMI 2.1 Adapter (Club3D CAC-1586)

EDID report generated by SwitchResX Version 4.11.2 (Apple Silicon) (Build 2101) for display LG TV SSCR


-----------------------------------------------------
------------------- RAW DATA ------------------------
-----------------------------------------------------
00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F
-----------------------------------------------------
0 | 00 FF FF FF FF FF FF 00 1E 6D C8 C0 01 01 01 01
1 | 01 1F 01 03 80 A0 5A 78 0A EE 91 A3 54 4C 99 26
2 | 0F 50 54 A1 08 00 31 40 45 40 61 40 71 40 81 80
3 | D1 C0 01 01 01 01 08 E8 00 30 F2 70 5A 80 B0 58
4 | 8A 00 40 84 63 00 00 1E 6F C2 00 A0 A0 A0 55 50
5 | 30 20 35 00 40 84 63 00 00 1E 00 00 00 FD 00 18
6 | 78 1E FF 77 00 0A 20 20 20 20 20 20 00 00 00 FC
7 | 00 4C 47 20 54 56 20 53 53 43 52 0A 20 20 01 77

-----------------------------------------------------
00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F
-----------------------------------------------------
0 | 02 03 62 F1 5E 61 60 76 75 66 65 DB DA 10 1F 04
1 | 13 05 14 03 02 12 20 21 22 15 01 5D 5E 5F 62 63
2 | 64 3F 40 2C 09 57 07 15 07 50 57 07 01 67 04 03
3 | 6E 03 0C 00 20 00 B8 3C 2C 00 80 01 02 03 04 68
4 | D8 5D C4 01 78 80 5B 02 E2 00 CF E3 05 C0 00 E3
5 | 06 0D 01 E2 0F FF EB 01 46 D0 00 48 03 76 82 5E
6 | 6D 95 00 00 00 00 00 00 00 00 00 00 00 00 00 00
7 | 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 BE

-----------------------------------------------------
< 00FFFFFF FFFFFF00 1E6DC8C0 01010101 011F0103 80A05A78 0AEE91A3 544C9926 0F5054A1 08003140 45406140 71408180 D1C00101 010108E8 0030F270 5A80B058 8A004084 6300001E 6FC200A0 A0A05550 30203500 40846300 001E0000 00FD0018 781EFF77 000A2020 20202020 000000FC 004C4720 54562053 5343520A 20200177 020362F1 5E616076 756665DB DA101F04 13051403 02122021 2215015D 5E5F6263 643F402C 09570715 07505707 01670403 6E030C00 2000B83C 2C008001 02030468 D85DC401 78805B02 E200CFE3 05C000E3 060D01E2 0FFFEB01 46D00048 0376825E 6D950000 00000000 00000000 00000000 00000000 00000000 00000000 000000BE >

-----------------------------------------------------
{ 0x00, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x00, 0x1E, 0x6D, 0xC8, 0xC0, 0x01, 0x01, 0x01, 0x01, 0x01, 0x1F, 0x01, 0x03, 0x80, 0xA0, 0x5A, 0x78, 0x0A, 0xEE, 0x91, 0xA3, 0x54, 0x4C, 0x99, 0x26, 0x0F, 0x50, 0x54, 0xA1, 0x08, 0x00, 0x31, 0x40, 0x45, 0x40, 0x61, 0x40, 0x71, 0x40, 0x81, 0x80, 0xD1, 0xC0, 0x01, 0x01, 0x01, 0x01, 0x08, 0xE8, 0x00, 0x30, 0xF2, 0x70, 0x5A, 0x80, 0xB0, 0x58, 0x8A, 0x00, 0x40, 0x84, 0x63, 0x00, 0x00, 0x1E, 0x6F, 0xC2, 0x00, 0xA0, 0xA0, 0xA0, 0x55, 0x50, 0x30, 0x20, 0x35, 0x00, 0x40, 0x84, 0x63, 0x00, 0x00, 0x1E, 0x00, 0x00, 0x00, 0xFD, 0x00, 0x18, 0x78, 0x1E, 0xFF, 0x77, 0x00, 0x0A, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x00, 0x00, 0x00, 0xFC, 0x00, 0x4C, 0x47, 0x20, 0x54, 0x56, 0x20, 0x53, 0x53, 0x43, 0x52, 0x0A, 0x20, 0x20, 0x01, 0x77, 0x02, 0x03, 0x62, 0xF1, 0x5E, 0x61, 0x60, 0x76, 0x75, 0x66, 0x65, 0xDB, 0xDA, 0x10, 0x1F, 0x04, 0x13, 0x05, 0x14, 0x03, 0x02, 0x12, 0x20, 0x21, 0x22, 0x15, 0x01, 0x5D, 0x5E, 0x5F, 0x62, 0x63, 0x64, 0x3F, 0x40, 0x2C, 0x09, 0x57, 0x07, 0x15, 0x07, 0x50, 0x57, 0x07, 0x01, 0x67, 0x04, 0x03, 0x6E, 0x03, 0x0C, 0x00, 0x20, 0x00, 0xB8, 0x3C, 0x2C, 0x00, 0x80, 0x01, 0x02, 0x03, 0x04, 0x68, 0xD8, 0x5D, 0xC4, 0x01, 0x78, 0x80, 0x5B, 0x02, 0xE2, 0x00, 0xCF, 0xE3, 0x05, 0xC0, 0x00, 0xE3, 0x06, 0x0D, 0x01, 0xE2, 0x0F, 0xFF, 0xEB, 0x01, 0x46, 0xD0, 0x00, 0x48, 0x03, 0x76, 0x82, 0x5E, 0x6D, 0x95, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xBE }

-----------------------------------------------------
Valid EDID block #0: checksum ok
Valid EDID block #1: checksum ok

-----------------------------------------------------
------------------- MAIN EDID BLOCK -----------------
-----------------------------------------------------

EDID Version........1.3
Manufacturer........GSM (1E6D)
Product ID..........49352 (C0C8)
Serial Number.......01010101

Manufactured........Week 1 of year 2021
Max H Size..........160 cm
Max V Size..........90 cm
Gamma...............2.20

Display Supported Features:
---------------------------


Display type:
-------------
RGB color display
Display is non continuous frequency
Default color space is not sRGB standard
First detailed Timing is preferred Timing


Input signal & sync:
--------------------
Digital Input


Color info:
-----------
Red x = 0.640 Green x = 0.300 Blue x = 0.150 White x = 0.312
Red y = 0.330 Green y = 0.600 Blue y = 0.060 White y = 0.329

Established Timings supported:
------------------------------
# 0: 720 x 400 @ 70Hz
# 1: 640 x 480 @ 60Hz
# 2: 800 x 600 @ 60Hz
# 3: 1024 x 768 @ 60Hz

Standard Timing supported:
--------------------------
# 0: 640 x 480 @ 60Hz (3140)
# 1: 800 x 600 @ 60Hz (4540)
# 2: 1024 x 768 @ 60Hz (6140)
# 3: 1152 x 864 @ 60Hz (7140)
# 4: 1280 x 1024 @ 60Hz (8180)
# 5: 1920 x 1080 @ 60Hz (D1C0)

Monitor Description blocks:
---------------------------
Descriptor #0 - Timing definition:
Mode = 3840 x 2160 @ 60.000Hz
Pixel Clock............. 594.00 MHz Not Interlaced

Horizontal Vertical
Active.................. 3840 pixels 2160 lines
Front Porch............. 176 pixels 8 lines
Sync Width.............. 88 pixels 10 lines
Back Porch.............. 296 pixels 72 lines
Blanking................ 560 pixels 90 lines
Total................... 4400 pixels 2250 lines
Scan Rate............... 135.000 kHz 60.000 Hz
Image Size.............. 1600 mm 900 mm
Border.................. 0 pixels 0 lines

Sync: Digital separate with
* Positive vertical polarity
* Positive horizontal polarity

Descriptor #1 - Timing definition:
Mode = 2560 x 1440 @ 119.998Hz
Pixel Clock............. 497.75 MHz Not Interlaced

Horizontal Vertical
Active.................. 2560 pixels 1440 lines
Front Porch............. 48 pixels 3 lines
Sync Width.............. 32 pixels 5 lines
Back Porch.............. 80 pixels 77 lines
Blanking................ 160 pixels 85 lines
Total................... 2720 pixels 1525 lines
Scan Rate............... 182.996 kHz 119.998 Hz
Image Size.............. 1600 mm 900 mm
Border.................. 0 pixels 0 lines

Sync: Digital separate with
* Positive vertical polarity
* Positive horizontal polarity

Descriptor #2 - Monitor range:
Horizontal frequency range.......30-255 kHz
Vertical frequency range.........24-120 Hz
Maximum bandwidth range..........1190 MHz
GTF supported

Descriptor #3 - Monitor name:
LG TV SSCR


-----------------------------------------------------
------------ EXTENSION EDID BLOCK 1 ---------------
-----------------------------------------------------

CTA-EXT: CTA 861 Series Extension:
----------------------------------
Revision:...........................3
First DTD block at offset...........94
Display Supports:
Underscan on PC modes
Basic audio
YCbCr 4:4:4
YCbCr 4:2:2

Data Block #1
Video Type: standard CTA Timings
VIC # 0: ( 97) 3840 x 2160 @ 60Hz - 16:9
VIC # 1: ( 96) 3840 x 2160 @ 50Hz - 16:9
VIC # 2: (118) 3840 x 2160 @ 120Hz - 16:9
VIC # 3: (117) 3840 x 2160 @ 100Hz - 16:9
VIC # 4: (102) 4096 x 2160 @ 60Hz - 256:135
VIC # 5: (101) 4096 x 2160 @ 50Hz - 256:135
VIC # 6: (219) 4096 x 2160 @ 120Hz - 256:135
VIC # 7: (218) 4096 x 2160 @ 100Hz - 256:135
VIC # 8: ( 16) 1920 x 1080 @ 60Hz - 16:9
VIC # 9: ( 31) 1920 x 1080 @ 50Hz - 16:9
VIC #10: ( 4) 1280 x 720 @ 60Hz - 16:9
VIC #11: ( 19) 1280 x 720 @ 50Hz - 16:9
VIC #12: ( 5) 1920 x 1080i @ 60Hz - 16:9
VIC #13: ( 20) 1920 x 1080i @ 50Hz - 16:9
VIC #14: ( 3) 720 x 480 @ 60Hz - 16:9
VIC #15: ( 2) 720 x 480 @ 60Hz - 4:3
VIC #16: ( 18) 720 x 576 @ 50Hz - 16:9
VIC #17: ( 32) 1920 x 1080 @ 24Hz - 16:9
VIC #18: ( 33) 1920 x 1080 @ 25Hz - 16:9
VIC #19: ( 34) 1920 x 1080 @ 30Hz - 16:9
VIC #20: ( 21) 1440 x 576i @ 50Hz - 4:3
VIC #21: ( 1) 640 x 480 @ 60Hz - 4:3
VIC #22: ( 93) 3840 x 2160 @ 24Hz - 16:9
VIC #23: ( 94) 3840 x 2160 @ 25Hz - 16:9
VIC #24: ( 95) 3840 x 2160 @ 30Hz - 16:9
VIC #25: ( 98) 4096 x 2160 @ 24Hz - 256:135
VIC #26: ( 99) 4096 x 2160 @ 25Hz - 256:135
VIC #27: (100) 4096 x 2160 @ 30Hz - 256:135
VIC #28: ( 63) 1920 x 1080 @ 120Hz - 16:9
VIC #29: ( 64) 1920 x 1080 @ 100Hz - 16:9

Data Block #2
Audio Data Block
Supported format: Linear PCM, on 2 channels
Supported sample rates (kHz): 32kHz, 44.1kHz, 48kHz, 96kHz, 192kHz
Supported sample sizes (bits): 16 bit, 20 bit, 24 bit

Supported format: AC-3, on 6 channels
Supported sample rates (kHz): 32kHz, 44.1kHz, 48kHz
Maximum bitrate: 640 kb/s

Supported format: Dolby Digital+, on 8 channels
Supported sample rates (kHz): 32kHz, 44.1kHz, 48kHz
Supports Joint Object Coding

Supported format: MAT (MLP), on 8 channels
Supported sample rates (kHz): 48kHz


Data Block #3
Vendor specific Data Block - Vendor Identifier: 00:0C:03 (HDMI)
Found HDMI IEEE Registration Identifier
CEC physical address........... 2.0.0.0
Supports AI (ACP, ISRC)........ Yes
Supports 48bpp color depth..... No
Supports 36bpp color depth..... Yes
Supports 30bpp color depth..... Yes
Supports YCbCr 4:4:4........... Yes
Supports dual-link DVI......... No
Maximum TMDS clock............. 300 MHz
Supported Content Types:
Cinema
Game
HDMI video capabilities........ Yes
Supports 3D.................... No
HDMI 1.4a specifics............ present
HDMI VIC # 1:................. 3840 x 2160 @ 30Hz
HDMI VIC # 2:................. 3840 x 2160 @ 25Hz
HDMI VIC # 3:................. 3840 x 2160 @ 24Hz
HDMI VIC # 4:................. 4096 x 2160 @ 24Hz

Data Block #4
Vendor specific Data Block - Vendor Identifier: C4:5D:D8 (HDMI Forum)
(HDMI Forum)
Version: 1
Maximum TMDS Character Rate: 600MHz
SCDC Present
Max Fix Rate Link: 3 and 6 Gbps per lane on 3 lanes, 6, 8 and 10 Gbps on 4 lanes
Supports 12-bits/component Deep Color 4:2:0 Pixel Encoding
Supports 10-bits/component Deep Color 4:2:0 Pixel Encoding
Supports Auto Low-Latency Mode

Data Block #5
Extended Data Block: Video Capability Data Block
CE scan behaviour: Support both over- and underscan
IT scan behaviour: Support both over- and underscan
PT scan behaviour: No Data
RGB quantization: Selectable (via AVI Q)
YCbCr quantization: Selectable (via AVI Q)

Data Block #6
Extended Data Block: Colorimetry Data Block
BT2020YCC
BT2020RGB

Data Block #7
Extended Data Block: HDR static metadata data block
Electro optical transfer functions:
Traditional gamma - SDR luminance range
SMPTE ST2084
Hybrid Log-Gamma
Supported static metadata descriptors:
Static metadata type 1

Data Block #8
Extended Data Block: YCbCr 4:2:0 capability map data block
VIC # 0: ( 97) 3840 x 2160 @ 60Hz - 16:9
VIC # 1: ( 96) 3840 x 2160 @ 50Hz - 16:9
VIC # 2: (118) 3840 x 2160 @ 120Hz - 16:9
VIC # 3: (117) 3840 x 2160 @ 100Hz - 16:9
VIC # 4: (102) 4096 x 2160 @ 60Hz - 256:135
VIC # 5: (101) 4096 x 2160 @ 50Hz - 256:135
VIC # 6: (219) 4096 x 2160 @ 120Hz - 256:135
VIC # 7: (218) 4096 x 2160 @ 100Hz - 256:135

Data Block #9
Extended Data Block: Vendor specific Video Data Block - Vendor Identifier: 00:D0:46 (Dolby)
Version: 2 (12 bytes)
DM Version: 4.x
Backlt Min Luma: 100 cd/m^2
Interface: Standard + Low-Latency
Supports 10b 12b 444: Not supported
Target Min PQ v2: 0
Target Max PQ v2: 2965
Unique Rx, Ry: 0.67578125, 0.32031250
Unique Gx, Gy: 0.25390625, 0.68359375
Unique Bx, By: 0.14453125, 0.05078125

Detailed Timing Blocks - 1 is native

I compared it to a direct connection to a MBP M1 Max HDMI 2.0 Port, and there were only a few differences:
LGC1_EDID_COMPARE.png
 
Last edited:

Monstieur

macrumors member
Oct 16, 2018
49
47
For 4K 120Hz I am confused how this works:
Thunderbolt -> DP 1.4 Monitor

But this doesn't:
Thunderbolt -> Dock/Adapter e.g. HP G2 with DP 1.4 -> HDMI 2.1

Doesn't OSX just think the Dock/Adapter is some generic DP1.4 monitor ?
4K 120 Hz 8-bit RGB requires all 4 DisplayPort HBR3 lanes. DSC is required to do 10-bit RGB at 4K 120 Hz over DisplayPort 1.4. It won't work through a dock as those consume 2 lanes for other ports.

The confusing part is then why it works for a direct DP 1.4 connection.
DisplayPort 1.4 supports 1200 MHz pixel clock. macOS restricts the HDMI 2.1 adapter to HDMI 2.0 600 MHz even though it presents itself as a DisplayPort device because macOS can still read the end display as HDMI. On Windows, the GPU drivers let the adapter run at 1200 MHz.
 
Last edited:

joevt

macrumors 604
Jun 21, 2012
6,963
4,257
It won't work through a dock as those consume 2 lanes for other ports.
Only true for USB-C (Non Thunderbolt) docks that support USB 3.x.
You can get USB-C docks that support only USB 2.0 so all four superspeed lanes can be used for DisplayPort.
Thunderbolt docks/hubs have enough bandwidth for two full DisplayPort connections (two HBR2 or one HBR3 with one HBR - 4 lanes each).
 

Monstieur

macrumors member
Oct 16, 2018
49
47
Only true for USB-C (Non Thunderbolt) docks that support USB 3.x.
You can get USB-C docks that support only USB 2.0 so all four superspeed lanes can be used for DisplayPort.
Thunderbolt docks/hubs have enough bandwidth for two full DisplayPort connections (two HBR2 or one HBR3 with one HBR - 4 lanes each).
Most Thunderbolt 3 docks can do only 2x streams of 4-lane HBR2 DisplayPort 1.2. They cannot do 1x stream of 4-lane HBR3 DisplayPort 1.4 even though the bandwidth required is the same. You must use a multi-stream DisplayPort 1.2 monitor if you want the full bandwidth. This is not DisplayPort Alt Mode and was a restriction of the original Thunderbolt 3 spec as DisplayPort 1.4 did not exist at the time.

The new Thunderbolt 3 controllers support DisplayPort 1.4 but I haven't seen 4-lane HBR3 work through any dock - all of them were limited to 2-lane HBR3 i.e. 4K 60 Hz. Some new docks were advertised with 8K 60 Hz support (implying 4K 120 Hz support), but the specs were soon changed to 8K 30 Hz.

You are correct about USB-C (non-Thunderbolt) docks with USB 2.0-only ports as those are just DisplayPort Alt Mode.
 
Last edited:

joevt

macrumors 604
Jun 21, 2012
6,963
4,257
Most Thunderbolt 3 docks can do only 2x streams of 4-lane HBR2 DisplayPort 1.2. They cannot do 1x stream of 4-lane HBR3 DisplayPort 1.4 even though the bandwidth required is the same. You must use a multi-stream DisplayPort 1.2 monitor if you want the full bandwidth. This is not DisplayPort Alt Mode and was a restriction of the original Thunderbolt 3 spec as DisplayPort 1.4 did not exist at the time.
Right, DisplayPort Alt Mode cannot put two full 4 lane SST connections over a USB-C connection. Only Thunderbolt can put two separate 4 lane DisplayPort connections on the same cable.

The Thunderbolt 3 docks you are referring to use Alpine Ridge - limited to DisplayPort 1.2 / HBR2. Hosts that use Alpine Ridge have the same limit.

Two DisplayPort 1.2 4K displays will use one 4 lane HBR2 connection each.
One DisplayPort 1.2 dual link SST 5K display (like the LG UltraFine 5K) will use two 4 lane HBR2 connections.

The new Thunderbolt 3 controllers support DisplayPort 1.4
Titan Ridge for Thunderbolt 3 hosts and devices; Goshen Ridge for Thunderbolt 4 devices; Maple Ridge for Thunderbolt 4 hosts; Ice Lake for Thunderbolt 3 hosts; Tiger Lake and Apple Silicon for Thunderbolt 4 hosts. All these can do 4-lane HBR3.

but I haven't seen 4-lane HBR3 work through any dock - all of them were limited to 2-lane HBR3 i.e. 4K 60 Hz.
I don't think you've tried a Mac/GPU/Thunderbolt dock/display combination that support 4-lanes of HBR3.
2 lane HBR3 is only a limit for USB-C docks that support USB 3.x.
I have a 4K display that can do 4K120 8bpc RGB using 4 lanes of HBR3 from a Thunderbolt 4 Hub (Goshen Ridge) and a Thunderbolt 3 dock (Titan Ridge).

Some new docks were advertised with 8K 60 Hz support (implying 4K 120 Hz support), but the specs were soon changed to 8K 30 Hz.
8K 60Hz actually implies 4K 240Hz support (8K is four times more pixels than 4K). 4K 120Hz is more like 8K 30Hz.
I think they changed their specs because of problems getting 8K 60Hz to work - but I think that's more of a driver support problem than a problem with the docks.
Anyway, there aren't a lot of 8K60 displays to do testing on.

You are correct about USB-C (non-Thunderbolt) docks with USB 2.0-only ports as those are just DisplayPort Alt Mode.
USB-C (non-Thunderbolt) docks with USB 3.0 are also DisplayPort Alt Mode.
 

Thinkscape

macrumors newbie
Feb 15, 2014
10
14
After reading all the current findings and the MacOS inherent 600 MHz (software) clock limitation, do you think we have a chance of:

1. Working around the limitation in MacOS, allowing 1200MHz and thus enabling Sabrent/CableMatters to start working in 4k@120Hz modes?
2. Or working with dongle/adapter manufacturers to write firmware that would prevent MacOS from limiting the clock, or enable the modes through reported mode trickery?
 

joevt

macrumors 604
Jun 21, 2012
6,963
4,257
After reading all the current findings and the MacOS inherent 600 MHz (software) clock limitation, do you think we have a chance of:

1. Working around the limitation in MacOS, allowing 1200MHz and thus enabling Sabrent/CableMatters to start working in 4k@120Hz modes?
2. Or working with dongle/adapter manufacturers to write firmware that would prevent MacOS from limiting the clock, or enable the modes through reported mode trickery?
1 yes
2 no

Someone needs to find the code that contains the limit and then come up with a way to change the code.
 

petchav

macrumors newbie
Nov 8, 2021
2
0
If it helps, I have tried 2 methods, neither of them works. ?
On an LG OLED55C1 display and a Mac Mini M1

HDMI 2.1 => Type-C
With this cable reference: HDMI-to-Type-C
and
HDMI 2.1 => DP 1.4 => Type-C
with these cable references: HDMI-to-DP and DP-to-Type-C
 
Last edited:

petchav

macrumors newbie
Nov 8, 2021
2
0
What chip does the HDMI-to-DP cable use? AGDCDiagnose output from an Intel Mac will give that info.
I don't know AGDCDiagnose, do you have a link to an explanation of how to use it? I specify that the test was done on a Mac mini M1
 

kasakka

macrumors 68020
Oct 25, 2008
2,389
1,073
Someone needs to find the code that contains the limit and then come up with a way to change the code.
Unlikely to be something we can fix ourselves because even on Windows Nvidia needed to fix their drivers for this to work. So best we can do is report it to Apple and hope their driver teams can fix it.
 

onlinespending

macrumors regular
Jul 17, 2014
103
187
Unlikely to be something we can fix ourselves because even on Windows Nvidia needed to fix their drivers for this to work. So best we can do is report it to Apple and hope their driver teams can fix it.
Of course, this assumes it’s not a deliberate limitation
 

joevt

macrumors 604
Jun 21, 2012
6,963
4,257

Thinkscape

macrumors newbie
Feb 15, 2014
10
14
Some fixes made in the past just involve changing a few bytes in the drivers. Just need to find them.
https://github.com/acidanthera/WhateverGreen
https://github.com/Floris497/mac-pixel-clock-patch-V2
https://github.com/KhaosT/tb3-enabler
That is indeed quite interesting... I wonder how the debugging for this goes.

Is there anyone here, following, who's got the skill set to make this happen?
(either a driver for available dongles/converters, or driver patch/workaround)

I'd be open to sponsoring such effort (even if it ends up being a binary patch that requires further maintenance)
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.