Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Status
The first post of this thread is a WikiPost and can be edited by anyone with the appropiate permissions. Your edits will be public.
@mindfulmonk "Do you know how dual 5k120hz monitors should behave when connected the way you describe? "

Daisy-chained to a TB5 Mac (or TB5 dock) everything should work fine...?
If 5K/120 isn't possible then one or both will drop down to 5K/60.

With TB4 I don't know what Apple has enabled - at least both at 5K/60.
I wouldn't expect one to exceed that, but who knows...

The M4 mini has three TB4 spec streams of video available out of it's TB4/HDMI ports.
Because of the laptop's screen, a M4 Air only has two.

"...the requirement for thunderbolt4 is due to how the thunderbolt controller is implemented on the Macs"
Early AS M1/P/M/U Macs had less sophisticated DSC capabilities (TB3-level?), so have restrictions when working with later hi-res monitors.
 
Last edited:
However if you connect two DSC-capable USB-C 5K/60 monitors to the two downstream TB4 ports - using 8K-specified USB-C cables or adapters (with 4 HBR3 video channels), then the DSC bandwidth needed for each monitor is below half the available TB4 bandwidth, and everything works fine.
yes, I also thought so but as @mindfulmonk said in his topic - this doesn't work even with M4 Mac mini. Just one monitor was working through the dock. Have I got it correctly, @mindfulmonk ?
 
Only 1 monitor would work through the dock or directly when using an mbp m2 pro or m4 air.

On the mac mini it worked. One monitor was 60hz and the other up to 165hz, both operating at 5k(my expectatation was that it would be 5k 60hz on both).
 
Only one of these channels can be terminated with a DP 1.4 output socket on a TB4 dock (or converted to 4K/60 HDMI).
A Thunderbolt 4 dock based on the Intel Goshen Ridge Thunderbolt 4 peripheral controller can have two DisplayPort outputs - DisplayPort or USB-C DisplayPort Alt Mode. Each DisplayPort output removes the use of a downstream Thunderbolt 4 port.

More than one DP or HDMI port indicates a dock with DisplayLink or MST functionality (for Windows, not Mac).
A Thunderbolt 4 dock based on the Intel Goshen Ridge Thunderbolt 4 peripheral controller that has a count of downstream Thunderbolt / DisplayPort / HDMI ports greater than 3 is using DisplayLink or MST for some of the DisplayPort or HDMI ports or the dock has more than one Thunderbolt controller.

Downstream Thunderbolt 4 ports greater than 3 means the dock has more than one Thunderbolt 4 controller.
DisplayPort / HDMI ports greater than 2 means DisplayLink or MST is being used, or the dock contains a second Thunderbolt controller. This is not useful unless connected to a Thunderbolt 5 host that has a Thunderbolt 5 host controller with more than 2 DisplayPort inputs.

I haven't seen a dock with more than one Thunderbolt controller (there are Thunderbolt 3 eGPU enclosures with more than one Thunderbolt 3 controller but they don't expose extra DisplayPort / HDMI outputs).

The two channel's bandwidth can be aggregated into a single (or twin*) channel output through a single downstream TB4 port - which can feed up to a single 6016x3456 6K/60 monitor uncompressed (just, very just, when using MacOS - Apple's secret Pro XDR sauce).
Apple did this secret sauce with a Titan Ridge Thunderbolt 3 controller in the 6K XDR display, and they don't allow this mode to work with a Thunderbolt dock or hub between the 6K XDR display and the Mac because reasons. It uses more than 38 Gbps of the 40 Gbps total. Using DSC 6K60 requires only 15.4 Gbps (less than uncompressed 4K60 = 16 Gbps).

With DSC you can feed two 5K/60 monitors from two TB4 out ports, but (most likely) not the DP 1.4 socket, because of the conversion hardware in the dock being 4K/60 - for PC users...
I don't know what "conversion hardware in the dock being 4K/60" means.
With DSC you can feed two 6K60 displays (not just 5K60) from two different ports because they use less than half the bandwidth of Thunderbolt 3/4.

However if you connect two DSC-capable USB-C 5K/60 monitors to the two downstream TB4 ports - using 8K-specified USB-C cables or adapters (with 4 HBR3 video channels), then the DSC bandwidth needed for each monitor is below half the available TB4 bandwidth, and everything works fine. 🙂
Only DisplayPort 1.2 (HBR2 x4 lanes) cables are required for 6K60 with DSC.

The vital factor is that the USB-C cables to the (USB-C) monitors must be fed from a TB4/5 port from a TB4/5 dock or Mac. That is the only protocol by which the Mac will deliver the two DSC data-streams via a single port.
A Thunderbolt 3 port should be capable. The Thunderbolt controller of the host needs to have its DisplayPort inputs connected (usually internally) to a GPU that can supply two 6K60 streams with DSC.

@PaulD-UK Do you know how dual 5k120hz monitors should behave when connected the way you describe? My expectation was that they would appear as 5k60hz on the mac, but instead I got different behaviour between a mac mini and a macbook air(both m4).
5K120 is 1902 MHz at least. That's 15.2 Gbps for DSC at 8 bpp. Or 22.8 Gbps for DSC at 12 bpp.
IF the GPU supports the resolution and refresh rate and DSC at 8 bpp, then you might be able to get two of these from Thunderbolt 3/4. Or you might not. macOS doesn't usually let you choose the DCS target bit depth. Maybe the DSC target bit depth can be forced by constricting the signal through a DisplayPort 1.2 hub.

A few people suggested to me that the requirement for thunderbolt4 is due to how the thunderbolt controller is implemented on the macs(and I think those commands from @joevt hint at that too?)
I don't think I hinted that?

@mindfulmonk "Do you know how dual 5k120hz monitors should behave when connected the way you describe? "

Daisy-chained to a TB5 Mac (or TB5 dock) everything should work fine...?
If 5K/120 isn't possible then one or both will drop down to 5K/60.

With TB4 I don't know what Apple has enabled - at least both at 5K/60.
I wouldn't expect one to exceed that, but who knows...

"...the requirement for thunderbolt4 is due to how the thunderbolt controller is implemented on the Macs"
Early AS M1/P/M/U Macs had less sophisticated DSC capabilities (TB3-level?), so have restrictions when working with later hi-res monitors.
Yeah, it's up to Apple. It should be possible with Thunderbolt 3/4 if DSC@8bpp is used for both displays and the GPU and drivers allow it (which they probably don't).

Only 1 monitor would work through the dock or directly when using an mbp m2 pro or m4 air.

On the mac mini it worked. One monitor was 60hz and the other up to 165hz, both operating at 5k(my expectatation was that it would be 5k 60hz on both).
5K60 is ≈ 930 MHz.
5K165 is ≈ 2700 MHz.
If both are using DSC at 8 bpp, then I suppose the latter got connected first using HBR3 x4 and the former got connected second at HBR x4.
Force them both to use HBR2 x4 by connecting an HBR2 x4 display first (usually 4K60 uncompressed) then connect a 5K60 display, then disconnect the 4K60 display and connect another 5K60 display.
Or use DisplayPort 1.2 hubs so that HBR3 x4 will not be chosen.
Or use HBR3 x2 cables (cut some wires).
 
If you have access and a method for saving the produced text file, then this will get all the info:
Code:
ioreg -fliw0 > ioregall.txt

This one doesn't save any properties - it just gets the list of nodes:
Code:
ioreg -iw0 > ioreg.txt

Then search for DPIn.

This M4 example has three Thunderbolt controllers (one per Thunderbolt port), each with two DisplayPort In Adapters.
Code:
    | |   +-o acio0@81F00000  <class IORegistryEntry:IOService:AppleARMIODevice>
    | |   | +-o AppleThunderboltHALType7  <class IORegistryEntry:IOService:AppleThunderboltGenericHAL:AppleThunderboltHALGenericACIO:AppleThunderboltHALType7>
    | |   | | +-o AppleThunderboltNHIType7  <class IORegistryEntry:IOService:IOThunderboltNHI:AppleThunderboltNHI:AppleThunderboltNHIGenericACIO:AppleThunderboltNHIType7>
    | |   | |   +-o IOThunderboltControllerType7  <class IORegistryEntry:IOService:IOThunderboltController:IOThunderboltControllerGenericACIO:IOThunderboltControllerType7>
    | |   | |     |   +-o IOThunderboltPort@5  <class IORegistryEntry:IOService:IOThunderboltNub:IOThunderboltPort>
    | |   | |     |   | +-o AppleThunderboltDPInAdapterOS2  <class IORegistryEntry:IOService:IOThunderboltTunnelDriver:AppleThunderboltDPAdapterFamilyBase:AppleThunderboltDPInAdapterBase:AppleThunderboltDPInAdapterCM:AppleThunderboltDPInAdapterOS:AppleThunderboltDPInAdapterOS2>
    | |   | |     |   +-o IOThunderboltPort@6  <class IORegistryEntry:IOService:IOThunderboltNub:IOThunderboltPort>
    | |   | |     |     +-o AppleThunderboltDPInAdapterOS2  <class IORegistryEntry:IOService:IOThunderboltTunnelDriver:AppleThunderboltDPAdapterFamilyBase:AppleThunderboltDPInAdapterBase:AppleThunderboltDPInAdapterCM:AppleThunderboltDPInAdapterOS:AppleThunderboltDPInAdapterOS2>

    | |   +-o acio1@89F00000  <class IORegistryEntry:IOService:AppleARMIODevice>
                -- same as above

    | |   +-o acio2@91F00000  <class IORegistryEntry:IOService:AppleARMIODevice>
                -- same as above
Attached from an Apple demo 16” M5 Max. Hope it’s helpful!
 

Attachments

  • Like
Reactions: joevt
This may not answer your question, but Apple's white paper is now available:


There are various components including "a newly developed cadmium-free quantum dot-based optical film stack, as well as a new set of color filters on the LCD panel itself. The optical film stack includes a color-conversion film that controls precise quantum dot particle size for color consistency and features a unique internal microstructure for long-term performance stability."

In the timing controller (TCON) section, they say "Apple-designed" but they don't say that here. They also say the mini-LED backlight system is "proprietary," but they don't say that either. Just that it's "newly developed" -- so we can probably infer something from that.

The section on color gamut is interesting, I gather it means it can display full coverage of both P3 and Adobe RGB simultaneously, and it can work in BT.2020 even if it can't display the entire gamut.
On the polarizer (part of the optical film stack), I take that back -- they say "An Apple-designed, industry-leading polarizer technology maintains color accuracy and contrast for a large viewing angle"
 
Two monitors connected directly to the m4 mac mini
thanks.

So yes, it turns out that for some reason this:
if you connect two DSC-capable USB-C 5K/60 monitors to the two downstream TB4 ports - using 8K-specified USB-C cables or adapters (with 4 HBR3 video channels), then the DSC bandwidth needed for each monitor is below half the available TB4 bandwidth, and everything works fine
is not working.
 
@thenewepic
Since @mindfulmonk didn't specify the spec of the monitor which worked and which didn't with his M4 Air, the reason the second monitor didn't work isn't obvious - unless it was set to 5K/120, which an Air can't support with another monitor.
 
Attached from an Apple demo 16” M5 Max. Hope it’s helpful!
Yes, it does look like each Thunderbolt 5 port can support 4 displays. That doesn't mean you can connect 12 displays.
Here's the relevant parts:
Code:
    | |   +-o acio0@A1F00000  <class IORegistryEntry:IOService:AppleARMIODevice>
    | |   | +-o AppleThunderboltHALType7  <class IORegistryEntry:IOService:AppleThunderboltGenericHAL:AppleThunderboltHALGenericACIO:AppleThunderboltHALType7>
    | |   | | +-o AppleThunderboltNHIType7  <class IORegistryEntry:IOService:IOThunderboltNHI:AppleThunderboltNHI:AppleThunderboltNHIGenericACIO:AppleThunderboltNHIType7>
    | |   | |   +-o IOThunderboltControllerType7  <class IORegistryEntry:IOService:IOThunderboltController:IOThunderboltControllerGenericACIO:IOThunderboltControllerType7>
    | |   | |     +-o IOThunderboltLocalNode  <class IORegistryEntry:IOService:IOThunderboltLocalNode>
    | |   | |     | +-o IOThunderboltXDomainServiceClientManager  <class IORegistryEntry:IOService:IOThunderboltXDomainServiceClientManager>
    | |   | |     | +-o AppleThunderboltIPService  <class IORegistryEntry:IOService:AppleThunderboltIPService>
    | |   | |     |   +-o AppleThunderboltIPPort  <class IORegistryEntry:IOService:IONetworkController:IOEthernetController:AppleThunderboltIPPort>
    | |   | |     +-o IOThunderboltPort@7  <class IORegistryEntry:IOService:IOThunderboltNub:IOThunderboltPort>
    | |   | |     | +-o IOThunderboltSwitchType7  <class IORegistryEntry:IOService:IOThunderboltNub:IOThunderboltSwitch:IOThunderboltSwitchOS:IOThunderboltSwitchOS2:IOThunderboltSwitchType7>
    | |   | |     |   | +-o AppleThunderboltPCIDownAdapterType5  <class IORegistryEntry:IOService:IOThunderboltTunnelDriver:AppleThunderboltPCIDownAdapter:AppleThunderboltPCIDownAdapterType5>
    | |   | |     |   | +-o AppleThunderboltUSBDownAdapter  <class IORegistryEntry:IOService:IOThunderboltTunnelDriver:AppleThunderboltUSBDownAdapter>
    | |   | |     |   | +-o AppleThunderboltDPInAdapterOS2  <class IORegistryEntry:IOService:IOThunderboltTunnelDriver:AppleThunderboltDPAdapterFamilyBase:AppleThunderboltDPInAdapterBase:AppleThunderboltDPInAdapterCM:AppleThunderboltDPInAdapterOS:AppleThunderboltDPInAdapterOS2>
    | |   | |     |   | +-o AppleThunderboltDPInAdapterOS2  <class IORegistryEntry:IOService:IOThunderboltTunnelDriver:AppleThunderboltDPAdapterFamilyBase:AppleThunderboltDPInAdapterBase:AppleThunderboltDPInAdapterCM:AppleThunderboltDPInAdapterOS:AppleThunderboltDPInAdapterOS2>
    | |   | |     |   | +-o AppleThunderboltDPInAdapterOS2  <class IORegistryEntry:IOService:IOThunderboltTunnelDriver:AppleThunderboltDPAdapterFamilyBase:AppleThunderboltDPInAdapterBase:AppleThunderboltDPInAdapterCM:AppleThunderboltDPInAdapterOS:AppleThunderboltDPInAdapterOS2>
    | |   | |     |   | +-o AppleThunderboltDPInAdapterOS2  <class IORegistryEntry:IOService:IOThunderboltTunnelDriver:AppleThunderboltDPAdapterFamilyBase:AppleThunderboltDPInAdapterBase:AppleThunderboltDPInAdapterCM:AppleThunderboltDPInAdapterOS:AppleThunderboltDPInAdapterOS2>
    | |   | |     |     +-o AppleThunderboltUSBType2DownAdapter  <class IORegistryEntry:IOService:IOThunderboltTunnelDriver:IOThunderboltTunnelDriverType2:AppleThunderboltUSBType2DownAdapter>
    | |   | |     +-o AppleThunderboltDPConnectionManager  <class IORegistryEntry:IOService:AppleThunderboltDPConnectionManager>
    | |   | +-o IOTBTTunnelClientInterfaceManager  <class IORegistryEntry:IOService:IOTBTTunnelClientInterfaceManager>
    | |   +-o acio1@A9F00000  <class IORegistryEntry:IOService:AppleARMIODevice>
    ... same as above
    | |   +-o acio2@81F00000  <class IORegistryEntry:IOService:AppleARMIODevice>
    ... same as above

I wonder what a AppleThunderboltUSBType2DownAdapter is? It also has a normal AppleThunderboltUSBDownAdapter.
Is AppleThunderboltUSBType2DownAdapter for supporting USB 3.2 gen 2x2 (20 Gbps) ?
https://eclecticlight.co/2026/02/11/what-has-changed-in-macos-tahoe-26-3/
I don't think a separate adapter type should be necessary for that.

The old adapter type is 0x200101 and has description "USB Adapter" and should support up to USB 3.2 gen 2x2. Previously, Apple Silicon Macs with Thunderbolt 5 ports were limited to USB 3.2 gen 2x1 unless you disable USB tunnelling by placing a Thunderbolt 3 dock between the Apple Silicon Mac and a Thunderbolt 5 hub/dock. This enables the Thunderbolt 5 USB controller in the Thunderbolt 5 dock which supports USB 3.2 gen 2x2.

The new adapter type is 0x210101 and has description "USB Gen T Adapter". The USB4v2 spec should describe this new type of USB tunnelling. I think it's USB data that can exceed USB 3.2 gen 2x2 limits. This is different than PCIe and DisplayPort and USB 3.2 gen 2x2 data.
https://www.graniteriverlabs.com/en-us/technical-blog/usb4-80-cio80

With the new inclusion of USB Gen T, does M5 now include USB 3.2 gen 2x2 support for the old Thunderbolt USB adapter type (now called USB Gen X) such that you can now get USB 3.2 gen 2x2 from a Thunderbolt 5 hub/dock without disabling USB tunnelling?

With USB Gen T, you should now be able to connect multiple USB 3.2 gen 2 devices and not be limited to 10 Gbps total, or even 20 Gbps total. This eliminates a bottleneck introduced with Titan Ridge and later docks. You have to check that a USB 3.2 gen 2 hub is not in the USB hierarchy or you need to connect the USB devices to different docks.

Previously, an Alpine Ridge Thunderbolt 3 dock would include multiple PCIe USB controllers such that the total bandwidth of all USB devices could exceed 20 Gbps. For example, the TS3 Plus Dock had two FL1100 (4 Gbps each), an ASM1142 (7.9 Gbps), and the Thunderbolt USB controller (10 Gbps) = 25.9 Gbps (but PCIe is limited to ≈22 Gbps).
https://www.caldigit.com/ts3-plus-interface-bandwidth-allocation-and-diagram/

Titan Ridge Thunderbolt 3 controllers and later Thunderbolt 4/5 controllers support USB and Thunderbolt upstream. Dock manufacturers usually did not add extra PCIe devices (especially additional USB controllers) because they cannot be used by a USB-only upstream host. These docks include USB hubs to add additional USB devices so the total data was limited to 10 Gbps. The benefit over Alpine Ridge docks is that all ports can support up to 10 Gbps just not all at the same time.
 
Last edited:
Another glossy entry around that $900 price point. *Same as 2025 Clarity 5K Fold Touch, but without Fold stand and Touch screen:

Alogic Clarity 5K 27C5KPD - 2026*
Panel: IPS - 27" - 5120x2880 - 218ppi - 60Hz - 400 nits - 1000:1 contrast
Glass: Standard (low-reflection)
Color: 8-bit depth + FRC - 99% DCI-P3 - 100% sRGB - 99% Adobe RGB - deltaE<1 accuracy
I/O: 2x HDMI 2.0 (30Hz), DisplayPort 1.4 (60Hz), USB-C (60Hz, DP Alt Mode, 65W); USB 3.0 Hub (2x USB-A downstream, 1x USB-B upstream)
## 3.5mm audio out, 2x 5W speakers
Alogic site $949
 
Got to say I've been enjoying the Asus XG27JCG I received earlier in the week. It replaces a 27" 4k 160hz monitor so I'm doing some comparisons between them. I can see the extra clarity of the 5k but with my eye sight and sitting back in my usual position I'm not sure if it's that noticeable or worth the £700. I'll swap back to the 4k monitor soon and make a decision if I'll keep it.

IMG_0664.jpeg
 
Got to say I've been enjoying the Asus XG27JCG I received earlier in the week. It replaces a 27" 4k 160hz monitor so I'm doing some comparisons between them. I can see the extra clarity of the 5k but with my eye sight and sitting back in my usual position I'm not sure if it's that noticeable or worth the £700. I'll swap back to the 4k monitor soon and make a decision if I'll keep it.

View attachment 2615563
Well, with a matte finish, that’s certainly understandable.
 
Gloss isn’t possible for me, my office has 2 large south facing windows to the right.

How is the brightness on this monitor during the daytime? The official specs only mention 350 nits which is obviously lower than a 500 nit Macbook, or a 600 nit Studio Display.
 
Another sighting of the Hisense (along with a new RGB MiniLED, not shown at CES), at the AWE 2026 trade show in Shanghai. The name seems to have changed (it was “GX Pro” at CES), and still no press release or anything official from Hisense that I can find, and note the error by the source (IT Home), where they misinterpret the OD in “0OD” as “Overdrive” when it should be “Optical Distance” …

Hisense GX Ultra - 2026 [?]
Panel: IPS (2,304-zone MiniLED backlight with Zero Optical Distance) - 27" - 5120x2880 - 218ppi - dual mode 165Hz/330Hz (5K/1440p) - 2000 nits
Glass: [?]
Color: [?]
I/O: DisplayPort 2.1
IT之家 (March 2026)
 
Last edited:
  • Like
Reactions: rafzamz and just!ns
LG UltraGear evo 27GM950B - Q2 2026
Panel: IPS (2,304-zone MiniLED backlight with Zero Optical Distance) - 27" - 5120x2880 - 218ppi - dual mode 165Hz/330Hz (5K/1440p) - DisplayHDR 1000
Glass: Matte (anti-glare)
Color: 99% DCI-P3 - deltaE<2 accuracy
I/O: DisplayPort 2.1 (UHBR20), 2x HDMI 2.1, USB-C (DP Alt Mode, 90W)
## AI upscaling
LG site

Hopefully not much longer to wait

 
So the MA270S has been out in some markets for a few weeks now but every single review has been of the "shiny shiny, rich colours, vibrant colours, lovely lovely" type. Telling me nothing I coudn't read on a spec sheet.

Finally, a Chinese guy has reviewed it and gives absolutely loads of technical details. Highly recommend viewing the whole thing (with subtitles). You won't get this level of detail from many English-language reviewers.

Results: 490 nits, and excellent brightness uniformity (420 nits in uniform brightness mode). I don't think you would get that uniformity from a Kuycon glossy.

So this is pretty much a like-for-like replacement for the LG Ultrafine and iMac 5K, although you need to sacrific a little brightness for that uniformity.

 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.