-
Just to track as known issue caused by Mac OS and discussed in Better Display Discord There is a bug in Sonoma 14.1 that will limit HDR availability depending on resolution and refresh rate. This issue is caused by Sonoma 14.0 and was fixed for M2 in 14.1. M1 Still have the bug as of 14.3 |
Beta Was this translation helpful? Give feedback.
Replies: 3 comments 9 replies
-
Hi there, The limitation is refresh rate and framebuffer size dependent (so what seems to matter is the framebuffer pixel clock). Not sure why Apple did it especially as everything seemingly worked just fine for 2+ years and everybody was happy (except for some people complained about a 8bit reported framebuffer color depth but that was a small minority, most ppl did not care). This same limitation applied to M2 initially (probably due to the M2 green screen of death bug) while M1 was unlimited (which was awkward as the older chip seemingly did better than the new one). When they fixed the M2 they also limited M1 (in the same move) - this indicates that probably the same engineer (or maybe a group of engineers) made the changes to both chip firmwares (the capabilities should depend on the rtOS - firmware - version running in a dedicated component of the Apple Silicon chip called "DCP"). Multiple (rather vague) replies from Apple support clarifies that the limitation is intentional and is here to stay. It would be great if the guys actually making the change explained things a bit better though. :) One can make 120Hz HDR work up to 2432x1368 HiDPI (you can use BetterDisplay to unlock that resolution) which is slightly higher than the stock 2304x1296 mentioned here. Would have been better if the engineers pushed a bit harder and allow 2560x1440@120Hz HiDPI for HDR. :) Note regarding the DCP firmware: this firmware is loaded at boot time and is macOS version specific - each macOS version has its own build. It's not the usual firmware in the typical sense that it is somehow permanently stored inside the chip itself in some kind of eprom. It's actually a separate real time operating system (called rtOS) that runs inside this specific part of the Apple Silicon chip called DCP (Display Coprocessor probably) - a chip inside the chip. Each macOS has build has its own matching build for the DCP firmware (OS) and the two work together as the functionality is divided up between them - some stuff is handled by macOS, some by rtOS and they interface with each other in a specific way - neither of them can operate properly without the other. DCP handles display modes, all negotiations with displays (and among other things manages also the built-in DP2HDMI hardware for first gen Apple Silicon Macs) and also manages video memory bandwidth stuff (this is probably where the limitation of modes come). Since there is nothing I can do to fix this with BetterDisplay (the limitation is at DCP firmware level), I'll move this over to discussions. |
Beta Was this translation helpful? Give feedback.
-
@waydabber is there any way to force detected HDR content to auto disable HiDPI and enable HDR, so I don't have to do manually to watch a video? I have Pro if it matters. If not, it looks like Apple will be forcing me to upgrade earlier than expected. |
Beta Was this translation helpful? Give feedback.
-
@ErinPo Is HiDPI off?
Awesome, thanks! My final script (save as .app in Script Editor and add to Login Items) which is working in case anyone else wants to do this:
|
Beta Was this translation helpful? Give feedback.
Hi there,
The limitation is refresh rate and framebuffer size dependent (so what seems to matter is the framebuffer pixel clock). Not sure why Apple did it especially as everything seemingly worked just fine for 2+ years and everybody was happy (except for some people complained about a 8bit reported framebuffer color depth but that was a small minority, most ppl did not care). This same limitation applied to M2 initially (probably due to the M2 green screen of death bug) while M1 was unlimited (which was awkward as the older chip seemingly did better than the new one). When they fixed the M2 they also limited M1 (in the same move) - this indicates that probably the same engineer (or mayb…