Upon updating from macOS 10.12.3 to 10.12.4 the following issues immediately began occurring consistently anytime the MacBook tries to initiate displaying on external monitors that are in power-save mode (i.e. on, but waiting for a signal and thus with the screen off). This includes when the MacBook boots up or wakes.

  • MacBook outputs intermittent signal to external monitors via Thunderbolt interface leading to race condition like situation where the external monitor wakes from power-save mode upon receiving a signal from the MacBook, but goes back into power-save mode due to not detecting a signal (or at least a consistent one) by the time it has finished waking. This process repeats cyclicly for some time until both external monitors happen to get in sync with the MacBook (or they don't, see below).
  • The MacBook enters an unresponsive state after being in the above cycle for some time without the external monitors getting in sync with the MacBook. This requires a forced shut down which can lead to loss of data and unsaved work.

Please see my answer below for further details.

  • For what it's worth: I've seen similar symptoms from a monitor with a faulty DisplayPort input.
    – user101978
    Commented Mar 29, 2017 at 18:25
  • In both cases the symptoms can be attributed to an intermittent signal. But in this case, it's pretty clear the issue is being caused by the output device, in particular the 10.12.4 update.
    – fvgs
    Commented Mar 29, 2017 at 20:40
  • Can you thin this into one question - probably how to troubleshoot display connections? Also, the edit on escalation should probably be added as an answer based on which question this ends up being.
    – bmike
    Commented Apr 1, 2017 at 13:49
  • @bmike This is a highly specific question relating to a bug that appeared as the result of a recent update. The question is simply "How to resolve this very specific, new bug?". There very well may be no straightforward answer at the moment. In which case, the eventual answer will be to update to a future, but as of yet unreleased, patch. The purpose of investigating this problem in the open is so multiple people can contribute valuable information which will help in that patch eventually being released, thus providing an answer to the question of "How to resolve this very specific, new bug?".
    – fvgs
    Commented Apr 1, 2017 at 14:16
  • I'll take a stab at editing to remove the extra questions. The question is clear and not you can put all the troubleshooting, bugs to apple, etc... in the answer section. Wrong answers and impartial answers are great for the site. Rambling, unclear, excessively long narratives with several questions at the end - not so much. I'll throw my experience in an answer - it might be totally wrong, but it might help you and it might help others. You may be on to the update, but please put your details in an answer - it's worth +1 as well for how to troubleshoot this.
    – bmike
    Commented Apr 1, 2017 at 14:21

5 Answers 5


The following paragraph is meant as a high level summary of the issue that is occurring, and I attempt to explain the relevant parts in more detail in the subsequent paragraphs.

Immediately after upgrading from macOS 10.12.3 to 10.12.4 I noticed that when my MacBook Pro (Late 2013) boots up or wakes while plugged into my two external displays via Mini DisplayPort, my MacBook has extreme difficulty getting them both working at the same time. The MacBook's screen will turn on and off as it tries to get in sync with the external displays (to start displaying on them as normal). Meanwhile, the displays seem to cycle between receiving a signal and receiving no signal as suggested by the fact that the displays will wake, but then display "No Signal" and go back into power-save mode. Eventually, after many cycles of this, one or both displays may "get in sync" with the MacBook and function normally from there on out. But sometimes it goes through many such cycles without getting in sync. When the displays and the MacBook fail to get in sync after many cycles and continue to cycle as described, I may unplug one or both external monitors and try to get them to sync up individually, which is usually easier than trying both at once. This makes sense because, as I explain below, this problem is effectively a race condition and appears to be compounded with every additional external display in use. Finally, the more serious case is when the cycle I've described ends in the MacBook entering an unresponsive state and requiring a forced shutdown which can lead to loss of data or unsaved work. Furthermore, this whole issue presents itself every time I boot up or wake the MacBook.

To be clear, the displays wake from power-save mode upon receiving a signal from my MacBook (e.g. when it boots or wakes from sleep). But after receiving that initial signal waking the display from power-save mode, the display shows "No Signal" and goes back into power-save mode. Meanwhile, the display on my MacBook goes from normal to a blank screen for a couple seconds and back to normal as it seemingly switches between outputting to the external display and not outputting. This is the cycle that continuously repeats until either each respective monitor gets in sync with the MacBook and works as normal, or the MacBook enters an unresponsive state as described below. In the case where the monitors do eventually get in sync, it appears to occur the moment the monitor waking happens to coincide with the MacBook outputting a consistent signal to that respective monitor. So, it's effectively a race condition. The reason I say the MacBook cycles between outputting to the monitor and not outputting to it is because, while it's in this cycle, I'll observe windows I had open on the external monitor prior to the MacBook sleeping toggle between being displayed on the MacBook's display and not being displayed at all (presumably they're being output to the external display).

As I mentioned above, there is an even more troublesome situation I've now encountered multiple times where, upon waking from sleep, the MacBook will enter the cycle of trying to get in sync with the monitors for a while and then enter an unresponsive state, showing a blank screen. Plugging in and unplugging the monitors from the MacBook or power cycling the monitors seems to have no effect in this situation. Even if the MacBook is recognizing one of the monitors while in this unresponsive state, that monitor is also just rendering a blank screen (but is not going into sleep mode therefore is receiving a signal). My only option has been to force shut down my MacBook after this has happened by pressing the power button and leaving it pressed.

I have two Acer S241HL displays each connected via Mini DisplayPort. I had never had this issue before and it started immediately after I upgraded from 10.12.3 to 10.12.4. In the past, my MacBook's screen might turn on/off once or twice when booting up (and maybe waking?) before all displays were on and rendering properly. But that was always quick and never an issue. This is a much greater extreme and a major inconvenience as I have to wait and sometimes mess with the monitors every time my MacBook wakes or boots up. Even more serious is that it is now dangerous for me to wake my MacBook if I have monitors plugged in due to the risk of it entering an unresponsive state, thus requiring a forced shut down. This can easily cause problems including loss of data and unsaved work.

I've contacted Apple Support to escalate this issue to engineering. They should be aware of this thread, so if you experience this same issue, please comment with details of your setup such as which video out port(s) you're using and what kinds of cables/adapters, as well as any other pertinent information.

  • Are you gathering sudo sysdiagnose when the issue first starts? Would be interesting to see how soon after a restart the video becomes unstable or loses sync. Would be very useful after 4 diagnose dumps are available for review.
    – bmike
    Commented Apr 6, 2017 at 10:36
  • I have this same issue but I am able to work around it by unplugging my monitor and rebooting my MacBook. Once the reboot has completed and I have logged back in, I plug the monitor back up and it works. If my MacBook ever goes to sleep I must repeat the process to get the monitor to work. I never experienced this issue until I upgraded to High Sierra. Commented Nov 10, 2017 at 1:51
  • I am still having this problem with my Macbook Pro late 2013 with OS-X 10.13.5. Commented Jun 18, 2018 at 21:45

It may be actually worth resetting the NVRAM and SMC (in that order) to see if this resolves your issues. Before doing so, unplug all external devices (including monitors, keyboard, etc).

Resetting NVRAM on your model MacBook Pro

Your Mac use Non-Volatile Random-Access Memory (NVRAM) to store a whole range of settings. Here’s how to reset this:

  1. Shut down your Mac. Yes, a full shut down, not just logging out.
  2. Press the power button and then press the commandoptionpr keys. You have to make sure you press these keys before the gray screen appears or it won’t work.
  3. Hold those keys down until your Mac reboots again and you here the startup chime.
  4. Let go of the keys and let your Mac reboot normally.

Note: When you log back in you may need to readjust some of your system preferences (e.g. speaker volume, screen resolution, startup disk selection, time zone information, etc).

Now proceed to reset the SMC.

Resetting the SMC on your model MacBook Pro

The System Management Controller (SMC) impacts a whole range of power management functions. Here's how to reset this:

  1. Shut down your Mac
  2. Keep the MagSafe adapter (power cable) plugged in.
  3. Press at the same time shiftoptioncontrol (on the left side of the built-in keyboard) and the power button
  4. Let go
  5. Turn your computer back on with the power button.

Let us know how you go.

  • 1
    I tried reseting both the SMC and NVRAM just to cover all the bases, although I didn't see a reason why doing so would help. Unfortunately, neither had an observable effect. I reached out to Apple Support to escalate the issue and they suggested reseting the SMC first before the NVRAM. This also had no observable effect.
    – fvgs
    Commented Apr 1, 2017 at 13:24
  • Well, the SMC is responsible for a whole range of power management settings/functions, including your Mac's sleep behaviour (a key part of your question), while NVRAM contains a whole range of system settings, some of which relate to your displays (e.g. resolution, depth, etc) and since your question is about your displays, I thought resetting both was worth a shot. It's good you've reached out to Apple Support. :) Evenso, I'd still follow Graham's advice and provide feedback to Apple.
    – Monomeeth
    Commented Apr 1, 2017 at 14:06
  • It was certainly worth trying, and I appreciate your proposing it as a possible solution. Based on my likely incomplete understanding of the problem, the underlying cause has to do with what is effectively a race condition having to do with an inconsistent signal being output by the MacBook, and the monitor being awakened from power-save mode only to find there is no signal, and thus going back into power-save mode. Only in the case where those occurrences happen to coincide does it work as desired. I've submitted feedback as well, so hopefully this will be resolved in the near future
    – fvgs
    Commented Apr 1, 2017 at 14:30
  • This is great advice and would work for most one time issues. This seems more a corruption over time or flakey cable / display situation. Hard to tell without a) detailed notes on timing and occurrence data b) system logs
    – bmike
    Commented Apr 6, 2017 at 10:37

Let Apple Know

If you have a reproducible test case (preferably after isolating for extraneous variables like sleep / wake / third party extensions and on a cleanly installed OS), the best way to alert engineering at Apple is via their Bug Reporter. You can use an Apple ID to log in and provide a detailed bug report.

If you lack access there, Apple Support can escalate issues to engineering. The least direct manner is the general Feedback page.

  • The Bug Report seems to require a developer account, not just an Apple ID?
    – fvgs
    Commented Mar 29, 2017 at 15:21
  • That is unfortunate.That leaves you with the feedback option. Another option is to talk to someone at an Apple Store but this is unlikely to get back to Apple's engineers. Commented Mar 29, 2017 at 16:02

We've had lots of issues with third party sync and displays since 10.11 so it might not at all be related to the update (or that the update enables new features that are exposing faulty cables more readily).

Our help desk now has several sets of known good cables and known good displays we take and systematically test along side our normal "known good clean OS install" drive on external bootable media.

Our triage process is:

  1. Reseat all cables, boot to safe mode and verify that the display itself works.
  2. Collect data on the timing of the issue - does it happen every single time, or on occasion?
  3. Perform a test - 5 sleeps and 5 wakes or three reboots and record the results.

At this point, we'll know if it's really something related to the OS or if it's an occasional problem.

If you don't have an IT help desk - you might need to do some of these in another order. If you're convinced it's the update, install a clean OS (same version) on external drive and test that for a few hours. That sort of detail usually gets Apple Engineering involved if you can reproduce the issue with a clean install.

Don't overlook cables as well - we're replacing a whole lot more cables than we did in past years. It's too soon to tell if we just bought lesser quality cables or if the new hardware and drivers are just more picky. The good news is cables are relatively cheap and you can mark the suspect ones for reuse later if it turns out to be software and not the cables.

  • What brands are the 'good cables'? Commented Oct 15, 2017 at 12:59
  • @AnthonyKong It's more about a specific good cable than there are many sketchy vendors. Anyone with product sold by Apple generally is good - belkin, cable matters, monoprice, and one sold by someone that lets you know you have 1 year warranty with them specifically (retailer or vendor direct) is usually a sign you've got a good manufacturer.
    – bmike
    Commented Oct 15, 2017 at 13:55

I am seeing something like this with an Acer monitor (H233H) on a direct HDMI connection from a late 2013 MacBook Pro on 10.14.6.

I'm not getting the crashes, but I do see a race condition here, to me it looks like:

  1. The monitor starts up, and pokes the HDMI connection
  2. The MacBook sees this and starts up dual screen mode
  3. ...
  4. The MacBook still hasn't returned anything on the HDMI interface, so the monitor gives up and shows "no signal"
  5. The MacBook finishes switching modes, and starts sending a signal
  6. But the monitor isn't listening any more
  7. Power cycling the monitor will break the HDMI connection and return the MacBook to single-screen mode (ie. goto step 1)

I've found that sometimes the monitor can restart fast enough to avoid triggering this last step, but apart from that I haven't figured out any way around this.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .