DevBiker

As that's likely the cause, yes, absolutely. If you control your case fans from the BIOS, try to see if you can tie them to that MB4 temperature and set a fan curve that starts increasing around 34C. Additional exhaust will also likely be helpful. The exact details are hard to say without knowing the layout of the fans and your case. We've seen cases where having the cooler on top as exhaust actually causes heat buildup in the top of the case as the warm air exhausted from the radiator doesn't get outside the case.

Yeah, so those logs show Link restarting the driver, which is what's causing the connect/disconnect sound. But why is the driver crashing? That's usually because multiple apps are reading the device at the same time; that'll crash the driver pretty quickly.

What is your ambient temperature? And the fans will run at 100% when it hits 40C. You might be able to change that with a custom fan curve but I think some coolers have an override for when it hits 40C; not sure myself as I've never had my coolant get that high. If your H115 settles at 45C max under a Prime95 load, then I don't think we're looking at a blockage situation here. What we have is an issue where your ambient temperature and internal case temperature is actually driving up the temperature of the coolant. It's thermodynamics 101 ... heat moves from warmer to cooler. If the air going through the radiator is warmer than the coolant, it's actually heat up the coolant. You want the air going through that radiator to be as cool as possible; that'll give you the best efficiency. But as long as you have high ambient and internal case temperatures, you'll have high coolant temperatures.

OK. So does it do this when you aren't playing games? When the system is idle? And after you stop playing, does the temperature drop? Another thing to do ... turn the system off and let it cool down. Restart the system and use a CPU-only stress tool (such as Prime). What happens? Watch MB Temp 4 and your coolant temperature. Your GPU should be low and keep your GPU fan off (in the default settings). Because right now, based on the data at hand, it looks like your problem is pretty clearly case air flow.

The chart tells the tale: Your cooler is hitting 40C ... which kicks your fans up to 100%. CPU and GPU temps indicate that you are putting a load on them but not too significant. Of course, it's difficult to tell because you've not filled out your system specs so I'm going to have to make guesses. You mention a Maximus X ... so Coffee Lake. It's very possible that the internal case temperature is rising due to the GPU waste heat which would indicate a problem with your overall case airflow. While identifying motherboard temps is always a challenge, I do notice that MB Temp 4 tracks pretty closely to your coolant temperature. If that's your internal case temperature, that is your problem right there. This is especially true if your radiator is configured in exhaust mode rather than intake; in that case, it would be trying to cool the radiator coolant with air that's over 40C. So, in accordance with the laws of physics, the coolant temperature would increase. Question: is this a new build, a new cooler or anything else new?

Can you show a screen cap when this is happening? Or perhaps the fan speed and coolant temperature graphs over a period of time when this is happening? Because from what I see in your screen shot, it's fine. Is it maybe another fan in your system?

What's the coolant temperature? What does your fan curve look like? Can you post a screen cap of your coolant temps and your fan curve configuration?

In Link, you can enable system logs ... you then specify the location of them.

No, I don't think that it's dangerous for the cooler, just annoying AF. I'm more concerned as to why it's happening too; it really shouldn't be. The only other thing that I can think of is that the driver is crashing ... but why? Can you post logs?

UPDATE: This appears to be a Chrome issue, not necessarily an iCue issue. I was able to reproduce this but only when running Chrome. It happens with other apps only when they have the input focus already, which is actually proper behavior under Windows CUI guidelines (don't steal input focus from apps that already have it). From there, I can assume that Chrome is taking control of the input focus whether you actually have focused the input cursor onto Chrome or not, which is pretty rude.

I am absolutely, positively, 150% sure of that. Does this happen when you aren't running Chrome? By the way, can you fill out your full system specs? First, this is the first I've seen this issue on the forum. One other user repro'd it in a very narrow, lucky use case. So now let's see if we can figure out why it's happening to you. If this was common, I can guarantee you that people would be screaming on the forums here. I understand that it's frustrating. But getting angry with folks that are trying to help out isn't really a good strategy, IMHO. YMMV.

Honestly, it really sounds like Link is restarting the driver. The driver can crash when it is accessed by multiple applications. Do you see this behavior when HWInfo is not running at all? And what kind of GPU do you have? You mention GTX but there are several makers. I ask because EVGA's Precision XOC has caused issues in the past.

The Profile you want should be on top. Not the settings in the performance tab. The first profile in the list is what is applied when iCue starts. It has nothing to do with the order that the settings are applied in the Performance tab. And what support for water coolers was removed from iCue? Where are you getting this? Considering that I have 4 Corsair AIO coolers running in different systems and they are all supported in iCue ... I've gotta say that this statement isn't accurate.

Not yet. What I've found works is to hold the mute button (turn sidetone off), release and then hold again (turn sidetone on). Turning Sidetone off and then back on in iCue also works but that's not always convenient. Also, it happens less frequently if you mute using the button vs. raising the boom.

I use to see this occassionally with previous builds of iCue - on my MM800 and the top bar of my K95. I've not seen it, however, in 3.5. Do you have any log files?

First, just because Corsair doesn't directly respond doesn't mean that they don't see what's going on. As far as this issue ... you mentioned that this happens with TeamSpeak. Do you have the same issue with other applications? I, for one, have programs set to my macro keys as well but I don't see this issue. At all. Ever. So perhaps it's a problem with TeamSpeak? Something that TeamSpeak does that's odd? Is there maybe a command line switch for TeamSpeak that would launch it in the foreground? As you noted, this isn't a common complaint.

I have a z270 and all of my virtualization options are on. I run HyperV VMs frequently.

Yet ... I have an Asus and don't see it. So it's not all Asus boards. But yeah, it's definitely something odd. The more info we get, the better the dev team can narrow down the exact root cause.

I wouldn't say that it's here to stay. Corsair has heard the feedback loud and clear and, from what I understand, they are looking into options to address this a bit better. No word, however, on any kind of ETA. For the record ... I'm with you. I'd like to see my profiles continue unabated, personally. I think a drop down option for how to handle workstation lock would make everyone happy (Current Profile ... Specific Profile ... Lights Off ... Unicorn Vomit). My daughter is a particular fan of unicorn vomit so it does have its fans.

Can you show us a screen cap of your profile list?

From what I understand, they are looking at a solution for this. Technically, it's not a bug as it's "by design" and "to spec". However, I would agree that the spec wasn't quite right in this case. I'd like to see a solution to this as well.

Not a fan of unicorn vomit, eh? :laughing:

I meant it sarcastically ... there's another thread where someone insists that Windows updates never cause issues.

If you have the temp sensors on the CoPro, you can use those as the temperature source for the fan curves. I don't see it in your system specs but I'm assuming that your system is liquid cooled - so the CPU temp isn't really a good source for your system anyway. Internal ambient temperature and coolant temperature would be better - they are less variable, less likely to be "jumpy", indicative of what the fans are actually cooling AND you'd be able to disable CPUIDSDK which seems to eliminate the black screen issue.

Crashes started with a Windows 10 Update?? Say it ain't so!! :evil: