c-attack

Did you recently update to CUE 5.14? This version has the new UI changes but critically for CUE Link users expands the device capacity from 14->24. This caused the fans to be renumbered and it's not going to match your previous configuration. Anyone with CUE Link gear needs to do a clean install to clean out any old config data. Make sure you export your profiles first so you can import them back later. Might need to reapply some lighting effects to specific fans or devices. https://www.corsair.com/us/en/explorer/diy-builder/blogs/addressing-over-current-protection-issues-expanding-the-icue-link-system-hub-to-24-devices/

Yeah, don't get too worked up over the CUE CPU temp reading or it's % load either. If you want detailed CPU temp info, use HWinfo. While we are talking about it, CUE's core temp numbering is incorrect for Raptor Lake and has been for a long time. I don't even leave my cores up on the dashboard anymore. You've already figured out it provides a different reading for package temp than HWinfo. Use HWinfo when trying to accurately asses core behavior, both for the temp and the frequency/VID/Voltage values you need to make sense of it. This is normal behavior these days for Intel. One core spiking is some minor task being executed and the CPU is doing what it's supposed to do -- respond rapidly. A slightly more complex issue is when all the cores respond to these minor tasks and this can sometimes happen with some manual overclocks. The CPU behavior has become very sophisticated and the brute force manual OC is not a great option. However, the temps are still fairly simple. When a core sparks up, it will warm up. No amount of pump speed, fan speed, or radiator real estate that prevent it from heating up as voltage is applied. The CPU IHS transfers the heat to the cold plate on the block, which in turn transfers to the water. The water transports heat to the radiator for dissipation. The cooling system is a waste heat removal service. It's fun to set up a massive case with yards of radiator length, but when your CPU is sitting at idle and only punching 17-23W, it doesn't make a difference whether you are running a 3x3m square external cooling system or a little 120mm radiator. The cooling system only has 20W to get rid off and all CPU temp results are voltage and conductivity alone. If we pair your motherboard and CPU with any cooling method or brand, the idle temps will be effectively the same aside from minor differences in thermal conductivity on the cold plate. It's when you load up your 600W GPU and 300W CPU that you can start showing off and blowing that heat all over the room while maintaining excellent hardware temps. If you get the mount wrong, you will see CPU temps 15-20C higher than expected. They will instantly hit that 95-100C mark when they should not. If you get the mount just a little imperfect, you might see a large disparity between core temps when at full load. However, this is no longer a reliable metric as these 8+16 CPUs all have a health 5-9C core differential between best and worst cores. Too many cores, too close together. There is some heat sharing and some of them will be better/worse for it. So a slightly imperfect mount might run just a touch higher than expected. Don't use idle temp to measure that. Use a full stress test with a linear load. I usually recommend CPU-Z "bench test" because it is under the 253W Intel power limit so you won't bang off the ceiling and downclock while providing an even load. This is a good way to test multiple things, like pump speed vs CPU temp or your heat transfer.

There is definitely something going on with fan/pump curves and the "new" Device Memory Mode as well as some glitches when running from normal software mode. There seem to be a lot of shades of this and new reports are coming in frequently for different gear. At this point I am just trying to categorize into controller type -> Does X, when I do Y. I am a little short on workarounds right now. Older devices still show a static color option under custom in DDM. My CUE Link stuff has no custom options. Presets only. I believe this to be an error as the CUE Link hub has higher hardware capabilities than most other devices and at a minimum there needs to be a static color option so you can do 0,0,0 (off).

You need to exclude Corsair internal devices like fan controllers, AIOs, and usually PSUs from HWinfo monitoring. RAM is usually OK. Both programs will get locked up trying to access the same device at the same time. I am trying to remember whether the 5000T came with a Commander Pro or Lighting Node Pro set up for the case RGB strips. Checking the CorsairLink & Asetek support option may work in certain cases, but your AIO is not Asetek and it doesn't resolve the monitoring conflict with other hardware.

Ok. So 1-3 are the QX rear temp sensor values. 36C intake air (assuming fans are on the front side of radiator) guarantees you will have coolant temp of 36-40C at idle. If intake air temp is 36C, then minimum coolant temp is also 36C. Most users float 4-7C above that value depending on unique case conditions and cpu idle power levels. So you’re within normal allowances for your intake temp. Just wanted to make sure there wasn’t a pump or flow issue.

Coolant temp? Or cpu/gpu temp?

Short Version: 1 & 3 -> Anyone with a pre-existing CUE Link hub that was set up prior to CUE 5.14 and the FW expansion to 24 devices needs to do a clean install of CUE. This has been discussed at length elsewhere, but in effect the fan enumeration changed when it went from 1-7 and 8-14 to 1-12 and 13-24. Lots of things tied to that identification. Export all your valuable profiles, clean install, then import your profile exports. Things will not work right until you do. https://www.corsair.com/us/en/explorer/diy-builder/blogs/addressing-over-current-protection-issues-expanding-the-icue-link-system-hub-to-24-devices/ 2) You should be able to save basic custom patterns like static, wave, ripple, etc. to DDM mode and the CUE Link hub and half the point of DDM is make it work like the keyboards with an active save mechanic (whether you like that or not). I’ll try and test some combinations later today. What’s your hardware? CUE Link AIO and QX fans?

Yes, unfortunately this seems to be the only option right now. There is another thread where we worked through it along with some additional issues, but if you have new fans or need to configure a Commander Core or XT, you likely need to uninstall 5.14, download 4.33 from the website, run the lighting wizard to auto detect, then reinstall CUE 5.14 over the top. There should be a correction in the next release, but for the moment this is an issue.

Yeah, unfortunately I’ve seen and reported that one as well. My tests were for the XD5 Elite, but not too surprising the CUE Link AIO does the same. I believe the hub is using CPU temp as the DDM control variable even when the custom curve specifies a native liquid temp from the AIO or XD5 Elite. So for your original issue, you create a DDM mode curve based on coolant temp (20-45C range), but it substitutes CPU temp as the sensor causing you to run at the high end of the curve at idle. If you set the actual Quiet fan preset in DDM mode, it seems to behave itself. This is probably another opportunity to remind Corsair the invisible presets don’t benefit anyone. Even if they don’t want us to be able to modify them, show the graph data “greyed out” so users can understand the behavior.

I am trying to get a look at those CUE Link Temps (orange). The ones in the low 30s are normal. The 4th one is 41C? If that is liquid temp, then that is warm. Be aware there is often a default CUE alarm set to flash fans red at 40C liquid temp. You can deactivate it and nothing happens at 41C except the CPU is +1C warmer. Regardless of the above, other users have reported some random CUE Link pump stops in the past. I am not sure we know why it happens, but most often it is a one off event. I guess just keep an eye on things.

You probably needed to clean out the registry files as well. For some reason Corsair has removed that part from their posted clean install directions. Maybe too many people deleting the wrong thing. At this point I would use to Revo or some other free professional software removal tool to clean out any remaining files. It should give you elevated permissions so you don’t have to back into safe mode again.

Definitely a necessary step because of the file corruption. Anytime you see it grayed out in the Windows menus, things have gone wrong. That is not likely specific to v5.14 and we have seen a steady stream of these type of install/registry errors since CUE 5 debuted. So the flashing red is after doing the clean install? If before, I am not surprised and hopefully it’s now cleaned up. If after, keep an eye on things. One of the difficulties with the new Link Tech is the device is entirely dependent on the software to supply power and control. If it thinks there is no device there, it might not supply power. I haven’t seen any other loss of power issues reported this far. One thing I can confirm is there is an issue with custom curves on the XD5 Elite models resetting to a “quiet” preset that keeps the pump at 800 rpm — too slow for any load situation. However, the same behavior on the AIO could only drop the pump to its 2000 rpm minimum. You can’t overheat the AIO at 2000 rpm. The pump definitely needs to stop and that’s not something you can do in the software. I don’t think there is a connection.

This is an actual overheat warning. Flashing red fans, especially before Windows and CUE loads is a hardware temp detection feature. There's not enough information here to guess why. However, if you have a CUE Link AIO you need to do the clean install to make sure all devices are detected and in proper sequence. If you have an Elite Capellix Model, most of this discussion doesn't apply to you and we still need to figure out why the AIO alarm is going off.

It seems like the AIO is actually overheating. That’s not something g anyone else has reported, but possible if it sets pump and fan speed to absolute minimum. Definitely possible if it’s not making the pump run and that is no longer a guaranteed factor without its own SATA cable.

That’s where the CUE Link 4 way splitter or Y-Splitter come into play. You don’t need to set it up like a traditional series circuit. All paths need to come back to either side of the hub.