c-attack

I don’t know what to suggest at this point. The Commander XT has independent rgb channels, unlike the older series circuit controllers. If one of the fans were bad on the rgb side, only that fan would be affected. I would not expect any controller to have a simultaneous 6 port fail without some type total power delivery problem affecting everything. These H100x Elite RGB models have been tricky with their rgb integration into CUE, but nothing tricky about the SP-Elite fans. You connect it to any RGB port and it should be auto-detected as 8 LED fan. I am not sure where that leaves you. It might be time to contact Corsair Support, but I’m not confident they will come to a resolution either. However, I am curious about the new CUE version. Did you update to CUE 5.15 with this new UI before you hooked up the Commander XT to the fans and pump?

Will it let you set up manually? It’s “8 LED fan series” for the SP-Elite. For the pump it needs 16 LEDs. Not sure if it will offer H100x as a choice but Hydro X -> XC7 will generate 16 LEDs too.

Don’t let the Commander Pro run from its presets. They are not really functional and use cpu temp on a curve designed for water temp. This keeps the fans at 1500 rpm all the time. You definitely need to use a custom curve.

I am using a Commander Pro for the same purpose and have not seen any deviant behavior on the current release, although there have been lingering control issues elsewhere the last several months. 25-27C is a normal idle coolant values, but be aware your maximum is going to be somewhere in the 35-40C range, so that curve may not be what you want. I can’t see the sensor choice under the graph, but the yellow line indicates the current value is 50-something and thus the fan speed is higher. When you create a new fan curve on the Commander Pro it always defaults to cpu temp. You need to get off that as it’s is not an appropriate or necessary control variable for water cooling. Click the drop down menu under “Sensor” and find Commander Pro Temp 1. It’s probably all the way at the bottom. Then in the lower right corner of the graph are 4 shape tools. Click the one in the left. This will give you a decent starting point for a water cooling curve. there are no automatic rules for this. Everyone’s normal idle temp will be based mostly on room and case temp. This also means you will see shifts based on condition changes. However, I am going to assume you are running 2x360mm rads and both cpu and gpu in the loop, thus a +10C increase during long term gaming (highest wattage) is probably about right.

You still must set CPU fan to 100% or Full Speed (Q-Fan) for the H80i GT as that is a power supplying connection unlike most other models. It will work with the current CUE software, although I am concerned the Device Memory Mode is going to cause problems on these older devices. My recommendation is to use the CUE 4 download (prior generation) to manage these older devices when there is no other Corsair hardware present. You also can skip cue entirely by putting the pump power on W_ Pump and the two fans on cpu/opt. Then in the BIOS set the fan delays to maximum for cpu fan (8.5 seconds). They will run cpu temp and I spend a lot of time explaining why that’s inefficient, but in these circumstances it’s a viable means of control.

Commander Pro Temp # is the data coming from the 4 temp sensors on the end of the device. A connected 10K thermistor for air temp or specialized one for water temp in custom loops is the normal purpose. What is connected to your Commander Pro temp register(s)?

W_Pump or AIO_Pump have no special properties on most Asus boards aside from a higher current capacity on premium boards. There are a CHAssis fan header preset to 100%. Most of the time this is wasted as nearly all AIO units these days use a SATA connector for power. The H80i GT was one of the last units not to do that and does take power directly from the motherboard. That 3 pin connector needs to connect to either AIO Fan, W Pump, or the CPU Fan connector set to “Full Speed” or 100% in the bios. If you use W pump or AIO pump, then you must disable the cpu boot protection system. Go to the Advanced BIOS (F7) -> Monitoring -> CPU fan RPM. Hit enter and then change it to Ignore to disable the boot protection. No matter which MB fan port you choose, that 3 pin must be connected for power. It will slide onto any 4 pin header in the correct orientation. The motherboard has an auto-detect for DC vs PWM or you can manually set it in the bios. The AIO should always be set to PWM and is meant to use a full 12v with speed adjustments handled by the internal controller.

Try shutting down the PC and flip the PSU off. Wait 30 seconds or hold down the case power button for 10 seconds to drain any residual power. Flip the PSU back on and start normally. This should put the Commander XT into a hardware level check for connections. This is a H100x RGB Elite, correct? All four fan rgb and pump rgb are connected 1-5? If the pump is #5, try moving it to 1. If the pump is 1, move it to 5. This should not matter at all since the rgb ports are independent, but we’ve seen several users report oddities with the pump rgb header on this model.

Coolant temp is the proper control variable for the pump and radiators fans. The cpu is cooled conductively as heat is physically transferred from lid to cold to water stream. That rate of heat transfer doesn’t change with fan speed. The fans help dissipate heat from the liquid. If you don’t remove heat from the liquid, then its temperature will increase over time. Liquid temp is the minimum possible cpu temp at zero voltage. +1C to liquid temp = +1C to CPU temp. If you start a stress test and the cpu temp instantly hits 90-100C+, you have a contact issue or in some cases the voltage is too high as a result of overlocking. If your fan speed is too low or the liquid flow too slow, you get a more gradual change. The counter starts at 75C, but then steadily creeps up to 90C over time. Based on your description, this going to be a contact issue. It can be really frustrating, but you need to remount again. The missing cpu package temp for AMD is a reported issue in this release. There were significant UI changes as well. In your screen shot above you are in “Device Memory Mode”. This is the new name for Hardware Mode or the device behavior when CUE is not running. Unlike before where leaving that screen return you to software mode, once you check the toggle to enter Device Memory Mode you stay there until you shut down the PC. Your control options, lighting, and choices will be more limited in DDM mode. Set the behavior you want for boot/shutdown/BIOS, then toggle DDM off and use CUE in its normal software mode.

C:Users/(name)/App Data (hidden)/Local/Corsair

You will never see a meaningful difference between the QX and RX on a multi-fan radiator setup. If you were running a 120mm radiator fan test you could create some differences at high watts and high fan speeds. On dual 360mm radiators (?) all of that yields to higher surface area and that’s the point of maximizing your radiator space. You can run whatever fans you like and speed becomes much less important. The RX cost less because it’s a lower tier model. Don’t get caught up the airflow and pressure spec game. Those are marginal differences. If you are running single sided fans on a 45-60mm radiator with a relatively high FPI (18+), then there could be a measurable difference (1-2C liquid temp), but if you’re using normal radiators or standard 30mm AIO rads it’s not going to matter.

It appears there may be an issue with CUE Link XD5 devices "resetting" their custom cooling curve at random intervals. This is a new system and brand new hubs and hardware with a clean CUE install. Each time I make a custom curve for the XD5 Elite, it resets back to the default custom cooling curve at a later interval. It does not seem to happen with fans on the same CUE Link circuit. Anyone running a custom curve with a CUE Link XD5 may need to pay attention. That default curve is 20% up to 45C. That's only 800 rpm on D5 pump and will lead to temperature problems when loaded.

There is going to be a lot of confusion with these changes. Device Memory Mode = Hardware Mode (Lighting and Cooling). They have changed the mechanic to work a bit more like the current keyboard lines with active saving required (at least you get a prompt), presumably as a way to tidy up or create more UI space. The functionality has not changed, but if you have older gear it's going to feel a bit strange. You go into DDM to set the behavior you want when CUE is not running. Toggle back out when you've finished your settings. Make sure you "leave" Device Memory Mode. Unlike the prior Hardware lighting/assignments etc where it goes back to software mode when you tab out, you must deactivate DMM when leaving or it stays on. So if you are running through your devices and don't toggle it off, you'll wind up with half your gear in hardware mode and half in software mode. I have to admit I liked the prior system better where the other devices previewed their own Hardware Lighting while you were changing one device. Everything went back on its own. Now in order to see that, you need to toggle all your devices into DMM and then all of them back out. At least all the CUE Link devices act in unison and perhaps that was the basis for the change.

You can’t roll back with CUE 5. It’s server based and always pulls down the current version, regardless of which installer you use. Not clear what those posters are referring to unless they went back to CUE 4.

You likely need to do a clean install. Vanishing from the OS menu strongly suggests registry corruption and this is one of the more common issues with CUE 5. Try the clean install directions below but export any valuable profiles first — if you can. This does erase everything. https://help.corsair.com/hc/en-us/articles/360025166712-iCUE-How-to-perform-a-clean-reinstallation-of-iCUE It’s also possible you won’t be able to install the program and it will be grayed out or the operation won’t succeed. If this happens, try one of the following: 1) Reboot to safe mode and uninstall + registry clean; 2) Use a professional software removal tool like Revo. The free version is sufficient and this also does the registry clean for you making it a better choice if you are not overly Windows savvy on getting into safe mode.