iCue and Commander Pro - two temperature sources for one fan

Hello, I am looking for an optimal solution for the fan management. It is nothing new for me that neither option is ideal. BIOS fan management has limited funcionality, third party software has often problems with communication with motherboard which results in some limitations too. Commander Pro looks very promising for me, because it doesn't even communicate with fans via motherboard and the whole loop relies on Corsair's hardware and software (I've got LL120 fans).

So, I consider buying Commander Pro, but I would like to ask about a presence of one function, which is important for me. In the link to thread below Corsair team states that they are working on posibility of a fan control in accordance to two temperature sources simultaneously. Is this function available already? Is it even still considered?

https://forum.corsair.com/v3/showthread.php?t=171225

Well, you have asked. Just be aware they tend not to reveal future plans and if you are looking for some kind of promise it will be implemented by xx/xx date, that's not likely forthcoming.

Not too many things need to be cross-linked to multiple temperature sources. What are you trying to set-up?

I want to make CPU and GPU temperature curves for one pair of fans. There is some software which allows that. For example Argus Monitor (has some problems with cooperating with my ASUS's mobo) or AISuite from ASUS, but it can not monitor my GPU temperature, because it is not from ASUS.

I had motherboards from 3 different manufacturers and this additional features, like fan control, always have to be broken in some way. Something is incompatibile, something does not work properly. They hardly ever fix such things. It is soo annoying......

You will be able to use GPU temp as a control variable with the Commander. There aren't many situations where running case fans off CPU temperature is helpful. I spend a fair amount of time suggesting people stop using it entirely, except for an actual CPU air cooler. If that is your situation, there is a third option.

The Commander comes with 4 10K thermistor sensor ports and the corresponding temp probe wires. These are the native control source for the device and work all the time, software running or not. That makes it really useful for water cooling where it can run from a coolant temp sensor. But it would also work for the theoretical air tower + GPU temperature issue. Both air cooler and GPU are sending the bulk of their heat toward the back fan. The temp sensor there will pick up on either source and more importantly the combination. Case fans move air in and take air out. That air exchange is the cooling. Removing the waste heat vs actually cooling the GPU or CPU. Chip cooling is down to fans on the GPU or CPU.

Oh, thank you for your advice. In fact my problem is not related with temperature. I am sure they will be fine without CPU/GPU fan curve. I just wanted to make temps even cooler. My GPU is not a hot one. Actually it has realy nice temps even after OC. It is RTX 2070 S from Gigabyte and it hits 70 °C in heavy load after OC, never higer so I think it is decent.

Good GPU thermal performance is also one of the reasons why in my configuration using CPU-temperature based, case fans is resonable. The second reason is 9700k from faaaar bottom of silicoon lottery. Hot as hell at 4,9 GHz, all cores. So in my pc case I have to intake fans based on CPU temp. They almost directly supply AIO outlet fans. The other two are based on GPU, and I wanted match one CPU in/out pair with CPU/GPU curve for better GPU cooling, but as I said GPU performs pretty well, so I let go.

With the Commander, you will also have the option to use AIO coolant temp if it is Corsair or even if it is not --- you run a temp sensor to the exhaust side the the CPU radiator. That will mirror the coolant temp and again give you the ability to match fan speeds. You have 4 of those (if needed) and they all can be independent sources for control or just informational.

Also, don't stress the immediate real time reaction to CPU or GPU changes from the case fans. Remember how the process works. Both your CPU and GPU have a metal copper plate bolted to their respective chip. That is where all the cooling is done. The plate conducts the heat away from the GPU/CPU chip. For the GPU, it radiates that heat inside the casing and then it's own fans blow the heat out (either rearward or all over depending on GPU type). For the AIO on the CPU, it dumps the heat into the coolant stream, which is then transported to the radiator and blown somewhere else. In either case and with any cooling method, it's about getting rid of the accumulated heat as to not make things worse. We are all voltage limited on our GPU and CPU. Otherwise you could clock up to 6GHz and run your fans at 4000 rpm. That of course doesn't work.

Where the case fans do help is keeping the environmental temp as low as you can. Whether you are at idle or running a fixed 100% load, raise the case temp or the room temp by 5C and your CPU, GPU, and everything else in the box also goes up by 5C. Your CPU AIO appears to be dumping heat directly out of the case. The GPU is bigger watt device by some measure and the clear heater in most set-ups. This is especially true if you have the open fan type of GPU that blows heat in every direction. Most of us do. Even so, running the GPU at 100% for 5 seconds won't heat up your case. It is a long term battle with intake vs exhaust air exchange that changes the case temperature. I am much more receptive to GPU temp based cooling because the load is usually more even. Any CPU from the last 4 years is going to be really jumpy in its behavior. Linking your case fans to that is pretty annoying.

However, to illustrate the point, imagine you have linked your case fans to GPU temp. You are having fun, playing whatever for 30-40 minutes and the case fans are moving at a good moderate pace as well to match that load. Then you pause the game, go to the title menu, pull up a map, or any of the many things that the GPU does not need full voltage for and it clocks down heavily. With the voltage drop, the temp immediately drops 30C and your fans go quiet too. This could be annoying for any game that does drop the clocks on frequently visited menus, etc. but it also means the fans aren't doing their job at that moment. Just because the GPU load stopped, doesn't mean the air temp inside the case also dropped 5C. If nothing changes and the GPU stays off it will eventually, but on the other hand if they continue to run at the moderate speed, you might actually cool down a bit while you are not loaded. A user preference on management, but ultimately the case fans work on the slow and steady process.

Not too many things need to be cross-linked to multiple temperature sources. What are you trying to set-up?

Coordinating intake fans with both GPU and CPU is commonplace in small volume builds. Also mini-ITX boards tend to have fewer fan headers to work with. This creates a need for iCUE and multi-source temperature control. I need all intake and exhaust fans to respond to a heat increase from either CPU or GPU.

With the Commander, you will also have the option to use AIO coolant temp if it is Corsair or even if it is not --- you run a temp sensor to the exhaust side the the CPU radiator. That will mirror the coolant temp and again give you the ability to match fan speeds. You have 4 of those (if needed) and they all can be independent sources for control or just informational.

 

Also, don't stress the immediate real time reaction to CPU or GPU changes from the case fans. Remember how the process works. Both your CPU and GPU have a metal copper plate bolted to their respective chip. That is where all the cooling is done. The plate conducts the heat away from the GPU/CPU chip. For the GPU, it radiates that heat inside the casing and then it's own fans blow the heat out (either rearward or all over depending on GPU type). For the AIO on the CPU, it dumps the heat into the coolant stream, which is then transported to the radiator and blown somewhere else. In either case and with any cooling method, it's about getting rid of the accumulated heat as to not make things worse. We are all voltage limited on our GPU and CPU. Otherwise you could clock up to 6GHz and run your fans at 4000 rpm. That of course doesn't work.

 

Where the case fans do help is keeping the environmental temp as low as you can. Whether you are at idle or running a fixed 100% load, raise the case temp or the room temp by 5C and your CPU, GPU, and everything else in the box also goes up by 5C. Your CPU AIO appears to be dumping heat directly out of the case. The GPU is bigger watt device by some measure and the clear heater in most set-ups. This is especially true if you have the open fan type of GPU that blows heat in every direction. Most of us do. Even so, running the GPU at 100% for 5 seconds won't heat up your case. It is a long term battle with intake vs exhaust air exchange that changes the case temperature. I am much more receptive to GPU temp based cooling because the load is usually more even. Any CPU from the last 4 years is going to be really jumpy in its behavior. Linking your case fans to that is pretty annoying.

 

However, to illustrate the point, imagine you have linked your case fans to GPU temp. You are having fun, playing whatever for 30-40 minutes and the case fans are moving at a good moderate pace as well to match that load. Then you pause the game, go to the title menu, pull up a map, or any of the many things that the GPU does not need full voltage for and it clocks down heavily. With the voltage drop, the temp immediately drops 30C and your fans go quiet too. This could be annoying for any game that does drop the clocks on frequently visited menus, etc. but it also means the fans aren't doing their job at that moment. Just because the GPU load stopped, doesn't mean the air temp inside the case also dropped 5C. If nothing changes and the GPU stays off it will eventually, but on the other hand if they continue to run at the moderate speed, you might actually cool down a bit while you are not loaded. A user preference on management, but ultimately the case fans work on the slow and steady process.

This is why a good fan controller has a timer for the uptake and roll off of fans based on their temperature sensing. Also the sensing is not from the load but from the temperature of the source.

So this might be a bit of an older thread but I might found a cheap solution to what you are looking for.

Its a programm called argus monitor that I stumbled across yesterday. Made by some tech students in Munich TU. Its free for 30 days and cost 8€ afterwards (13€ for a 3 year update subscription).

It allowed me to do things i previously thought impossible with my mobo, cause speed fan didnt pick it up for example.

You can control fan or pump speed by up to the median or max values from up to 4 sensors combined. set anything to semi passive and undervolt 3 pin outputs and turn down pwm to 5%. my noctua industrial 3000 rpm fan was able to go down below 300rpm.

Very useful for example custom water loops to control the fans by water temp and spin the pump up either if cpu or gpu get to hot.

check it out here: https://www.argusmonitor.com/en/index.php