RetroRambler

No it doesn't. Is this kind of logic routinely employed at Corsair to troubleshoot problems or write code? Motherboard manufacturers don't all implement AVX negative offsets exactly the same way (or at all), name the settings the same, or use the same default values, even among different models from the same manufacturer. And then there is (of course) the end user part of the equation.... not everybody is going to tune them their AVX knobs the same way either. Bedevere: "Exactly. So... logically... " Villager: "If she... weighs the same as a duck... then she's made of wood...." Bedevere: "And therefore...." All: "A WITCH!!!" Sorry if it ruffles anyone's feathers, but it fits.

just a guess but maybe need to turn off the checkbox that allows windows to put the USB to sleep to save power.

Not a pro here but the first thing that came to mind for me right off the bat (reading first post) was AVX instructions, which many others did as well. Offhand it eludes me but it should be relatively straightforward to verify whether iCue is or is NOT causing one of your AVX negative offsets to kick in, especially if you have something like HWinfo64 or another util that you can watch individual core clock multipliers on. You could probably even verify which core(s) iCue is running on and watch those clocks. You'll be able to tell when AVX negative offset is kicking in on those cores by the clocks/multipliers they come down to. I've seen that watching stresstests with Prime95, Realbench, etc. when testing my OC stability. Course, it also depends how you've set the BIOS to manage core clocks, using turbo boost max or not, etc. The other suggestion someone gave, try changing your AVX offsets, is a good one -- (change them downwards to a "more negative" offset to be safest) and see if those clocks change downwards by the new amount or not. That will more or less prove whether what you are seeing with iCue is due to use of AVX instructions by iCue on those cores. I'll be honest, the more of this kind of stuff I see, the more I am coming to the conclusion that I should only have iCue installed in a special "configuring my lighting" Windows boot/installation that I only ever boot into when I want to change the lighting scheme and keep iCue far, far away from my general/everyday Windows boot/install.

I can only hope that Corsair does not go so far in the direction of software control that we lose the ability to store a profile we like in hardware and not need a piece of software or a driver to use it. Why? Couple reasons I can think of: 1. Performance hit. I didn't build a 9980XE system because I am OK with donating half of its cores and performance to poorly written software and drivers. It may sound a little different but this is the same basic thing people complaining about the stutter/lag/FPS hit/interrupt overhead in games and elsewhere on their systems are saying. 2. Please don't impose an OS on me. I run that evil penguin OS much of the time and there isn't going to be any iCue for Linux. And I am OK with that, so long as when I am in Windows I can still store something besides the cheesy rainbow swirl default on my DIMM's that will start running as soon as I power up from a cold boot into Linux (or whatever other OS I choose to run on MY system). Which is to say, please do not make lighting control of your DIMM's completely dependent on a driver. Thank you for the improvements, which 3.20.80 seems to have by the consensus of early reports here. I am still on 3.19.120 for now.

Sadly that is most companies cranking out computer parts these days and the "high end" ones are some of the worst. But since so many "reviews" these days are published/written by fanboys who get their hardware for free in exchange for positive reviews (let's be honest mmmkay?) you have to really be careful believing positive reviews. I allowed myself to fall for one -- knowing Corsair's RGB control software has been bad in the past (we have bought and returned Corsair RGB keyboards over the poor software) -- because one reviewer of the Dominator Platinum RGB modules I read acknowledged those problems but claimed the software was much more reliable now. So I got chumped and until 3.19.120 release came out I was starting to think I would never get RGB lighting control working on my rig. For the lucky ones who are able to use older versions of the iCue software and still control/configure their RGB DIMM's, I would definitely say don't use this version. Me, unfortunately, this 3.19.120 seems to be the first iCue version they've released -- or at least the only one I have found of the last several -- that finds my Corsair RGB DIMM's on my Asrock X299 Overclock Formula (OCF) mobo. I hope they fix the interrupts/lag/latency problem with the next release. FWIW the Timings menu section doesn't show anything on my rig either.

While I can see a lot of people are having problems with this version, I did want to share some positive news about the 3.19.120 iCue release that was a first for me. I've been trying for weeks to find a way to get iCue to recognize my Corsair RGB modules on this "new-ish" build so I could change them away from the default "rainbow swirl" lighting scheme. I *DO* have a 4x8GB set of the Corsair Dominator Platinum RGB modules (the DDR4-3000 CL15 ones) in this rig along with an i9 9980XE processor sitting on top of an ASRock x299 Overclock Formula (OCF) board. This was the first version of the iCue software that was able to "see" the Corsair Dominator RGB modules at all. I've tried several BIOS versions old and new. With 1.60 BIOS and SPD Write Enable turned on in the PCH IO Advanced menu, as soon as I installed 3.19.120 it saw the modules. It also saw there was a firmware update available for them and successfully applied it - and I was also able to configure the lighting on the modules. So I do look forward to the updated release that fixes the interrupt issues -- evidently by the reports here I have the right mix of hardware to be afflicted by it -- but I am also glad to see progress is being made in getting the RGB modules to be detected on my board. BTW in case anyone else is wondering, after updating the firmware, back-leveling to 3.18.77 didn't work -- I tried restoring to the "never had iCue installed" backup I started my experiment with and installing 3.18.77 -- but just the updated firmware wasn't enough for 3.18.77 to see them, it had to be 3.19.120 for iCue to detect the modules and allow me to configure their lighting on my ASRock X299 OCF. ICue doesn't show me anything under Timings though, if that is supposed to be showing me Timings specs for these modules I guess that isn't working yet with my ASRock X299 OCF mobo. Glad to at least be able to configure the lighting and will keep fingers crossed that good things happen with updates. Cheers, RetroRambler