Jump to content
Corsair Community

New i300 owner here. Is there an issue with my system or are my expectations to high?


Recommended Posts

Hello all. 

 

New owner of a Corsair i300 here. In the week or so of ownership I've noticed some thermal issues arise. I'm curious if this is considered normal behavior for these units or if this may be a case of improper AIO installation, or some other hardware related issue?

 

Out of the box, running the i300 without changing any settings (ICUE default curve, windows power plan set to balanced) and the CPU rarely (if at all) boosts to its advertised clock speeds while gaming or even when not gaming and just doing regular non-intensive CPU tasks. Despite it not bosting, I've noticed that the CPU temps can spike up to 90c at times when just doing regular web browsing, etc. 

 

I changed the windows power plan from balanced to high performance, while leaving the ICUE fan curve set to default and noticed that the system now hits the advertised boost clocks often under load. However within seconds of launching a game or benchmark, the system over heats (Single red flashing right lightpipe) and then the fan actually ramps up (although seemingly way to late)  to a high enough speed to actually cool the system down. 

 

I then switched the ICUE fan curve to extreme, which seemed to do a good job of cooling the system with the power plan set to high performance. The boost clocks remained high when under load, system no longer over heats and results in various benchmarks I used to test it showed performance that is expected of a system with these components. 

 

So back to my main question: what is considered normal operating behavior of the i300? That if I want to get the advertised boost clock speeds, even when not gaming, I have to run the fan at full speed with the windows power plan set to high performance? Or is my only other real option to set the power plan back to balanced and with ICUE fan curve set to default, and just accept the slower performance and constant thermal throttling? 

 

Or is it possible that the AIO wasn’t properly installed on my unit? I am aware that small form factor builds have some compromises built in. I’ve built multiple custom ITX computers, including my last computer (NCASE M1, i7-7700 with 1080 ti) that the i300 is replacing, and not had any issues with throttling at stock CPU speeds. Granted that build had a custom loop and not an AIO solution as well a custom fan curve that seemed to do a munch better job than the default fan curve is doing on the i300. 

Link to comment
Share on other sites

If your ambient temperature isn’t extremely high, and haven’t blocked airflow somehow your findings seem to be different from mine.  I haven’t managed to get extreme cpu temperatures in normal usage or games.  Only extreme gpu junction temperatures (100-106 degrees Celsius), and that during full (or close to full) gpu load.  My iCUE software was installed wrong, not showing products, only sensor values, so there could be variations in what i300 versions are being shipped.

Link to comment
Share on other sites

My ambient temps are about 70f or so with the system sitting wide open in the living room on a desk. Nothing blocking it or next to it. 

Now that you mentioned ICUE, I'm starting to lean toward that being the main issue with the over heating in my system.  I didn't have the issue you had and that ICUE recognized my system, etc. However, I noticed that at random times the top fan wasn't running at all, despite the temps creeping up past 60c while doing very little more than web browsing..  To see if the fan would spin back up under load, I would run a game and very quicky the temp would rise 95c mark and only then did the top fan spin up full speed to cool the system down (while the right light tube would flash). Once the temp got under 60c or so, the fan would completely stop spinning again, and the process would just repeat itself. 

After that I decided that I should do a complete icue uninstall and to make sure it got rid of all the files, I used the windows clean uninstall tool. Then I rebooted and reinstalled most up to date version of icue.

For about an hour after the reinstall of icue, the system seemed to be working as it should. The fan was running at low RPM at idle, keeping the temps in the high 30's to low 40's. So I started playing the ascent and an hour or so into the game, I got the flashing right light tube warning again. Checked Hardware Monitor Pro, and some of the cores got up to 100c while the GPU core never got higher than 83c. 

Not really sure what to do about this. I really like the system, but I'm not really interested in doing an RMA for a system I've only had for a week that cost more than 5k. And returning it and buying it again also sounds like a hassle, that's why I'm hoping there is an easier solution to fix this. 

Link to comment
Share on other sites

Unfortunately, it seems to be getting worse. Now I get two red flashing red on the right, followed by two flashing red on the left. I decided to export the icue logs to see if there is something in there and I noticed these areas that seemed strange. I'm not very experienced with reading logs like this, but I noticed the multiple lines with the temp and pwm showing no value, as well as the malfunction: GPU thermistor fail message.  Is this more of an indication that there is a hardware failure? Or is it still possible this is an icue caused issue?

spacer.pngspacer.png

Link to comment
Share on other sites

I'm half asleep right now reading this.  I personally have kept the windows power plan to the default Corsair Balanced setting.  After that I would reinstall iCue with both options to clean install removing prior presets selected see what happens, and yes always use custom75+%/extreme fan settings

Link to comment
Share on other sites

Also not sure why this really makes a difference but download iCue from the i300 product page not the iCue software page.  I have my reasons why ro do this just to tired to explain atm.

Edited by zguy85
Link to comment
Share on other sites

Hello kilroy81,

did you try zguy85 suggestions and if so, did it improve your system' temps ?

In the meantime I came across the discussions about the washer mods. Do you think it is something feasible in the C1 (due to the restricted access to the MB) ?

Link to comment
Share on other sites

I have uninstalled icue a few times now and reinstalled it with the version found on the Corsair's site under the downloads area. To ensure that there was no files left behind when I uninstalled the ICUE software, one time I used windows uninstallation tool and the second time I used revo uninstaller. 

The clean ICUE version did fix the issue with the top fan not spinning at all (at least for now). However, when using the default fan curve profile, there is no change in the system overheating (even with windows power plan set back to balanced). I'm not only getting the overheating light pipe warning, but I'm also getting the pump failure light pipe warning now too. But only when gaming on the default profile (after about 20-30 minutes or so). 

I can and have ran the fan curve at extreme (which seems to just be 100 percent on, and no actual curve is present... at least for me). And yes, it does keep the system really cool and no over heating happens with this setting turned on.  I have also ran the system at 75 percent as you suggested zguy which also worked fine.. When running both of these settings, there is no over heating error or pump failure errors present, even with prolonged gaming. 

But I have to be honest here.... I didn't spend more than 5k for a system to sound like a small jet taking off in my living room. If corsair isn't going to allow us to create a basic fan curve... then they need to make sure the default one actually meets the systems temperature needs. I can't think of any good reason for the default fan curve to wait until after the system is overheating to run ramp the fan up. In fact, I'm pretty certain at least in my case, the default ICUE fan curve doesn't actually ever bring the fan speed up high enough to cool the system under gaming load.. And that it just keeps gradually getting hotter until the system watchdog overrides ICUE to cool the system down. 

On a side note, I'm still curious about the logs saying that I have a malfunction: GPU thermistor fail. Ironically, from what hardware monitor has sown me, the GPU temps have stayed well within normal operating limits, and it's only the CPU thermals that have been having issues. 

I opened a support ticket yesterday and now I'm just waiting to hear back from corsair about my situation. 

  • Like 1
Link to comment
Share on other sites

I agree with you. The default fan curve should assure the computer works without overheating. We know Corsair wants it to be as silent as possible (which I really like), but a silent computer that overheats is not useful.

  • Like 1
Link to comment
Share on other sites

Unfortunately, the top fan stopped working again. While sitting mainly idle with a screen saver running, they system over heated, flashing the light pipe led message suggesting pump failure and overheating. Checked the temp in hardware monitor pro, and the system was idling in the 70’s and had shown a couple cores hit 100c again. 
 

Even with the fan not spinning, at idle I don’t think the system would climb up to 100c in a room that is just around 73f if the coolant is still moving through the system. So I’m leaning toward the pump failure error message being correct. But that wouldn’t explain the top fan not working, etc.  Maybe the icue software is causing the pump and fan to stop? 

 

Link to comment
Share on other sites

3 hours ago, jupiterstar said:

I agree with you. The default fan curve should assure the computer works without overheating. We know Corsair wants it to be as silent as possible (which I really like), but a silent computer that overheats is not useful.

When I emailed Corsair support about this several times they kept completely ignoring the issue, while my understanding of what was going on was improving and improving.  When the 2 weeks return-window had passed he congratulated me on the sales and summarized the whole matter by saying: “Because of the limits imposed by the unit's compact form factor, Corsair Ones will tend to run on the hotter side.”  And that was after pointing only at the 71 degree gpu core temp in the screenshots I had included, not the 106 degree gpu memory junction temps I wanted to discuss.

Link to comment
Share on other sites

Sounds like you're experiencing all the problems we have all had with only solution setting fan 75-100%.  I must be Def because of the actuall jets I do work on lol so doesn't bother me.

Does sound like you might be having additional complications with the top fan.  This is when set to custom/extreme or only default?

I am really HOPING corsairs first set of BIOS/iCue updates makes some improvements.

I completely agree to the T with what you said about default setting not properly cooling or acting too late when already overheating.  Well put.

This is the biggest setback for the system and could easily be addressed if Corsair would just address these issues to the team that writes the software.

Actions speak loudly waiting to see what they decide to do otherwise I'll keep my gaming headset on to drown all noise till then.

Let's fix this Team Corsair! 

Edited by zguy85
Link to comment
Share on other sites

One of my main complaints was the lackluster gpu fans, until I now know they don’t really cool the gpu mem as the grill they’re on stays relatively (50-60 degrees) cool, even under full load.  Evidently Corsair support doesn’t want to go into detail too much about weaknesses of their design, otherwise they could have just stated this fact.  All major 3080 TI designs have their memory chips way closer to the gpu than those outer grills.

Edited by User 1
Link to comment
Share on other sites

Just to add my experience on this

I just get my corsair one I300 several weeks ago, I check also the temps and the biggest concern is around the memory that is running hot (around 105°C).

During my test I noticed that the small GPU fan start when the memory temp reach about 95°C and the GPU temp about 60 - 65°C. At full load and with a stable clock the temps stabilyze around: 105°C for the memory, 80°C the hot spot and 72°C for the GPU (with the cooling set on extreme).

It seems that the cooling configuration is set to follow the GPU temp or the hot spot but it should be the memory. For this we need to wait on Corsair for a BIOS or ICUE update as mention by zguy85.

 

I never encounter any overheating issue because I always run with a custom fan setting at least at 65% because or higher. On my previous A140 the SSD was always on the hot side with the default fan setting (I noticed this after the ssd died after 24 months).

I hope Team Corsair will hear us and work on a fix soon.

Link to comment
Share on other sites

It would be great if they let us control the GPU fans manually as well like previous models.

Link to comment
Share on other sites

On 3/24/2022 at 1:07 PM, Jean-Mi said:

Just to add my experience on this

I just get my corsair one I300 several weeks ago, I check also the temps and the biggest concern is around the memory that is running hot (around 105°C).

During my test I noticed that the small GPU fan start when the memory temp reach about 95°C and the GPU temp about 60 - 65°C. At full load and with a stable clock the temps stabilyze around: 105°C for the memory, 80°C the hot spot and 72°C for the GPU (with the cooling set on extreme).

It seems that the cooling configuration is set to follow the GPU temp or the hot spot but it should be the memory. For this we need to wait on Corsair for a BIOS or ICUE update as mention by zguy85.

 

I never encounter any overheating issue because I always run with a custom fan setting at least at 65% because or higher. On my previous A140 the SSD was always on the hot side with the default fan setting (I noticed this after the ssd died after 24 months).

I hope Team Corsair will hear us and work on a fix soon.

As I just explained in the post above you those fans don’t directly cool the memory, since those chips are much closer to the gpu chip itself.  Just look at any 3080 TI card (the one in the i300 is from Palit) and you can see where they are placed.

Link to comment
Share on other sites

I don't know ya'll.  Nothings exactly how we want it and purchasing this system either has limits or further from normal requirements due to the small form factor.  I know it's expensive but I IMO.... overalll, its best in class including design and I think we all could agree on that otherwise you probably wouldnt be reading this on such a niche forum. 

My experience with i300 has been pretty solid so far.  After delid I pushed the processor even further and have a solid stable system with no issues arguably other than the top fan having to be custom set. Performs in and out if gaming with no issues currently. Maybe I got a lucky one?  Let's see how Corsair moves forward, let's see if they respond with action letting us know we are heard.

Edited by zguy85
Link to comment
Share on other sites

No question that C1 is best in class among pre-build SFF's. Not that there are many more out there but still best in class.

The issue for me is that, by design, if you are unlucky with your system (and lately a few people have been complaining), you are stuck with a system that is hard to work with yourselve (locked BIOS, custom AIO, etc).

And Corsair support seems to be rather unresponsive when bad luck strikes... imagine having paid a small fortune, having problems with you system, and not getting the 24x7 support that Corsair preaches in their website for C1 owners...

This is what is driving me away from buying one. And as I said previously, I would really love to get a top spec i300.

Link to comment
Share on other sites

Having just bought a Neo G9 I am pretty happy with how my i300 performs.  I didn’t like seeing those high gpu memory junction temperatures, and hated the unresponsive support I got, but as a “tiny” pc it performs well.  I try keeping the junction temps below 100 degrees, and sofar manage 110-120 fps in both God of War and Horizon Zero Dawn, highest settings.  This is on the Neo G9.  Pretty awesome if you ask me.

I work at a research and product development department and when I told one of the senior Electronics staff about those 104/106 degrees junction temps he wasn’t much put off.  In electronics under stress these values he considered normal. (we deliver products that have to endure heat on roofs etc. in the hottest climates)

I read many articles about lowering gpu-related temperatures, but I found those people were nearly always talking about the same gpu products.  Reading that Palit was complimented in a review about their cooling solution in their GamingPro card (similar/same as card in i300) I now guess that the slow junction temperature built-up to 104/106 degrees comes from slowly over-heating the gpu back-plane.  It would be great if people would have taken the gpu cooling apart but for now I have rested most of my desires for improvement.

I have put my attention to other matters and now for instance backup off-line from a Hiren backup USB-stick (Macrium Free or AOMEI) to 300 Euro 18TB Toshiba hdd.  Backing up the entire 1.2 TB filled NVME is done within an hour.  I have Kasperski, which is the best AV, but not without risks nowadays, so I try to have many angles covered.  Also don’t have the pc 100% connected to the internet as I use iPads etc. for most things.  Over-heating risks of a gpu is just one of many concerns nowadays.

My concern with buying an i300 right now would be the price drops of gpu cards and ddr5, which should result in cheaper gaming) pc’s.  Good luck with whatever you choose.

  • Thanks 1
Link to comment
Share on other sites

On 3/25/2022 at 3:04 PM, User 1 said:

As I just explained in the post above you those fans don’t directly cool the memory, since those chips are much closer to the gpu chip itself.  Just look at any 3080 TI card (the one in the i300 is from Palit) and you can see where they are placed.

Thanks for the information about the video cart manufacturer. On the picture (in an other topic) the heat pipe should be on the memory chip and should transfert the heat to the cooler with the fan. That's why I was thinking the cooling configuration should follow this sensor and not other.
Open question, that's mean do we need a mb bios or a gpu card bios update ?

On the other hand I was looking to undervolt the CPU because I had some really short spike at 100°C on some core during gaming. After tweeking some parameters on the bios I keep now all core around 85°C.

Next I will try to change the main fan that have better performances.

Link to comment
Share on other sites

  • 1 month later...
On 3/21/2022 at 7:04 AM, kilroy81 said:

Unfortunately, it seems to be getting worse. Now I get two red flashing red on the right, followed by two flashing red on the left. I decided to export the icue logs to see if there is something in there and I noticed these areas that seemed strange. I'm not very experienced with reading logs like this, but I noticed the multiple lines with the temp and pwm showing no value, as well as the malfunction: GPU thermistor fail message.  Is this more of an indication that there is a hardware failure? Or is it still possible this is an icue caused issue?

spacer.pngspacer.png

I'm also a new i300 owner and have the same issue. I enjoyed about 4 days of normal performance, and then one day the fan didn't turn on until the system had already begun to overheat. Extreme and Custom fan curves in iCUE work normally, but under the Default curve, it does nothing until the CPU heats up to 90-100c, and the fan goes to 100% and LED panels alternate between blinking red on right or left sides, indicating Auto Detection of Pump Failure or Overheating.

Looking at my logs, I see the same thing as OP. Normal default fan curve performance, followed by a string of errors, followed by a default fan curve set to 0 PWM across all temps.

After going back and forth with Corsair support on this, and twice doing a clean install of iCUE with no fix, the final word from support is:

"This is an issue that we are aware of at the moment. Recommended that you change to a more aggressive fan curve or just leave it at a constant percentage. There may be a fix pushed through BIOS later on for this issue."

  • Like 1
Link to comment
Share on other sites

17 hours ago, Rakshasa said:



"This is an issue that we are aware of at the moment. Recommended that you change to a more aggressive fan curve or just leave it at a constant percentage. There may be a fix pushed through BIOS later on for this issue."

There may be a fix pushed through the bios later on…. That sounds very reassuring. 😂

I just ended up returning my defective one. Bought a replacement i300. Didn’t install any updates for icue and it’s working fine so far. Default profile seems to kick on at the appropriate times (like before the system melts itself).
 

I’m nearly 100 percent certain this is an icue software issue and not related to the bios. icue is an shockingly bad application. It’s a shame that it has such a high level control over the system. 

  • Like 1
Link to comment
Share on other sites

Posted (edited)

 Well that's the first official response I've heard of regarding this issue.  An issue that we all knew about, come to find out they knew about, was brought up numerous times on this forum, and not brought up once from Corsair.  This is the BS lack of transparency and complete breakdown of communication between Corsair and their Customers that is an absolute fail and doesn't help at all with trust in the company.

Corsair please learn from this example not to point fingers but we all want you to do better.  A win for your customers is a win for you period!

I also want to add I love how they state pretty much we MAY come out with a fix for a known discrepancy for you $4000-$5000 premium purchased designer computer.  If you know there's a problem fix it otherwise your telling your customers they are not the priority.  No debating that.

Edited by zguy85
Link to comment
Share on other sites

I concur the issue is most likely a buggy iCue. I no longer use iCue, but should the need ever arise, I find version 3.38.88 relatively better in ramping up the top fan earlier as the coolant temperatures rise.

As regards waiting for Corsair BIOS updates, i gave up and used AMIs afuwin tools to overwrite their BIOS with the latest from Asrock for my i182 and replacing the motherboard altogether for the A100. Hoping to get Aquacomputer's Quadro to handle fan control.

Lesson learnt for me is to keep away from Corsair pre-builts. The cases are good, but their restrictive software lets them down big time.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...