Well, the maximum GPU temperature (in C) is 89 C so your reading of 84 C is within the thermal limit for a RTX 2080 Ti.  If you bumped up the fan speed from 61% to 80 or 90%, that should cool things down to a more prudent temperature; but that's just me.  You're not at the point of any damage or throttling down; I just prefer not to run GPUs any hotter than necessary.  In other words, your choice--leave things alone or bump up the fan speed for cooler temps but more fan noise.

What's interesting is that Titan Quest Anniversary Edition is a 2016 game that originally was released in 2006.

I'm very surprised that it is giving your RTX 2080 Ti such a workout. 

Enjoy the game! 

I'm suprised as well. In fact when I run Cyperpunk 2077, the GPU temp and fan speed are only slightly higher than those when I run Titan Quest. However, the requirements of these two games are quite different. The only problem when I run Cyperpunk is the random extremely loud fan noise. This is known issue in Aurora R8. I have asked dell to figure out a solution since I have the warranty.

I'm just curious that why these two different games can give the GPU the similar workload.

I personally think there is something in the most recent two nvidia drivers causing an anomaly as since I bought cyberpunk and installed them “every” game I play has my intake fan going mad and it’s annoying me

What FPS do you get at the moment?

What is the frequency of your monitor?

Do you have GSYNC off?

Just a wild guess, your GPU is heating up too much the inside of the case at that temperature (84°), the AW sensor is detecting it and revving up the case fans, this exhausts the heat inside and slows down the fans subsequently.

Can you monitor the temperatures and the fan speeds in AWCC Thermal Profiles ?

@Scallop 

Rgr, see what you mean now you posted those graphs in the other thread.

I guess cards react differently when they go past 83°, afterall that is the default software setting for throttling before overclocking, I saw it in AWCC and EVGA Precision X1 on my rig.

Running a GPU over 80° is not what I would recommend. I would try to get the GPU at least under 75°. Like @Doghouse Reilly said, max out your case fans to get cooler air inside and around your card.

Hence my questions, FPS, Monitor Frequency and GSYNC. If you are pushing your card beyond your physical FPS limit (monitor), then Vertical Sync it, avoids tearing and lowers GPU temps.

Well in fact the first time that the fan noise showed up is about one year ago. So I think this is not related to the drivers. And not only Cyberpunk, others games like ARK and Monster Hunter can trigger the fan noise.

Following along with this thread. Also have the 2080 Ti in my R8 and in the last week or so have started getting massive fan rev spikes out of the GPU.

I can feel hot air coming out of the back of the case/GPU when it happens, but I don't understand why it has started happening.

I got Cyberpunk at release, and it runs really nicely. This rev spiking only started within the last week or so and with Cyberpunk it happens every ~10 seconds now so it's super annoying. Even started happening with some older games like Farcry 5 in the last week.

I replaced my front intake fan and have tried running it hard steady state but it makes no difference on the GPU fan going into overdrive. 

Someone up above mentioned could be an nvidia driver change, maybe that's it? Planning to open a ticket with Dell to rule out the GPU being a problem.

https://forums.evga.com/m/tm.aspx?m=2931484&p=4

Here is a link.

Those people all have similar problem. They have different model GPU or same GPU from different manufacturers. It seems they all fix the problem by repasting the thermal paste. Some people post with pictures so that you can have a look and know why repasting is necessary.

In fact, you can find some similar posts in other websites like Reddit.

I guess this is the best solution. 
You can have a try if you don’t have the warranty.

@markburv 

I didn’t notice the FPS. But I didn’t feel any change of gaming performance. So I guess the FPS is not affected. I don’t play games a lot. Most of time, the GPU temp can be just 40-50. That’s totally fine. Since I have the warranty, I know Dell will help me out. I don’t worry about this. Just really curious, is this a tough problem? From my aspect, Dell don’t really know how to fix it. 

And at the beginning, I have the same idea with you. The loud fan noise should be related to some software settings, like BIOS and AWCC. However, how you explain that the GPU can run quietly and has no sign of high temp for 5 month? Just someday, it changes.

After the replacement of the GPU, things become worse. Before the replacement, I can run games like Planet zoo quietly. Only games like Monster Hunter can make that noise. After the replacement, same problem when I run Monster Hunters. However, the GPU temp  increases to 84 and the fan speed increase to about 3000 RPM immediately when I open planet zoo. There is no change of any software settings. So I start to think about heat sink. The difference performance between two GPU should be caused by the difference of heat sinking capacity.

If you consider my original card as a card with medium heat sinking ability, and the replacement as the one with poor ability. Then it can make sense. When I purchase the desktop, the card should have a good capacity. Somehow, it changes. And alone with the default setting, the fan noise happens. This is the story now I believe. Then what can affect this ability, I think the most common answer is thermal paste. However, Dell will not do the check of thermal paste. I guess I can only hope they will give me a good replacement eventually.

@gtdiesel 

You can use the link above to have a look.

And to be honest, what Dell will do is giving you a refurbished GPU. And what you can do is to pray this one is pasted well.

Ah, thank you very much for that link. That indeed sounds like a bunch of folks with the same issue. I have the same behavior, GPU hitting 80+ without trying very hard. 

My R8 is still under warranty so I might roll the dice with Dell, and then give this a shot if that doesn't help.

---

@markburv wrote:

@Scallop 

Running a GPU over 80° is not what I would recommend. I would try to get the GPU at least under 75°. Like @Doghouse Reilly said, max out your case fans to get cooler air inside and around your card.

Oh, Oh,...my fault for not being clear...I meant to specify to max out the GPU fans...I let the case fans run as the computer overlords allow.  When running an intensive game, the GPU fan is running around 96+ percent which gives me maximum GPU temps of 75 C / 167 F.  I use Afterburner's automatic scan to over clock the video cards and set the fan speed to vary with need per the GPU temperature.

Lots of fan noise, especially with my 2 GPUs running in SLI, but I also don't recommend running GPU temps any higher than necessary.

Unfortunately, this is a side effect of a small case that is stuffed with components and little room left for air flow.

Well max out all the fans and can't get better than that for temps ...  we have liftoff !!! 

That paste is awful in that EVGA link!

Having said that, imho, no factory pasted chip is better than redoing it yourself with high quality paste. I have a slightly less than satisfactory CPU temperature on my 2 year old R8 and will be repasting this weekend.

In the end, in less than a year, if you are buying the latest games, soon enough you will have to set to lower quality settings or pushing the clocks too high or you will finally resort to getting a new graphics card to have it playable. We are in a consumer market where games are always hungry for more power, a graphics card 3 years old is nearing it's end-of-life in the power hungry game market.

If you want to get a year or two extra out of your GPU at always the max settings without Sync, then repaste, then OC for more life but then maybe have to repaste more regularly. That will be your choice.

I looked at those graphs, after spiking the temperature dropped to what looked like < 80°. Given that GPU fans do not slow down rapidly, that boost brought your cards to a more acceptable temperature for it to drop back to an acceptable silence.

This is the point, silence, the market demands more silent computers and manufacturers know that the most cost-effective way is managing better the fans, that may explain the spikes when it hits a threshold.

I will always make sure I limit FPS instead of pushing out more frames than my cable/monitor can handle and find the balance depending on game type, for example multi-player shooters up max FPS and single-player campaign type games like this to 60 FPS is more than enough tradeoff for temperatures > 75°.

Pushing your GPU above your FPS limits of your setup is imo totally pointless, it just increasing your temperatures making your paste wear out more.

I understand you do not know why, you have a great card still, just don't blindly ignore VSYNC is my suggestion and get more life out of your GPU like I intend to.

Sorry for the long read, I hope it helps you, in short, don't ignore FPS, an good luck