High GPU Hot Spot Reading

[Cloudy]

HWiNFO64 reads that my Asus Strix RTX 2080 Ti has a GPU Temperature of 70°C while gaming, but the GPU Hot Spot Temperature reads 100°C. Should I be worried?

[Illusion of Progress]

Hm, not sure, because not long ago these hot spots weren't reported on nVidia hardware so whatever was fine before should be fine now. It was known that one reading on one spot won't always equal the warmest spot. However, a difference of ~30C seems like a lot. Most comments I've found on the subject report a difference closer to 10C or so (which is about what i tend to see on my GTX 1060), but I guess it may depend on what you're doing on the GPU at the time since not everything will utilize it the same.

Is this difference consistent or have you so far just used one situation (game) and noticed the difference then? It might be worth reapplying the thermal paste in such a situation?

[Supafly]

Biggest difference I've seen is 14 and my current is 12. 30 seems like a lot and could simply be down to a dodgy sensor.

If it's throttling then you'd want to improve airflow. If it's not throttling I wouldn't worry to much

[Schrute_Farms_B&B]

A difference of around 10-15 degrees between gpu and hot junction/hot spot temperature is widely considered as normal but everything over 20 degrees, let alone 30-40 degrees indicates that the cooler isnt evenly touching the the surface / uneven spread of thermal paste.
You could try to tighten the screws a tad and see if that helps. A new thin layer of thermalpaste could also help. You could also ignore it, if youre able to stay within the 80-90 degrees range.

100-110 degrees on the other hand are way too high for hot spot temps, even for a 2080TI. High junction temps also affect the lifespan of your gpu.
It might be marginal, or more distinct, but the impact is undeniable.

[Guydodge]

seems to warm for my liking i know with the Nvidia founders addition 3000 series Nvidia themselves said 100c was ok as their cards memory is hitting 100c.i honestly dont believe
100c to be ok.the partner cards are not hitting 100c.my 3090 was hitting 75-85c under heavy load.about 10c difference from gpu temp which was 65-75c.so i set a 92mm noctua 2000rpm fan on the opening in the backplate and it knocked it down about 10c overall.not sure the 2080ti would do the same.but at 100c i'd try something.
(those temps are with a 100mhz OC)

also i have my fan curve set to ramp up at 45c to as high as they would go without being
above what i considered to loud.for me thats 93%.easier to keep cool than to cool it down

[plat]

For verification, you can download the trial version of AIDA64. This will also give you a hotspot reading (good thing too, a lot of readings aren't available in the trial). The hotspot reading is under the Computer icon, and then you click on the "Sensor" icon at the top.

If it's the same as HWiNFO's, you can also search the issue online and see what others say about it.

https://www.aida64.com/downloads/NzlmZTllNmE=

The difference on here betw. base sensor and hotspot is usually about 11-20 deg C but that's contingent on usage. This older gpu doesn't get a very heavy workout anymore unless it's being benchmarked which ain't often.

Here's a Reddit thread that suggests thermal pads; I'm sure there's more. You can also search the NVIDIA forums.--but unsure what's on there pertinent to your issue.

https://www.reddit.com/r/techsupport/comments/lsgwk2/is_100_degrees_celsius_normal_for_hotspot/

https://www.nvidia.com/en-us/geforce/forums/discover/

[_I_]

and be aware that running multiple temp monitoring softwares at the same time can cause some incorrect readings

like 2 asking 1 device questions at the same time

[N0REGARD4LIFE]

Originally posted by Cloudy Canadian:

HWiNFO64 reads that my Asus Strix RTX 2080 Ti has a GPU Temperature of 70°C while gaming, but the GPU Hot Spot Temperature reads 100°C. Should I be worried?

A typical hotspot/core temp gap is around 10-15c. That’s definitely not normal. Did you recently open the cooler and replace the thermal pads? If so you might have gotten pads that are too thick causing poor gpu to cooler contact. If not, maybe open it and repaste. Maybe that and a reseating of the heatsink will drop your temps down some. 30c difference isn’t normal or good though.

[Cloudy]

A few things to note about my build:
- As of now I have never opened my GPU before, so it couldn't be that I pasted it or screwed it together incorrectly
- Case has a mesh front with high airflow, and even when I do remove side panel the temps almost don't improve at all
- Case fans are running at a medium RPM, and all three GPU fans are running around 70%-80% while under load
- These temps I said are with any demanding games, not just one or two particular titles
- My GPU's VRM temps are not high, only "Hot Spot" reading
- My game's framerate has no issues

[Schrute_Farms_B&B]

Originally posted by Cloudy Canadian:

As of now I have never opened my GPU before, so it couldn't be that I pasted it or screwed it together incorrectly

But the manufacturer could have done it wrong. You know how many times I had to send a GPU back, because they did something wrong ? Doesnt matter if AMD,Nvidia, Sapphire, Asus or whatnot, mistakes happen but it gets dangerous if the consumer doesnt pay attention.
Your card is also a few years old and the thermal paste could need a little "freshen up"

Originally posted by Cloudy Canadian:

- My GPU's VRM temps are not high, only "Hot Spot" reading

Doesnt matter. The junction temperature is way more important then other values. People dont even pay attention to the regular temps since it became almost meaningless.

Originally posted by Cloudy Canadian:

- My game's framerate has no issues

Thats good for you but also doesnt matter.

Plenty good suggestions and solutions in the comment section. I think you should know what to do by now.

[Cloudy]

So I took the time to take the GPU out and open it up, the paste was very dry. After cleaning it up and very carefully reapplying new paste (Thermal Grizzly Kryonaut) I made sure everything was seated evenly and screws were fairly snug when put back in. Running some tests for awhile I can say it was a success! GPU temp is now 66°C and Hot Spot is only 81°C! I didn't change the stock thermal pads since they were still "fresh" feeling and clean.

Thanks guys for the help, I feel much better now knowing my GPU is running cooler.

[N0REGARD4LIFE]

Originally posted by Cloudy Canadian:

So I took the time to take the GPU out and open it up, the paste was very dry. After cleaning it up and very carefully reapplying new paste (Thermal Grizzly Kryonaut) I made sure everything was seated evenly and screws were fairly snug when put back in. Running some tests for awhile I can say it was a success! GPU temp is now 66°C and Hot Spot is only 81°C! I didn't change the stock thermal pads since they were still "fresh" feeling and clean.

Thanks guys for the help, I feel much better now knowing my GPU is running cooler. :Hatfall_Master:

Good to know and look, you’re in that 10-15c range now. If you ever do decide to change the pads, make sure you get the correct ones. VRAM on almost every 2080ti is 0.5mm but do some research on YOUR card just to make sure.