• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Is 75C GPU 90C hotspot normal for a 3070 TUF in a well-ventilated case?

NewbieOneKenobi

NewbieOneKenobi

NewbieOneKenobi

Associate
Joined
7 Sep 2020
Posts
107
CPU: i5-9600KF @ ALFII 420, OC removed for testing
mobo: Aorus z690 Pro, so solid value board / bottom of the high end
mem: Ripjaws V DDR4/3000MHz/CL14 (2x8GB)
storage: SSD/NVM/PCIE3 (CS3030)
PSU: Corsair RM750 (the yellow-sticker one)
case: Enthoo Pro
fans: four P14 @ 1000 rpm intake, four P14 exhaust (3 from the AIO, top-mounted @ push @ PWM, and 1 sysfan @ 1000 rpm)

The TUF still manages to heat up to 75C GPu / 85–90C hotspot in a single benchmark run, marginally cooler in games.

Performance is a bit meh. It's within range for a 3070 but below the average score of a FE. TimeSpy says I'm below average for this CPU+GPU combination. In Kingdom Come: Deliverance the game feels smoother than with my (now-dead) 1070ti, but I see bigger fps dps than with the 1070ti in Rattay. I now see lot more <40 fps moments than I did with the 1070ti. The 1070ti had been heavily OC'd, boosting at >2200 and all, but a 3070 should still be much stronger than a full 1080 and still noticeably stronger than a 1080ti.

The card worked in a data centre for 1.5 years, and it's already a replacement. The first one was so loud and hot that the company owner apologized and sent me a personally tested unit, which is this unit. Less loud and marginally less hot than its brother but still not what I'd expected from a TUF given its stellar reviews, like TechPowerUp's.

So, am I being a princess or is something off? I'm leaving on a holiday in a couple of days, so I wonder if I should keep the card, swap the paste and pads when I return, or return it right now just to be on the safe side before my 14 days run out.

If this came from a gamer or designer, I'd be inclined to take the risk and just swap the paste, but after 1.5 years in a data centre perhaps the card might be damage? I have no information as to whether that was 8 hours a day or 24/7/365 for 1.5 years, but still… 1.5 years isn't an awful lot of time. Shouldn't really be enough to suffer discernible degradation.
 
Might be worth running the card with case side off, just to rule out any airflow issues.


My Inno3D runs quite hot, even after reapplying thermal paste which was awful from factory, no issues though with Time Spy and am in the Legendary results when oc'd.
 
3070 FE I had fully stressed benchmark would hit 75c , hotspot 85c , clocks 1840-1920mhz at stock I would have thought tuf with beefier cooler would be lower
 
13oots2 said:
Might be worth running the card with case side off, just to rule out any airflow issues.
Click to expand...
Yeah, tried both open and closed, though I don't remember the difference, sadly. I'd presume my airflow to be at least decent, given the fan count and lack of obstacles. Two P14 for front intake, another two for bottom intake, one for rear exhaust (sysfan) and three through the radiator in the roof. In the bay section you can feel it's really windy. In the mobo section less so, also given how the length of the CPU basically cuts the case in two compartments, and then there's the discreet soundcard, but I'd still presume the airflow to be not worse than most users and reviewers probably have.


Alt0153 said:
its probably just got ****** factory paste thats been running 24/7 hot as balls and needs replacing
Click to expand...

I certainly hope that's the case and that there's no more to it. I messaged the company at the night, and they actually replied right away, saying that they're going to take a look at my thermal results and respond in detail tomorrow but are also open to accepting a later return, especially if a professional repair shop finds some issue with the card itself rather than just the paste and pads needing replacement.

I've been looking at other deals pending my original 14-day window, as that's fair game IMHO if there are any sort of problems with the purchase, but this little TUF still manages to beat all other deals by at least a narrow margin. Nowhere close to being a suspiciously good deal, it's still convincingly better than the alternatives. So I hope it's just paste and pads, though I'd rather not end up with a significantly degraded card.

Joxeon said:
You sure it's from a date centre and not come from a mining farm?.
Click to expand...

Yeah, though I suspect the data centre used the cards to mine crypto at night anyway, unless their data-centre operations were 24/7. One thing I'm sure about is that their cards have managed to wear their paste off after 1.5 years of use.

uscool said:
3070 FE I had fully stressed benchmark would hit 75c , hotspot 85c , clocks 1840-1920mhz at stock I would have thought tuf with beefier cooler would be lower
Click to expand...

Indeed. I get similar temps with 1950MHz at stock, which is not a great difference. And we're talking about a massive heatsink and three fans versus two.

For the record, it seems I might be getting better thermals with the GPU after applying a hefty CPU overclock, so perhaps there was a bit of a problem with my 9600kf being too weak and bottlenecking the card. But I need to give it an outrageous 1.45V to sustain 5 GHz, at which point it reaches 80C under load with a 420 AIO (my Phanteks TC14PE with three P14s would probably have melted)… Not a good overclocker.
 
NewbieOneKenobi said:
Yeah, tried both open and closed, though I don't remember the difference, sadly. I'd presume my airflow to be at least decent, given the fan count and lack of obstacles. Two P14 for front intake, another two for bottom intake, one for rear exhaust (sysfan) and three through the radiator in the roof. In the bay section you can feel it's really windy. In the mobo section less so, also given how the length of the CPU basically cuts the case in two compartments, and then there's the discreet soundcard, but I'd still presume the airflow to be not worse than most users and reviewers probably have.

Just a thought, a few reviews will be open bench.
Click to expand...
 
You must log in or register to reply here.
Back
Top Bottom