Topic: Radeon VII temperature (Read 497 times)

I hope so, but it looks so difficult to imagine caused by these reasons

That's only the first reason and you can read more through this link https://www.quora.com/Why-is-AMD-so-behind-NVIDIA

Someone has very well explained about that.

I'm hoping that AMD will come out with some awesome Navi gpu with low power consumption that is worth buying relative to price/power. They're killing it with their CPU's, hopefully GPU's will be great at some point.

But for me that's normal to see radeon VII gets junction temp around 80 - 90. I have a friend that was using it for play a game in 4k resolution and it was increasing more than 100 jtemp.
I heard some people in another forum said that's normal as long as it has not yet passed the throttle limit.

Personally I prefer to get the temperature below 65 degrees celsius when mining. Above that I think I feel a chemical strange smell coming from the computer and or gpu's. Don't think I will ever buy any Radeon 7's for mining if that temp is the norm.

There was this video which showed where you can undervolt the card to get it to 70c only if you adjusted it in the overclocking software from ati control panel.
https://www.youtube.com/watch?v=5emUHO_Eur4
Not sure if it would work for mining since you are pushing these cards to their limits to make mining with them profitable.

Try to reset core and memory to default and anyway turn fans at 100% for some time. Just to see the result temperatures.
PS. Fan? Not fans? What a model is it?

Not quite sure, but I think I remember something about the Radeon 7 measuring temperatures slightly different than the usual norm, point of the matter is the Radeon 7 will manage higher temps than your previous amd gpu's.

Radeon VII has production issues with making the die package level. This made the company have to resort to using higher end thermal pads between the die/hbm and the heatsink.  This is a good bit inferior than a properly thin layer of thermal paste, however since not all die packages were level this level of thinness couldn't be garunteed. So, in order to ensure all cards perform at least within spec, AMD used a thicker solution, a thermal pad. 

Many have modded their cards to replace the pad with paste, hoping their dies are level enough. Others are tightening the mounting screws a bit more, and of source there are water cooling plates out there now, as well as people tossing on aftermarket coolers like the morpheus II.

Add this to the stock voltages being way to high for every clockspeed, and needing a bit of external airflow over the backplate (as would be had in a well ventilated case) and it results in many ways to address cooling. Don't just have this card sitting on a testbench....it needs a bit of airflow over it.

Finally don't forget that throttling temp is based of the new tjunct value, which is the highest temp in a much dense array of sensors as compared to hot spot temp value of vega 56/64.  This difference is more about a change in measurement resolution rather than the chip actually running any hotter or cooler than last gen. As long as your tjunct isn't throttling, pay more attention to your general core temps.

Open rig with single GPU. No fans on rig case. Other Rx570 GPUs on a separate rig showing low 60s temperature.
Fan are purposely not running at 100 pct. It is to lower noise and not breaking fans.

Is it in rig or in a case? How MSI Afterburner is configured?
First of all - fans ought to be working at 100%. What else problem can be - i need answers to say smth.

Are they measuring the temp different them other cards?  Aren't the higher temps normal?

There is a miner in cold canada who has been vlogging on his channel about this very card.
I think he mentions he has two of those expensive cooler fans Noctua to keep the cards cool cause they do run up high when you are mining with them which runs them at full throttle.
Here is the starting video where he starts then it progresses to the profits of how much it makes for running it for a month.
https://www.youtube.com/watch?v=1Bse_fX6SmQ
https://www.youtube.com/watch?v=q5KK7uGwsJ4
Shows the open case with those 3000prm 120mm brown black noctua fans on the front
https://www.youtube.com/watch?v=U34p8v8UkBE
And the profits made from last month
https://www.youtube.com/watch?v=o1VdvfGkbR0

He does mention the fans on the cards themselves are loud when they are run full tilt so some kind of external fan setup wouldnt hurt.

Thanks everyone for help.

It is an open rig and only one GPU in the rig. Windows are open and other Rx580/570s are all showing less than 60 degree. I could increase fan because noise is not a problem. I try to undervolt till stability became concern.
I noticed that air outflow is from top of the GPUs.  I could try to see if putting some external fan helps.

The Radeon VII is not the "coolest" card outthere. In fact, AMD is taking to 2020 Navi 20 because the flagship of the company, to compete with 2080ti has problems with thermals and power consumption according to some reports. AMD has problems with thermals even at 7nm.

Why do you think all Radeon VII have a 3 fan config? Because 2 was simply not enough. Even the 2080ti can do ok with 2 fans. Doesn't this tell you anything on thermals with AMD???

And why can't you see models of Radeon VII with MSI, Sapphire, ASUS, Zotac , EVGA, PNY, and others?? Well, because thermals are just a complete pain in the ass.


I understand you. You bought what you thought would be a good deal to mine. In the hottest days, you may probably see 100ºC on the cards unless you go 100% rpm fan at 3.000 rpm which is the loudest shit you will hear from a card.

Solutions:

1º increase manually the fan rpm to near 100%.
2º increase airflow in the room by opening windows. if you have the card inside a case, open one panel so that there is much more airflow
3º if your case has place for fans, install them. If not, it would be a nice idea to buy a new case with very good airflow to keep the card as far as possible from dust.
4º you can also try decreasing core voltage till it crashes, to see the min voltage in which your core is able to work at a given speed.
5º reduce the core clock = reduction in voltage = reduction in heat
6º i don't think you can see 90ºC on an open bench unless you live in hell so.....

Ohhh btw, it's possible to see 100mh/s with Radeon VII , so you still have work to do.

The fan doesn't seem to be the issue here. I think you have it inside an enclosed PC case without adequate cooling. Basically you need a good flow intake and exhaust fans mounted inside your case, if you don't want to do this then at least just remove the cover. Might be a little louder but as long as you don't sleep in the same room it should be fine.

If the GPU was near throttling then your fan speed would actually even be higher, and in general the junction temperature can always be slightly higher.

Only other thing you can do is to keep reducing the core voltage until your stability is bad, then you reached the lowest stable voltage and can't go lower without reducing your engine clock.

It is still in range if the card temperature reach to 115°C then it may start throttling so keep it at the current value and try to provide sufficient ventilation/cooling in room too.

I don't think a junction temperature of 90 degrees C is a problem though. The core temperature should be much lower, but the junction temperature should be below 100 or 110, which you already have. If you still wanna get it lower, you could always undervolt, but I think it is not that hot for a junction temperature.

Use some tool like Msi to set fan speed manually and if that doesn't work there may be some problem with your card fan.

The temeprature limit should be 90 is normal but the working temperature should be a lot less than that.

Currently mining Ethereum using Phoenix and on minerstat os.
FAN70%
CORE1750 MHz
MEM1100 MHz
VDDC990 mV
VDDCI0 mV
It is stable but junction temperature is 90. Hashrate is 89.
Any suggestion how to reduce temperature?
Thanks