"Overclocks Dream" The Fury X

shankly1985 · 11 Jul 2015 at 23:14

Even if we was to go by TomsHardware

They showing Gaming vs Stress Testing..

Infrared Temperature Measurements

We see the consequences of that conservative fan setting in our infrared temperature measurement results. During gaming, the VRMs stay reasonably cool, even though they're only covered by a small heat sink that touches a heat pipe above it. The board hits 60 °C at the slot, meaning the VRM’s heat travels across the PCB under the rubberized back plate.

The story changes during our stress test. The water-cooling rule of thumb comes to mind right away: use one centimeter of radiator length per 10W of power. Almost 90 °C at the motherboard slot indicates that the VRM pins have passed 100 °C. This certainly isn’t a great way to run the card long-term, but then again, stress tests aren’t an everyday usage scenario. Still, it would have been nice to see some reserves for overclocking.

So 60c when gaming!

pgi947 · 11 Jul 2015 at 23:17

You should take a thermal guns reading over afterburner to be fair, it's giving you real time, on surface temps, not what a diode is reporting. Temps will move about, but a big hit strip right down the vrm bank only tells us one thing....that bank is running hot.

shankly1985 · 11 Jul 2015 at 23:19

pgi947 said:
You should take a thermal guns reading over afterburner to be fair, it's giving you real time, on surface temps, not what a diode is reporting. Temps will move about, but a big hit strip right down the vrm bank only tells us one thing....that bank is running hot.

But the problem with that is.. What the temp reading is on the cards casing isn't showing how well the GPU is moving that heat away and then showing its running temps.. A case could very well show 90+c but the core could really be running 60c

Davedree · 11 Jul 2015 at 23:23

shankly1985 said:
Even if we was to go by TomsHardware

They showing Gaming vs Stress Testing..

Infrared Temperature Measurements

We see the consequences of that conservative fan setting in our infrared temperature measurement results. During gaming, the VRMs stay reasonably cool, even though they're only covered by a small heat sink that touches a heat pipe above it. The board hits 60 °C at the slot, meaning the VRM’s heat travels across the PCB under the rubberized back plate.

The story changes during our stress test. The water-cooling rule of thumb comes to mind right away: use one centimeter of radiator length per 10W of power. Almost 90 °C at the motherboard slot indicates that the VRM pins have passed 100 °C. This certainly isn’t a great way to run the card long-term, but then again, stress tests aren’t an everyday usage scenario. Still, it would have been nice to see some reserves for overclocking.

So 60c when gaming!

70c vrm for a fury x at stock voltage 1.21v, what about when increasing voltage? !
For a fury on air vrm is 100c gaming at 1.21 v

Davedree · 11 Jul 2015 at 23:24

shankly1985 said:
But the problem with that is.. What the temp reading is on the cards casing isn't showing how well the GPU is moving that heat away and then showing its running temps.. A case could very well show 90+c but the core could really be running 60c

The heat is still there at the vrm's if it wasn't then the infrared wouldn't show a heatspot.

shankly1985 · 11 Jul 2015 at 23:29

Davedree said:
70c vrm for a fury x at stock voltage 1.21v, what about when increasing voltage? !
For a fury on air vrm is 100c gaming at 1.21 v

60c is nothing, vrms can hit 120c and be safe.. Hell I used my other 290x for couple weeks without any vrms heat sinks on them when using kraken g10.. Vrms was 90+ with hot air passing over them.
Never had a issue..

Davedree · 11 Jul 2015 at 23:32

shankly1985 said:
60c is nothing, vrms can hit 120c and be safe.. Hell I used my other 290x for couple weeks without any vrms heat sinks on them when using kraken g10.. Vrms was 90+ with hot air passing over them.
Never had am issue..

Depends on the quality of the vrms and chokes used, but as a general rule for long term 100-110c max, 110c+ short term yeah.

On air fury 3584 is at it's limits.

Gerard · 11 Jul 2015 at 23:33

The vrm's on fury x are supposedly good for 150c. Its on one of the reviews though i can't remember which, it was brought up after the thermal image surfaced.

pgi947 · 11 Jul 2015 at 23:52

Being safe and being efficient are two different things, they are very safe at those temps, they are also very inefficient at those temps, as this thread is about the oc capabilities, my point is they are reduced due to the vrms being hotter then those of an actively cooled card.

When overclocking the vrm temps can become as important as the core temps.

Also define stress test for me, is this furmark or 3dmark?

NirK · 11 Jul 2015 at 23:58

mm bit of a flop unfortunately the AMD cards...

Actually the non x version + a freesync monitor is pretty good on a budget, but a 980ti + Gsync is a lot better.

Nutella33 · 12 Jul 2015 at 00:00

pgi947 said:
Being safe and being efficient are two different things, they are very safe at those temps, they are also very inefficient at those temps, as this thread is about the oc capabilities, my point is they are reduced due to the vrms being hotter then those of an actively cooled card.

When overclocking the vrm temps can become as important as the core temps.

Also define stress test for me, is this furmark or 3dmark?

Alrite. what are the 980ti's VRM temps? we know it clocks well so they should be low 60's right?

pgi947 · 12 Jul 2015 at 00:15

Lots of variables are included, I'd presume gm200's vrms run quite hot, it's very easy to hit the 110% power limit with minimal voltage increase

Nutella33 · 12 Jul 2015 at 00:58

pgi947 said:
Lots of variables are included, I'd presume gm200's vrms run quite hot, it's very easy to hit the 110% power limit with minimal voltage increase

So if they are running at 90+ they are to be assumed inefficient and the 980ti a bad clocker?

8 Pack · 12 Jul 2015 at 01:06

Power limit is tdp of the card measured by draw through pci X connectors. It's nothing to do with vrm temps.

Kaapstad · 12 Jul 2015 at 01:08

8 Pack said:
Power limit is tdp of the card measured by draw through pci X connectors. It's nothing to do with vrm temps.

Ian have you had a go with any of the Fury chips on LN2 ?

My gut feeling is they are not good for benching but very good for 2160p gaming.

JediFragger · 12 Jul 2015 at 01:20

Kaapstad said:
Ian have you had a go with any of the Fury chips on LN2 ?

My gut feeling is they are not good for benching but very good for 2160p gaming.

8 Pack · 12 Jul 2015 at 01:21

If the vrm is alive who cares.

No not tried ln2 need to work out volt mods. Much time at the moment testing other stuff.

8 Pack · 12 Jul 2015 at 01:26

My gut feeling is also not good for Ln2. Maybe I waste a lot of time I don't have. Possibly many mods needed which I don't have time to trouble shoot and AMD can't help with.

Nutella33 · 12 Jul 2015 at 01:29

8 Pack said:
My gut feeling is also not good for Ln2. Maybe I waste a lot of time I don't have. Possibly many mods needed which I don't have time to trouble shoot and AMD can't help with.

Interesting. Do I sense a bit of a falling out between OCUK and AMD?

8 Pack · 12 Jul 2015 at 01:34

Not at all. LN2 Overclocking is no brands priority all this stuff has to be figured out and tested. No one has yet tried with Fury X. I asked some questions but AMD themselves did not test sub zero at all.

For me more efficient drivers should be and is there priority.