Overclock.net banner

161 - 178 of 178 Posts

·
Registered
Joined
·
1,557 Posts
Thanks for the quick responses.
Quote:
Originally Posted by The Stilt View Post

Increasing the value will direct more load toward the PCI-E connector, but the difference between 0xB and 0xF is extremely minor and AMD is already using 0xD in the 16.7.1 driver. Lowering the GPU power draw will definitely lower the power draw from the bus, but not by the same amount as the total consumption decreases (due VRM being split 50/50). So if you manage to decrease the GPU power draw by 10W, with "F" programming value you should see around 6W decrease in power draw from the PCI-E slot (5W at stock settings, i.e with "0" programming value).

The memory VRM is fed only from the PCI-E power connector, so the memory clocks don't affect the power draw from the PCI-E slot significantly (most likely the difference is measured in mW).
So what you're saying is that the difference between "F" and "D" is negligible, and decreasing core voltage will significantly decrease the PCI-E slot power draw.

Thanks,
 

·
Registered
Joined
·
1,557 Posts
Quote:
Originally Posted by JackCY View Post

Or have a mobo that already has that by default without any extra.
Or a powered riser no?
Problem with powered riser's for me is that I don't have enough molex cables (only one molex cable with 4 connectors on it, and if i connect all four to the gpus I would be overloading the cable), so I'm stuck with 4 gpus that need to be powered via SATA connectors, which i dont think can handle the 65W load. Perhaps with sufficient undervolting I can manage it.
Currently have one molex cable with 4 connectors, and 3 sata cables with 4 connectors each.
 

·
Meddling user
Joined
·
7,423 Posts

Attachments

·
Meddling user
Joined
·
7,423 Posts
Yes, GPU-Z, etc show monitoring data.

Just for the purpose of if someone view the attached data and think what the other changes are I stated it
wink.gif
.
 

·
Meddling user
Joined
·
7,423 Posts
Quote:
Originally Posted by The Stilt View Post

That's correct...
A mistake IMO, since I've killed some CPUs accidentally when the limit has not been set.

1.36875V was set for enhanced safety and should be more than enough, even on custom water cooling.
So VMAX register as is per loop and loop 2 not used on RX 480 you went 5Fh instead of 53h on MLU?

** edit **

Tested it
smile.gif
, changed Fury X 66h to 64h.

 

·
Registered
Joined
·
23 Posts
Noob here. I am trying to overcome throttling on my MSI RX 480 Gaming X once I overclock past a certain clock rate/voltage/PL. Someone suggested I was being limited by the power limit even though I have the newest Trixx release set to +50. I can get to a point where the 480 throttles the core clock quite severely once I push so much voltage to it to get past screen anomalies. There appears to be no 'balance' where I can get a stable overclock and no throttling. For instance I can get to 1370Mhz +50 PL and +66mV but running Firestrike loop stress test it throttles after a few minutes hitting past 73c (max temps I get are 76/77c tops). I am sure it has more headroom there temperature wise? So will this get me past that throttling without burning the card so to speak?

Please correct me if you think I am being dense
smile.gif
 

·
... not new ...
Joined
·
751 Posts
Quote:
Originally Posted by deadman3000 View Post

Noob here. I am trying to overcome throttling on my MSI RX 480 Gaming X once I overclock past a certain clock rate/voltage/PL. Someone suggested I was being limited by the power limit even though I have the newest Trixx release set to +50. I can get to a point where the 480 throttles the core clock quite severely once I push so much voltage to it to get past screen anomalies. There appears to be no 'balance' where I can get a stable overclock and no throttling. For instance I can get to 1370Mhz +50 PL and +66mV but running Firestrike loop stress test it throttles after a few minutes hitting past 73c (max temps I get are 76/77c tops). I am sure it has more headroom there temperature wise? So will this get me past that throttling without burning the card so to speak?

Please correct me if you think I am being dense
smile.gif
Power draw increases with temperature, so if you were running cooler you could run more frequency in the same power.

I can't speak for the MSI card, but this tool will allow you to trick your VRM logic into thinking it is consuming less power. You can push more voltage and tweak many other things as well.

The setting you want s "Current scale."

If you watch the "GPU Power" in GPU-z you will see that it drops when you press "Half."

Keep the temperatures under 85C and you should be good, but 80C or lower is better.
 

·
Registered
Joined
·
23 Posts
Thanks. It helped a little bit I think. It still throttles in Firestrike stress test loop though after a certain point. I can run Heaven/Valley and ran Rise of the Tomb Raider (Very High setting) for a while at 1390Mhz using +96mV with a temperature of 80c and did not see any throttling. That was with the fans going full speed. I wonder if I should take it apart and put some decent Arctic TIM on it (I would need to ask MSI if that is OK though. I think they are OK with it as long as you don't damage anything).
 

·
... not new ...
Joined
·
751 Posts
Quote:
Originally Posted by deadman3000 View Post

Thanks. It helped a little bit I think. It still throttles in Firestrike stress test loop though after a certain point. I can run Heaven/Valley and ran Rise of the Tomb Raider (Very High setting) for a while at 1390Mhz using +96mV with a temperature of 80c and did not see any throttling. That was with the fans going full speed. I wonder if I should take it apart and put some decent Arctic TIM on it (I would need to ask MSI if that is OK though. I think they are OK with it as long as you don't damage anything).
That possibly means the VRM controller is not the same as the reference (or my XFX GTR).

But... I can only get my card to pull 170W from the GPU itself, which is only a little higher than what I can get with +50% power target... so you're probably not missing much. 1390Mhz is pretty awesome for air cooling. 80C temps seems about right - the small footprint of these GPUs makes it more difficult to cool higher heat outputs. Changing the TIM could be worth a few extra degrees - might let you hit that magic 1400Mhz number we all know you're craving (I hit it with mine, but just barely could get it stable no matter how much voltage I threw at it - despite staying at or under 80C (100% fans)).

Good luck!
 

·
Registered
Joined
·
23 Posts
Yeah as I said it's only really throttling in a synthetic that pushes it to the limit like Firestrike stress test. Realistically, in games it seems to do it a lot less although something like RotTR might start thottling after running for more than 10 minutes. I would have to suck it and see. I will probably contact MSI directly before considering changing the TIM (there is a seal screw) so I am on their system.
 
161 - 178 of 178 Posts
Top