Contribute
Register

gpu temperature info rx 580 8gb

Joined
Apr 3, 2019
Messages
68
Motherboard
ASRock Z390 Phantom Gaming ITX/ac - v4.40
CPU
i7-9700
Graphics
UHD 630 + RX 590 Nitro+
Try setting FrameBuffer to Orinoco.
Sorry but I dont get in your link how I am supposed to set the framebuffer to Orinoco. I think I don't need to remap my GPU ports as I only use one. Unless a wrongly identified connector, even if not in use, might cause higher power consumption? In my case I have 2 HDMI + 2 DP ports are working. I don't know for the DVI port (I have no DVI cable).

But if you said that you saw your power consumption drop to 30W with only one monitor, I'm willing to try to achieve the same result!

Also, my dGPU is alreaydy properly displayed as a RX590 in macOS. I just have the WEG kext with no additional injection/fix/boot argument), and I'm using OpenCore 0.5.7 (but I got the same result with the same kext and no extra patching/injection using Clover).
 
Last edited:
Joined
Jan 18, 2018
Messages
104
Motherboard
GA EP45-UD3P
CPU
Q9550
Graphics
Sapphire RX 580 Nitro+ L.E
Mac
  1. iMac
  2. MacBook
Mobile Phone
  1. iOS
Sorry but I dont get in your link how I am supposed to set the framebuffer to Orinoco. I think I don't need to remap my GPU ports as I only use one. Unless a wrongly identified connector, even if not in use, might cause higher power consumption? In my case I have 2 HDMI + 2 DP ports are working. I don't know for the DVI port (I have no DVI cable).

But if you said that you saw your power consumption drop to 30W with only one monitor, I'm willing to try to achieve the same result!

Also, my dGPU is alreaydy properly displayed as a RX590 in macOS. I just have the WEG kext with no additional injection/fix/boot argument), and I'm using OpenCore 0.5.7 (but I got the same result with the same kext and no extra patching/injection using Clover).
Screen Shot 2020-04-27 at 6.22.12 PM.png
 
Last edited:
Joined
Jan 19, 2018
Messages
221
Motherboard
Gigabyte Z390UD
CPU
i9-9900
Graphics
UHD 630 + RX 580
Mac
  1. iMac
Mobile Phone
  1. Android
Hi all,

I just bought a used Sapphire Nitro+ RX580 for my hackintosh with dual monitor setup running i5-8600 32GB RAM and Mojave 10.14.6 with OpenCore 0.5.9.

When used in dual monitor, the core clock keeps holding at 1.340GHz which I understand is the maximum clock based on the specification on Sapphire website (no throttling at all under idle with dual monitor setup?), while the idle power is at 143W and temperature at 50degC. Quite high consumption in idle condition.

Then I run with single monitor setup (literally disconnecting the 2nd monitor), and the core clock goes down to around 300MHz but power consumption still quite high at around 90W and temperature is around 45degC.

Using terminal command also shows similar result with what is shown by HWMonitorSMC above.

In contrast, I tested with Unigine Heaven, and Unigine shows that my temperature moves around 40-50degC and never above that, while comparing with data at HWMonitorSMC gives around 70degC temperature.

Understand that the Unigine only gives temperature and not the power consumption, but we have two different information given by 2 application. Yet I believe, if the temperature is lower, for sure the power consumption is also logically lower.

Anyone has further information on this issue? I run only macOS in this computer, so I cannot check it under Windows. Maybe later I will try to see it in Windows when I want to do a fresh install, but if anyone has some information about what is actually happening here, I will be very happy.

Thank you.
 
Joined
Oct 19, 2013
Messages
51
Motherboard
Asus Prime Z270-A-0906 - OpenCore 0.63 - Big Sur 11.1
CPU
i7-7700K
Graphics
XFX Radeon RX 580 8GB GTX Core GDDR5 HDMI DP
Mac
  1. iMac
  2. MacBook Pro
Mobile Phone
  1. iOS
Hi all,

I just bought a used Sapphire Nitro+ RX580 for my hackintosh with dual monitor setup running i5-8600 32GB RAM and Mojave 10.14.6 with OpenCore 0.5.9.

When used in dual monitor, the core clock keeps holding at 1.340GHz which I understand is the maximum clock based on the specification on Sapphire website (no throttling at all under idle with dual monitor setup?), while the idle power is at 143W and temperature at 50degC. Quite high consumption in idle condition.

Then I run with single monitor setup (literally disconnecting the 2nd monitor), and the core clock goes down to around 300MHz but power consumption still quite high at around 90W and temperature is around 45degC.

Using terminal command also shows similar result with what is shown by HWMonitorSMC above.

In contrast, I tested with Unigine Heaven, and Unigine shows that my temperature moves around 40-50degC and never above that, while comparing with data at HWMonitorSMC gives around 70degC temperature.

Understand that the Unigine only gives temperature and not the power consumption, but we have two different information given by 2 application. Yet I believe, if the temperature is lower, for sure the power consumption is also logically lower.

Anyone has further information on this issue? I run only macOS in this computer, so I cannot check it under Windows. Maybe later I will try to see it in Windows when I want to do a fresh install, but if anyone has some information about what is actually happening here, I will be very happy.

Thank you.
It's about the same I have in idle
#ioreg -l |grep \"PerformanceStatistics\" | cut -d '{' -f 2 | tr '|' ',' | tr -d '}' | tr ',' '\n'|grep 'Temp\|Fan\|Clock\|%\|Power'
"Fan Speed(%)"=34
"Core Clock(MHz)"=1366
"Memory Clock(MHz)"=2000
"GPU Activity(%)"=0
"Fan Speed(RPM)"=782
"Temperature(C)"=51
"Total Power(W)"=141

I never really tested with other apps, but are confident that my numbers are within accepted levels
 
Joined
Oct 19, 2013
Messages
51
Motherboard
Asus Prime Z270-A-0906 - OpenCore 0.63 - Big Sur 11.1
CPU
i7-7700K
Graphics
XFX Radeon RX 580 8GB GTX Core GDDR5 HDMI DP
Mac
  1. iMac
  2. MacBook Pro
Mobile Phone
  1. iOS
Hi all,

I just bought a used Sapphire Nitro+ RX580 for my hackintosh with dual monitor setup running i5-8600 32GB RAM and Mojave 10.14.6 with OpenCore 0.5.9.

When used in dual monitor, the core clock keeps holding at 1.340GHz which I understand is the maximum clock based on the specification on Sapphire website (no throttling at all under idle with dual monitor setup?), while the idle power is at 143W and temperature at 50degC. Quite high consumption in idle condition.

Then I run with single monitor setup (literally disconnecting the 2nd monitor), and the core clock goes down to around 300MHz but power consumption still quite high at around 90W and temperature is around 45degC.

Using terminal command also shows similar result with what is shown by HWMonitorSMC above.

In contrast, I tested with Unigine Heaven, and Unigine shows that my temperature moves around 40-50degC and never above that, while comparing with data at HWMonitorSMC gives around 70degC temperature.

Understand that the Unigine only gives temperature and not the power consumption, but we have two different information given by 2 application. Yet I believe, if the temperature is lower, for sure the power consumption is also logically lower.

Anyone has further information on this issue? I run only macOS in this computer, so I cannot check it under Windows. Maybe later I will try to see it in Windows when I want to do a fresh install, but if anyone has some information about what is actually happening here, I will be very happy.

Thank you.
In the 4th post you will see factory burned in values and a medium value is 65°, you should not worry, also the Watts are based on the DC power used in the card, properly a combination of 5v/12v. Also take a look at the POWERTUNE on same image. There are max temps burned into card there will shutdown card etc....
 
Joined
Jan 19, 2018
Messages
221
Motherboard
Gigabyte Z390UD
CPU
i9-9900
Graphics
UHD 630 + RX 580
Mac
  1. iMac
Mobile Phone
  1. Android
It's about the same I have in idle
#ioreg -l |grep \"PerformanceStatistics\" | cut -d '{' -f 2 | tr '|' ',' | tr -d '}' | tr ',' '\n'|grep 'Temp\|Fan\|Clock\|%\|Power'
"Fan Speed(%)"=34
"Core Clock(MHz)"=1366
"Memory Clock(MHz)"=2000
"GPU Activity(%)"=0
"Fan Speed(RPM)"=782
"Temperature(C)"=51
"Total Power(W)"=141

I never really tested with other apps, but are confident that my numbers are within accepted levels
Thanks for sharing your numbers.
I got the following in idle with dual monitor.
Screen Shot 2020-09-21 at 22.03.11.png


While in single monitor setup see from HWMonitorSMC2, Core and Memory Clock is at around 0.300GHz with temperature around 40degC and power shown around 100W.

I do a comparative checking using direct power meter from the socket of the computer (excluding monitors) and the power consumption of the monitor fluctuates around 100-120W in dual monitor setup while around 70-80W in single monitor setup. Which makes the measurement in macOS rather doubtful, i.e. does not make any sense if the GPU measured at 140W while the computer setup only draws 100-120W measured from the socket.

Either one might be faulty (the HWMonitorSMC or my cheap power meter), but I now understanding that nothing is strange with this measurement (considering you have similar measurement as well as my secondary measurement method), then I am ok with it. While I might learn and do something on PowerPlayTables in the future to tweak the performance when I feel necessary.

Thank you very much.
 
Joined
Oct 19, 2013
Messages
51
Motherboard
Asus Prime Z270-A-0906 - OpenCore 0.63 - Big Sur 11.1
CPU
i7-7700K
Graphics
XFX Radeon RX 580 8GB GTX Core GDDR5 HDMI DP
Mac
  1. iMac
  2. MacBook Pro
Mobile Phone
  1. iOS
Thanks for sharing your numbers.
I got the following in idle with dual monitor.
View attachment 488939

While in single monitor setup see from HWMonitorSMC2, Core and Memory Clock is at around 0.300GHz with temperature around 40degC and power shown around 100W.

I do a comparative checking using direct power meter from the socket of the computer (excluding monitors) and the power consumption of the monitor fluctuates around 100-120W in dual monitor setup while around 70-80W in single monitor setup. Which makes the measurement in macOS rather doubtful, i.e. does not make any sense if the GPU measured at 140W while the computer setup only draws 100-120W measured from the socket.

Either one might be faulty (the HWMonitorSMC or my cheap power meter), but I now understanding that nothing is strange with this measurement (considering you have similar measurement as well as my secondary measurement method), then I am ok with it. While I might learn and do something on PowerPlayTables in the future to tweak the performance when I feel necessary.

Thank you very much.

The watts you measured on computer socket (110/220 VAC) has nothing to do with the watts calculated on Graphics Cards there mostly calculate watts based on 12V e.g 12 colts *12 amps = 144 watts.

Find more on these to links


Again take a look at PolarisBiosEditor output in Post 4
 
Joined
Jan 19, 2018
Messages
221
Motherboard
Gigabyte Z390UD
CPU
i9-9900
Graphics
UHD 630 + RX 580
Mac
  1. iMac
Mobile Phone
  1. Android
The watts you measured on computer socket (110/220 VAC) has nothing to do with the watts calculated on Graphics Cards there mostly calculate watts based on 12V e.g 12 colts *12 amps = 144 watts.

Find more on these to links


Again take a look at PolarisBiosEditor output in Post 4

Thanks for the reading resource. Intresting information for me.
What I see in the HWMonitorSMC2 about total power on my RX580 is (I assume) total board power which shows 145W in my above picture. Meaning, this is the instantaneous measured power consumption of the GPU board seen by macOS/HWSMC2, it draws 145W of power from the Power Supply.
Why I said this number might not be correct is because, I measure the power drawn by the power supply unit of the computer, meaning the total power consumption of the computer, including the GPU. If total power seen from the power supply reads at around 100W while GPU Power from HWMonitorSMC shows 145W for the GPu only, then the math does not add up here. Total power from the power supply should be bigger than what is measured from the GPU only, since that is how electricity works, i.e. PS supplies power to CPU, RAM, fans, RGB LED, and GPU.
Else, the Total Power shown in HWSMC2 is the maximum allowable power and not the instantaneous measured power (which I doubt this also, since it is moving with the load of the GPU especially when I am using it in single display configuration.

Thanks on the PolarisBIOSEditor here. I am learning on it to look at how I can tweak my RX580 to my liking.
 
Joined
Oct 19, 2013
Messages
51
Motherboard
Asus Prime Z270-A-0906 - OpenCore 0.63 - Big Sur 11.1
CPU
i7-7700K
Graphics
XFX Radeon RX 580 8GB GTX Core GDDR5 HDMI DP
Mac
  1. iMac
  2. MacBook Pro
Mobile Phone
  1. iOS
After change to OpenCore 0.64 and Big Sur 11.1 things changed a bid, most numbers dropped and occasionally even fan stops

ioreg -l |grep \"PerformanceStatistics\" | cut -d '{' -f 2 | tr '|' ',' | tr -d '}' | tr ',' '\n'|grep 'Temp\|Fan\|Clock\|%\|Power'
"Device Utilization %"=0
"Fan Speed(%)"=36
"Core Clock(MHz)"=667
"Memory Clock(MHz)"=2000
"GPU Activity(%)"=0
"Fan Speed(RPM)"=944
"Temperature(C)"=56
"Total Power(W)"=85
 
Joined
Oct 10, 2015
Messages
890
Motherboard
Z370 AORUS Gaming 3
CPU
i5-9600K
Graphics
RX 580
Mac
  1. MacBook
  2. MacBook Pro
Mobile Phone
  1. iOS
After change to OpenCore 0.64 and Big Sur 11.1 things changed a bid, most numbers dropped and occasionally even fan stops

ioreg -l |grep \"PerformanceStatistics\" | cut -d '{' -f 2 | tr '|' ',' | tr -d '}' | tr ',' '\n'|grep 'Temp\|Fan\|Clock\|%\|Power'
"Device Utilization %"=0
"Fan Speed(%)"=36
"Core Clock(MHz)"=667
"Memory Clock(MHz)"=2000
"GPU Activity(%)"=0
"Fan Speed(RPM)"=944
"Temperature(C)"=56
"Total Power(W)"=85
The fan of RX 580 stops at a temperature below 50ºC by default.
 
Top