Intel 13600K Using more power than a 2950X Threadripper?

[mbc0]

Hi, 

I just changed out my hardware from an AMD 2950X Threadripper on a Gigabyte Designaire X399 Mobo for an Intel 13600K on an Asrock Z790 Pro RS/D4 to find it is consuming more energy than before? 

Am I missing something? everything I read said the Intel would be more effecient to run than the AMD 

13600K 39W Idle

2950X 90W Idle 

 

Installed in the system...

1 X HBA Card with Expander

1 X Mellanox 10gbe Card

24 X Mechanical Hard Drives

3 X NVME

1 X NVidea GPU (Now Removed For 13600K)

1 X 4 Port NIC (Now Removed For 13600K)

 

With the 2950X I was using approx 150W idle with 3-4 hard drives spun up and 190-225 when being put to task.

With the 13600K I am using approc 250W idle with 3-4 hard drives spun up and 420-450 when being put to task

 

Quite the difference!! even though I have removed 2 cards as well I am consuming more power, have I missed something? I wanted to do the change for 2 reasons, Power Consumption & Quicksync but using more power makes no sense

 

Can anyone else with a 13600K share their experience? I am using a Tasmota Smart Socket to measure the power consumption.

 

Many thanks!

 

 

 

[Frank1940]

I3 or I5 or I9 processor?  As the 13600K is available in all three lines.  (I could also ask if you are overclocking this processor as that has an impact of power consumption...)

[Vr2Io]

Check BIOS setting and the CPU freq does clock down during light-load idle.

[mbc0]

38 minutes ago, Frank1940 said:

I3 or I5 or I9 processor?  As the 13600K is available in all three lines.  (I could also ask if you are overclocking this processor as that has an impact of power consumption...)

Sorry, I did not know that, it is an i5

[mbc0]

28 minutes ago, Vr2Io said:

Check BIOS setting and the CPU freq does clock down during light-load idle.

I did look in the BIOS but did not see anything like that apart from undervolting, I will check again tomorrow

[mbc0]

[Yankton]

you should be able to set power limits in the bios.  Worth looking into.

[mbc0]

Thank you, I have set the power limit to 125W I don't think power limits will help as it is not getting anywhere near stressed (see pics on my previous post) 

 

I cannot understand how it is consuming more power than my threadripper? as you can see it is sitting around 3%-5% load and even now with only 3 hard drives powered on it is drawing 215W?

 

I am using the same hba card, 10gbe nic and removed a graphics card and still using more power than before, Anyone got any ideas?

[Vr2Io]

Pls provide Tasmota whole screen dump, from your last pic. it indicate "efficiency" near 100% .... this almost impossible.

Edited February 5, 2023 by Vr2Io

[mbc0]

6 hours ago, Vr2Io said:

Pls provide Tasmota whole screen dump, from your last pic. it indicate "efficiency" near 100% .... this almost impossible.

 

The Effiency jumps around form 73% to 100 every refresh, not sure what it relates to?

 

[Yankton]

Do you have a kill-a-watt meter or a ups you can plug into and monitor?  Maybe the Tasmota is bad.

[mbc0]

No, but I do have a lot more Tasmota's I will try another, but the fact will still remain that it is using more than the Threadripper as it is the same plug.  I will let you know! thank you!

 

Edited February 5, 2023 by mbc0

[Vr2Io]

4 hours ago, mbc0 said:

The Effiency jumps around form 73% to 100 every refresh, not sure what it relates to?

If without calc. power factor ( efficiency ) then power reading will be larger than real power. Anyway I seldom found efficiency swing that much.

 

To identify does CPU consume huge power, you can set PL to 65w, if reading have ddro, that means it consume more then 65w, next try 40, 30 ..... if reading not drop significant, that means power draw not on CPU.

Edited February 5, 2023 by Vr2Io

[S80_UK]

I was reading the above and wondered about the power factor too.  Is this using the same power supply as the Threadripper build?  If it's not, then it might just be that the power factor is different in the new build and the Tasmota device (not sure what that is btw) is reading VA rather than true power in Watts.  Some power supplies only manage decent power factor at reasonable loads, and at light loads the readings might be quite misleading.  Using a kill-a-watt kind of device that can read both Watts and VA might help clarify what's going on.   

[Decto]

 

250W idle is insane

 

The CPU temp is only 1C higher than the motherboard temp, that would confirm to me that this isn't the CPU drawing any signficant power.

I'd question if the kernal has full support for power saving for Z790 / 13xxx yet, so may be a power state issue for connected peripherals.

I've always been a generation or 2 behind to make sure I get decent Linux support.

 

Can you detect anything getting hot in the case, 200W doesn't disappear, are 'spun down' HDD actually cold.

Do you have additional drives / devices connected, but not in the array. 

Do you have a UPS inline?

 

Can you remove the HBA / Expander, remove the power to the HDD's then start the system (not the array).

Give it a few minutes to idle.

That should give you a baseline for the platform, I'd expect this to be 50W or lower, even my XEON E5 V3 (10C)  idles at 35W for just the board, memory, cooler with an 850W PSU.

 

 

 

[mbc0]

Hi @Decto

 

All the cards/drives were installed with the Threadripper motherboard/cpu, I have removed 2 pci cards (GPU & 4 Port NIC) as well so the new motherboard and 13600k are in place in the same system, same smart socket (that monitors the power) as before.

I have checked the HBA and extender at 14W, you are correct about the cpu temp and this is because the CPU is just barely ticking over as it is running incredibly light tasks.

I do not unfortunately have a UPS but will be looking at an alternative method for measuring power draw although whatever it is, it is still srawing more power than the old threadripper setup as I am using the same mornitor smart socket.  

 

Currently, as I type I am drawing 198W with.....

3 active Hard Drives (Mechanical)

Asrock Z790 Pro RS/D4

Intel i5 13600K CPU

4 X Corsair 8GB RAM DDR4

Noctua CpU air cooler

1 X NVME Samsung 980 Pro 1 TB

2 X NVME Rocket 1TB

1 X HBA Card with Expander

1 X Mellanox 10gbe Card

24 X Mechanical Hard Drives (21 inactive)

EVGA Gold 650W PSU

It is in a 4U rackmount with 3 large fans that cool the 24 hard drives but they are spinning slowly and again, were there before I removed the threadripper

 

I was pretty sure this would come in less that 100W so maybe you are right regarding the kernel? nothing is hot within the case

 

 

[mbc0]

3 hours ago, S80_UK said:

I was reading the above and wondered about the power factor too.  Is this using the same power supply as the Threadripper build?  If it's not, then it might just be that the power factor is different in the new build and the Tasmota device (not sure what that is btw) is reading VA rather than true power in Watts.  Some power supplies only manage decent power factor at reasonable loads, and at light loads the readings might be quite misleading.  Using a kill-a-watt kind of device that can read both Watts and VA might help clarify what's going on.   

 

Hi, as described in my response just now in more detail, basically EVERYTHING is the same except the motherboard/cpu and the removal of 1XGPU & 1X4PORT NIC

[Frank1940]

You might try installing the Tips and Tweaks plugin.  It does have some settings for Power Saving Settings. 

 

[mbc0]

@Frank1940 You are on a roll!!! 

 

Thank you so much mate, enabled power saving in tips and tweaks and now down to 110W!!! Thank you buddy!

[Decto]

54 minutes ago, mbc0 said:

Hi @Decto

 

All the cards/drives were installed with the Threadripper motherboard/cpu, I have removed 2 pci cards (GPU & 4 Port NIC) as well so the new motherboard and 13600k are in place in the same system, same smart socket (that monitors the power) as before.

I have checked the HBA and extender at 14W, you are correct about the cpu temp and this is because the CPU is just barely ticking over as it is running incredibly light tasks.

I do not unfortunately have a UPS but will be looking at an alternative method for measuring power draw although whatever it is, it is still srawing more power than the old threadripper setup as I am using the same mornitor smart socket.  

 

Currently, as I type I am drawing 198W with.....

3 active Hard Drives (Mechanical)

Asrock Z790 Pro RS/D4

Intel i5 13600K CPU

4 X Corsair 8GB RAM DDR4

Noctua CpU air cooler

1 X NVME Samsung 980 Pro 1 TB

2 X NVME Rocket 1TB

1 X HBA Card with Expander

1 X Mellanox 10gbe Card

24 X Mechanical Hard Drives (21 inactive)

EVGA Gold 650W PSU

It is in a 4U rackmount with 3 large fans that cool the 24 hard drives but they are spinning slowly and again, were there before I removed the threadripper

 

I was pretty sure this would come in less that 100W so maybe you are right regarding the kernel? nothing is hot within the case

 

 

 

Hi,

 

The UPS question was in case you had a UPS inline and it was pulling excess current to charge a failing battery.

 

How are the drives powered? are they all fed from the EVGA or do you have a powered drive shelf etc.

 

I think there are two main options

 

1) the Tasmota is giving a totally incorrect power reading. Can you read this from any other device, smartphone / web interface etc?

I'd swap this out or add a second power meter inline to validate. Even a $10 plug in meter will differentiate between 50W and 200W.

 

2) The HDD's are idle but not in power save. Typcially HDD are approx 5W in idle so that is easily 120W or 140W from the wall. 

In power save they would be ~ 0.8W or ~ 20W total hence the question of what happens if you remove the HDD power.

 

 

 

 

[Decto]

7 minutes ago, mbc0 said:

@Frank1940 You are on a roll!!! 

 

Thank you so much mate, enabled power saving in tips and tweaks and now down to 110W!!! Thank you buddy!

 

That's more sensible for 24 drives!

[mbc0]

2 minutes ago, Decto said:

 

How are the drives powered? are they all fed from the EVGA or do you have a powered drive shelf etc.

All Powered by the single PSU in the rackmount

 

2) The HDD's are idle but not in power save. Typcially HDD are approx 5W in idle so that is easily 120W or 140W from the wall. 

In power save they would be ~ 0.8W or ~ 20W total hence the question of what happens if you remove the HDD power.

I did not know that! so if they are spun down they are still using approx 5W? or do you mean just spinning and not read/writing?

 

 

 

 

 

[Decto]

18 minutes ago, mbc0 said:

 

 

Idle is still spinning but not active

Powersave is spun down