![]() |
Taichi X570 wrong Total Socket Power (PPT) |
Post Reply ![]() |
Page 123> |
Author | |
foxx1337 ![]() Newbie ![]() ![]() Joined: 19 Jul 2019 Location: Munich Status: Offline Points: 89 |
![]() ![]() ![]() ![]() ![]() Posted: 19 Jul 2019 at 5:10am |
According to software such as Ryzen Master or HWiNFO64, the total socket power (PPT) on my Ryzen 3900X is about half of its real value.
Do I have a deffective CPU or motherboard? When I fully load the CPU, the reported PPT is 50-52% out of 142W in Ryzen Master while HWiNFO64 reports circa 60 watt power. As I know, the 3900X is a 105 watt processor. |
|
Ryzen 3900X, Taichi X570
|
|
![]() |
|
kschendel ![]() Newbie ![]() Joined: 21 Aug 2018 Status: Offline Points: 86 |
![]() ![]() ![]() ![]() ![]() |
While I don't know what is going on in your specific situation, don't confuse power input with TDP (Thermal Dissipation Power). The 3900X is a (nominally) 105w TDP chip, meaning that at peak it can be expected to emit about 105 watts of heat. Actual power to the chip may be more or less depending on what's going on.
|
|
![]() |
|
foxx1337 ![]() Newbie ![]() ![]() Joined: 19 Jul 2019 Location: Munich Status: Offline Points: 89 |
![]() ![]() ![]() ![]() ![]() |
Clear. Still, my 3900X runs full burn at 50% out of 142W PPT, while 3700X-es I could find in screenshots or videos burn at around 100% out of 88W PPT. |
|
Ryzen 3900X, Taichi X570
|
|
![]() |
|
garych ![]() Newbie ![]() Joined: 09 Dec 2017 Status: Offline Points: 62 |
![]() ![]() ![]() ![]() ![]() |
TDP is thermal design* power
|
|
![]() |
|
![]() |
|
garych ![]() Newbie ![]() Joined: 09 Dec 2017 Status: Offline Points: 62 |
![]() ![]() ![]() ![]() ![]() |
Anyway, what's your VRM current output reading?
I assume HWiNFO64 displays that just like on X370 Taichi or similar. |
|
![]() |
|
![]() |
|
foxx1337 ![]() Newbie ![]() ![]() Joined: 19 Jul 2019 Location: Munich Status: Offline Points: 89 |
![]() ![]() ![]() ![]() ![]() |
Hey garych. It's 33 Amps, as reported by Ryzen Master in the TDC field.
Here's a screenshot - https://imgur.com/a/UZRd5wa - I took yesterday at the same time with running CineBench R20. Both PPT and TDC are at around half of what I FEEL they SHOULD be (I might well be wrong here, I'm totally clueless about modern hardware, this is why I'm asking for help). This is what Ryzen Master looks like during the same test run: https://imgur.com/a/8nXjor0 The last gauge there, the EDC, is probably correct. The guys at Gamers Nexus have a very different picture in their Ryzen 3900X test here: https://www.youtube.com/watch?v=B7NzNi1xX_4&t=11m28s (the video is linked at the 11:28 exactly on Ryzen Master while running CineBench R20). To compare the numbers: - Gamers Nexus result: ~7000, same as me - clock: ~ 4 GHz, same as me - temperature: they get 85 Celsius, I get 82, same as me - PPT: they get 18% of 1000w = 180w, I see 42% of 142w = 60w (!!!) - TDC: they get 24% of 490A = 118A, I get 35% of 95A = 33A (!!!) - EDC: they get 25% of 630A = 158A, I get 66% of 230A = 152A, same as me |
|
Ryzen 3900X, Taichi X570
|
|
![]() |
|
foxx1337 ![]() Newbie ![]() ![]() Joined: 19 Jul 2019 Location: Munich Status: Offline Points: 89 |
![]() ![]() ![]() ![]() ![]() |
Thing is - AMD first line of support told me to fill an RMA form with them. But I don't think the processor is broken. I also don't think the Taichi is, at most some firmware bug somewhere in between the 2. If I run Prime95 small FFT test I INSTANTLY reach 95 degrees celsius at ~4 ghz all core and ~"71W" Total Socket Power (PPT) with ~"33A" of Sustained Current Limit (TDC). This would make my 3900X half as power hungry as what Gamers Nexus have and also around half of what AMD are marketing, while at the same performance. Due to it not even coming close to the 142 watt PPT, it never stops asking for more since it's halfway through the socket power limitation, but it does stop due to the other ceiling - the 95 Celsius temperature limit. As I understand from a different Gamers Nexus video where they explain PBO, the PPT number is baked into the mobo, while AMD recommend 142 watts for 105 watt TDP Zen 2s. |
|
Ryzen 3900X, Taichi X570
|
|
![]() |
|
garych ![]() Newbie ![]() Joined: 09 Dec 2017 Status: Offline Points: 62 |
![]() ![]() ![]() ![]() ![]() |
Scroll down in HWiNFO64 and you'll see something like this [URL=https://i.imgur.com/6k75bOw.png]https://i.imgur.com/6k75bOw.png[/URL]
It will show you the current and power readings from vrm by motherboard. They might not be accurate, but probably close to actual power draw. |
|
![]() |
|
![]() |
|
foxx1337 ![]() Newbie ![]() ![]() Joined: 19 Jul 2019 Location: Munich Status: Offline Points: 89 |
![]() ![]() ![]() ![]() ![]() |
Today I learned! Here's mine, looks credible - https://imgur.com/a/5UcjYXr So what's borked in my case then? Software bug in the Taichi BIOS, the 3900X, or the Taichi itself? Or maybe somebody forgot and they're just showing one CCD (out of the two inside this processor) in the HWiNFO64 CPU package power and Ryzen Master PPT readouts? |
|
Ryzen 3900X, Taichi X570
|
|
![]() |
|
garych ![]() Newbie ![]() Joined: 09 Dec 2017 Status: Offline Points: 62 |
![]() ![]() ![]() ![]() ![]() |
I don't know what might be the issue here, but on X370 Taichi there's similar problem with Ryzen 7 1700 power draw readings. But instead it would show ~50% higher power and current draw than one would expect, while on other boards it's working properly.
I also noticed that your VCore and SOC VRM input Power looks strangely identical. Might be HWiNFO64 bug or motherboard not reporting correctly. |
|
![]() |
|
![]() |
Post Reply ![]() |
Page 123> |
Tweet
|
Forum Jump | Forum Permissions ![]() You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You cannot vote in polls in this forum |