Discussion Ryzen 3000 series benchmark thread ** Open **

Page 32 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Topweasel

Diamond Member
Oct 19, 2000
5,326
1,524
136
I was going to say a bunch of things about this being an artificial test. But really most of the stuff can be summed up below. In the end they are stair stepping the test through the cores and not working or analyzing anything as it would be scheduled through the Windows schedular or anything AMD would have control over.

But the science behind that as devices get hotter, they leak more power, requiring more power to maintain performance.

Look back at Gamers Nexus review of the Vega64/FE Edition. They found that the chip was using +50w more with base cooling than when cooled the card down other ways. They did the test to show how beyond the efficiency range and the available cool that AMD was taking the cards to hit clock speeds to be competitive with Nvidia. The point being is that we don't have all the information on their tests. But an under used CCD could be maintaining as good if not better clocks with power usage per core because the CDD is cooler to begin with. This is why when the cores are maxed out we tend to see a drop little by little for all core clocks over time. It is also why I think I see with an all core usuage a increase and temp and power about 5-6 minutes into a test. At that point I think that AMD sends more power to try to keep the clocks up for a little while longer before it has to start dropping it.

Besides testing methodology here. We don't actually have core clock information. Trying to analyze this other than to say the CPU hits peak power usage around 10 cores. Anything other than that is a fools errand.
 

joesiv

Member
Mar 21, 2019
74
24
41
But an under used CCD could be maintaining as good if not better clocks with power usage per core because the CDD is cooler to begin with. This is why when the cores are maxed out we tend to see a drop little by little for all core clocks over time. It is also why I think I see with an all core usuage a increase and temp and power about 5-6 minutes into a test. At that point I think that AMD sends more power to try to keep the clocks up for a little while longer before it has to start dropping it.
Good point on the efficiency drop at higher temperatures.
 

TheGiant

Senior member
Jun 12, 2017
666
282
106
That's system power, not CPU power.

With that said, both AMD and Intel allow for sustained boost levels at stock as long as your cooling solution and power delivery can handle it. That means that both 95W intel chips and 105W AMD chips will use ~140W all core if your setup can handle it. Additionally, both AMD and Intel allow for essentially unlocked power limits which will obviously use more power for higher sustained boosts. Some motherboards (especially on the intel side as of late) will unlock these power limits by default so CPU power use will depend largely on the motherboard you choose. This behavior has evolved over time and I think both Intel and AMD need to be more transparent about it (like listing both a base TDP and boost TDP) but this is where the unfortunate lack of traditional technology scaling has brought us.
yep I know, but that is what is important, the 12V rail is just informative
what I tried to tell was that unless yuu understand TDP spec from intel or AMD, the number itself doesnt tell you anything
9900K (91W TDP), 9920X (165W TDP), 3900X and 3950X (105W TDP) eat the same power
try to explain it to your wife :)
You got so caught up in trying to prove your point that you forgot about the 16 cores vs the 12 in the 3900X. Nothing special, eh? OK.
well enjoy the win, that is what you are here for isnt it?
if you read the techreport article, they say what I am saying...if you underclock the 3900X by 200MHz and undervolt to the levels of 3950X, it will have 30-40% less power consumption
so 16C 3950X configured in a way that draws power like 3900X (which is clocked and volted higher) is nothing special, that is the standard in the server world
I tried to do it with my 3900X and ....205W while video encoding wall power instead of 240W and loosing 200MHz, which I wont even notice
 

jpiniero

Diamond Member
Oct 1, 2010
7,929
1,209
126
  • Like
Reactions: lightmanek

Asterox

Senior member
May 15, 2012
333
330
136
Please, do comment. Some people don't want to watch a crappy video with crappy music just because and not even explain what it is trying to say but just giving numbers.
Well, you only need numbers or CPU sales numbers, or uber verkauft=sold over.

Mindfactory(the largest German PC retailer) has been mentioned frequently on this forum.


 
  • Like
Reactions: DarthKyrie

ASK THE COMMUNITY