Aida is lying?

[WildCAt]

Hi all! ^^

 

I've looked into AIDA64 Engineer "sensors" section and during idle 12V voltage value is 11.8V

And ingame with OC it's 11.3, sometimes even 11.28 or so 

wut?  

NZXT CAM says, there is pure 12V on motherboard. Maybe it's rounding, but 11.3 ~ 11, and there is no 11 in MIN section

PSU is Silverstone Strider Plus 600W bronze

 

I've heard that AIDA lies sometimes. I haven't experienced any issues, OC is good, power is stable.

So, is Aida lying or I need to start worrying?

 

 

[AlwaysFSX]

Your 12V isn't going to run at exactly 12V all the time. It can vary. You can also have a bad reading with multiple programs open that read the same sensor.

[W-L]

2 minutes ago, WildCAt said:

-SNIP-

Software based voltage readings are never accurate so they aren't very reliable, only way to check would be to wire up a multimeter to the PSU to get readings directly.

[PRTI]

Its not going to run 12V at every millisecond 

[Energycore]

It's not unheard of for PSUs to let the 12V rail drop to 11.8V.

 

As for 11.3, probably not, ATX spec maximum drop is 11.4V

[WildCAt]

Also, GPU +12V is 11.4

BTW, screenshot:

 

[Energycore]

1 minute ago, rrubberr said:

Haha, spec. That power supply isn't even within spec if it's a single rail.

I'm afrad I don't follow.

[Energycore]

Just now, rrubberr said:

Spec means nothing to most PSU and mobo manufacturers

[citation needed]

[Energycore]

2 minutes ago, rrubberr said:

http://www.formfactors.org/developer/specs/atx12v psdg2.01.pdf

 

I think you mentioned the ATX spec by name, earlier, but here's what a PSU made to spec looks like.

 

That link is from the ATX standard version 2.01, published June 2004.

 

The latest version is 2.38 of mid-2008

[Energycore]

Just now, rrubberr said:

I can't seem to find on the latest Intel spec sheet where it says over 18 amps is permitted on a simgle 12v line.

Eh that's fine. Most mobos have UVP which will shut the computer down at 11.3V so I've a feeling that reading is wrong

[quan289]

23 minutes ago, rrubberr said:

"System components that use 12V are continuing to increase in power. In cases where expected current requirements is greater than 18A a second 12 V rail should be made available."

 

Power supplies are all so far out of spec it isn't even funny. It's not that I care, but it's funny to argue over spec compliance when the thing that isn't complying is out of spec by nature.

Hmm...The current specs is not ATX12V 2.01. It's ATX12V 2.4 which is based on the Revision 1.31 of the Design Guide for Desktop Platform Form Factors that was release in April 2013. The PDF file had been removed for some reason but just paying attention to some recent changes shows why and when it was change. In this case, in order for a PSU to be truly compatible with Haswell (line of CPU release in 2013) by Intel, the voltage must stayed in specs when you load the minor rails to it max and put a 0.15A load on the 12V rail (many group-regulated PSU would not passed this). This is why you have PSU that is compliant to the "ATX 2.4 / EPS 2.92" standard like the Fractal Design Integra M: http://www.fractal-design.com/home/product/power-supplies/integra-m/integra-m-450w

 

Regarding the 240VA limit that you had mentioned.

The ATX12V 2.01 guide you linked to from 2004 as well as the ATX12V 2.2 guide from 2005 specifically states under the Over Current Protection section on page 27 and 29 respectively that "Overload currents applied to each tested output rail will cause the output to trip before reaching or exceeding 240 VA. For testing purposes, the overload currents should be ramped at a minimum rate of 10 A/s starting from full load".

 

That's 20A on the 12V rail.

 

Then the Revision 1.1 guide or ATX12V 2.3 that was release in 2007 quietly removed the 240VA limit. They did, however, brought the typical power distribution for PSU up to 450w that may have followed the 240VA but it is not required to meet the design guide - only recommended to follow.