Intel's Vdroop Explained

[dasickninja]

Just a little read for those interested.

Linkage

 

[kwalker]

Nice find
Thanks Ninja 8)

 

[dasickninja]

Always happy to find a decent read or two. Half the stuff goes over my head though...

 

[dasickninja]

Prozac, buddy, nice to see you journey to upper hell. I see you've broken 10K. Going to post a celebratory Babe of the Month?

Yeah. Wusy recommended that site to me. Top notch when it comes to CPU stuff. Packed end to end with industry insiders that get the low down before most others do.

 

[SirCrono]

Thx for the link, it was quite useful

I don't like your new avatar though, it's kinda... sucky.

Thx again!

 

[dasickninja]

It stays till my birthday.

 

[kwalker]

Always happy to find a decent read or two. Half the stuff goes over my head though...

That means the other half is retainable.
I usually have to re read these articles a few times to comprehend the data.
Some people soak it up like a sponge.
The bottom line of the article is Voltage doesn’t necessarily degrade the circuitry, but current flowing through the circuit can destroy components unregulated or modified.
Think of it like a shock absorber on your vehicle.
Less wear and tear on components and a smoother performance curve.

 

[dasickninja]

Oh... [/simpleton]

 

[Dante_Jose_Cuervo]

Interesting find, ya long time... I haven't been on the forumz for a while.

 

[akahuddy]

Good find Ninja.

A good E&M Physics course goes a long way to understanding all of the stuff in there.

 

[1Tanker]

Just a little read for those interested.

Linkage

Nice article. I may have misunderstood, but is he saying that vDroop is designed-in for Intel processors, and is actually needed? Because, i have always seen that ASUS boards have the worst vDroop....but maybe that's not such a bad a thing then? :?

 

[SuperFly03]

I think it has to do with his new avatar. Just my guess.

Ok I read it... and most of it went over my head. Honestly, I am not an engineer nor do I have a good grasp on electrical subsystems. That is beyond my current skill level. I do understand the basic concept... its cheaper lol.

My problems is I have to keep my vcore at 1.41875 (CPU-Z reports 1.44v 8O ) to achieve 1.375vcore underload on my current board (which is outside the standards for the board, so I am exchanging it). So it seems to me it causes problems when overlocking, but they don't design processors to be OC'd necessarily.

Can't wait for my replacement RD600... back to .002v vdroop.

DFI FTW!

 

[kwalker]

The problem you’re describing is inherent among certain motherboard designs.
The ability to maintain the current flow through the circuitry as described in the article by FCG expresses this design concept very clearly.
Although a little on the dry side for a lot of readers the point still comes across.
The Vcore set at 1.44 then drooping to 1.375 under load is a droop of .065 volts.( a lot in comparison to most)
This could be caused by poor circuitry design or even an insufficient PSU or not so good CPU.
The other possibility is that the regulator circuitry is doing its job keeping the VCC at a constant.
This is where one motherboard to another can make all the difference in the world in over clocking.
The Intel based chipset boards are more likely to have better regulatory circuitry as per Intel guidelines where as another developer may skimp on this area and offer another aspect of importance like ram over clocking or SLI.
As for your Voltages being correct
The only way to actually tell is by using a digital volt meter and an oscilloscope with a logic pod to trace these frequencies and monitor the duty cycle.
Or you could use the overclockers tools we all use, make the system stable, and call it a day.
Oh yes
I almost forgot.
All CPU’s are made to overclock. :tongue:

 

[dasickninja]

What is the deal --- who the heck did you p!ss off?

The operative question is who didn't I piss off...

@kwalker
Thanks for the explanation.

 

[TehRokHead]

I have been SEARCHING for this for awhile. Even though I have a EE background, I haven't used that stuff in over 10 years. However, I am familiar with most of the theory behind this document. Never the less, this will seriously help me to overclock my board in a way that I actually know what I'm doing.

Thanks again.

TehRokHead.

 

[djcoolmasterx]

Its from 2007 so make sure you have the up to date info...