What is the effect of CPU VID on Overclocking?

Status
Not open for further replies.
G

gamepagol

Contributor
recently i just found some data about CPU Vcore i.e. CPU core voltage set in BIOS (variable on overclocking) and VID i.e. default CPU core voltage set by Intel (fixed on any case).... now coming to the question in short.... let us think two system showing diff' VID (fixed) on CoreTemp (still dnt know whether its showing VID or not or even if its VID does it showing correct value or not)

1. Q6600 G0 ----- VID:1.1125v
2. Q6600 G0 ----- VID:1.3250v

everything else are entirely same for say, now my question which one will go better on overclocking or does both will go same or does it entirely on the expertise of the overclocker or does it depends on some other factors?:huh:
 
^^the first one will overclocking much much better than the 2nd one

basically there is a max voltage you can go to....because temperatures become an issue

for any clock speed, the first one will be at lower voltage then the 2nd one which means that for the same max voltage, it will be higher clocked

the temperature depends more on the voltage than the clock speeds btw...
 
gamepagol said:
recently i just found some data about CPU Vcore i.e. CPU core voltage set in BIOS (variable on overclocking) and VID i.e. default CPU core voltage set by Intel (fixed on any case).... now coming to the question in short.... let us think two system showing diff' VID (fixed) on CoreTemp (still dnt know whether its showing VID or not or even if its VID does it showing correct value or not)

1. Q6600 G0 ----- VID:1.1125v
2. Q6600 G0 ----- VID:1.3250v

everything else are entirely same for say, now my question which one will go better on overclocking or does both will go same or does it entirely on the expertise of the overclocker or does it depends on some other factors?:huh:
Click to expand...

On the very same motherboard, if those voltages are reported by coretemp or the differences in the voltages being the same as reported from other monitoring s/w, then,

U can bet the 1st has a better headroom for overclocking. There's a limit to push those voltages.

And where did u see a Q6600 G0 having a VID of 1.1125v???
 
^^ Read the thread properly. I didn't get to the second page, but the lower voltages aren't due to brilliant processors.

Speedstep drops the CPU voltage (VID) so if you manage to open coretemp or CPU-Z at the time when the CPU is idle, you'll get a whack reading, resulting in readings below the correct voltage.

The only way to take an accurate reading is to load the CPU to 100% using TAT and then take a reading from the software or motherboard.

The BIOS is programmed to identify CPU and set the operating voltage correctly. This usually results in a fixed range of VID on the same board/processor combo if you're not overriding it for overclocking. You'll notice the good guys are taking proper shots with Orthos, with readings ~1.4-1.5V under overclocked conditions.

That said, the overclockability of processors is related to their sensitivity to voltage. In general a better processor will hit a higher clock at lower voltage, all other conditions remaining equal. There are exceptions and other permutations, but let's not get into that area. The best clocks are achieved by overclockers, not CPUs. Experience will trump equipment every time.
 
sangram said:
^^ Read the thread properly. I didn't get to the second page, but the lower voltages aren't due to brilliant processors.

Speedstep drops the CPU voltage (VID) so if you manage to open coretemp or CPU-Z at the time when the CPU is idle, you'll get a whack reading, resulting in readings below the correct voltage.

The only way to take an accurate reading is to load the CPU to 100% using TAT and then take a reading from the software or motherboard.
Click to expand...

but that actually didn't happnd with me not at least in my board... though i hv heard that Intel Speedstep has no effect on VID but as u r said i run prime95 and take a shot ... here u can see Speedstep is working fine...

ON load


Idle


so whats wrong in it?
 
^^for some reason i cant see those pics u just posted

the speedstep does reduce the voltage at idle thought, if you want ti to be permenantly higher clocked you disable speedstep and C1E

waise my bios reads the VID mentioned by the proccy, i think its fixed to the same value for all preccys...not sure though
 
vID (max) read by coretemp is the one that is programmed by intel itself after uP testing (as mentioned in link you gave). Itll be constant irrelevant of load.This is the max value which intel thinks the processor needs,to run at stock speed (without power saving features and other variances).

what CPU-Z reads is the actual voltage being fed to core at an instant ,which varies according to load, power saving tech and other factors like droop et al.

EDIT:just noticed your cpuz reads VID instead of core voltage
 
^ he has it,lesser VID means less voltage for the cpu at stock speed,and theoretically oveclocking a lesser VID chip should yield similar overclocks at lower core voltage than a chip with higher VID.
though practically,it isn't that straightforward,for a higher VID chip may go on to post a higher overclock on its stock voltage while you might need to up the voltage on the lesser VID chip just to go pass stock levels.so basically it boils down to searching forums and finding out what chips generally overclock better. ;)
 
CPU with lower VID may OC better

but it doesnt guarantee U cool temps.

Coz many experts believe that larger difference between stock and OCed vCore means higher temps, eg. U have 2 CPU, 1 = 1.225, 2 = 1.285 and both require 1.345v for say 3GHz then CPU with 1.285 will run cooler coz its only .06v bump.
 
sangram said:
That said, the overclockability of processors is related to their sensitivity to voltage. In general a better processor will hit a higher clock at lower voltage, all other conditions remaining equal. There are exceptions and other permutations, but let's not get into that area. The best clocks are achieved by overclockers, not CPUs. Experience will trump equipment every time.
Click to expand...

You are wrong there sangram, ofcourse between novice OCers and experienced OCers it will be always the experienced guy who will win. But between 2 relative experienced OCers its always about the chip and other surrounding hardware. Some of the well known OCers go through 100's of chip to find their Gem.
 
True, but an experienced OC'er can push a given chip farther than a newbie (like me). The highest clocking CPUs are a matter of luck, but a veteran can push even a dog quite far. A novice will not be able to fully exploit a 'Gem' either. And that is what I meant.
 
Status
Not open for further replies.
Top Bottom