How high a voltage can a 2500+ Barton take?

[bupkus]

I set the vcore voltage to 1.65 for an OC to 2.2GHz, but the bios reads it as 1.71.
Question: how high can I safely take the voltage?

 

[Shimmishim]

i've run mine as high as 1.95 on air.

these things are durable.

 

[bupkus]

I tried about 1.7+ but just became more unstable. Had to drop fsb to 166.

 

[Mogadon]

I run mine with 1.95V. The recommended max would be 2.035V (10% more than 1.85 which is their rated max). There's a guy on this board, can't remember who that runs his at 2.05V constantly.

 

[BigCoolJesus]

Really, it all depends

as stated, 99% of the time, the maxiumum is 2.03v or so....but thats not to say you dont get a chip that is "special" and can run higher


its like the Venices, most people run thiers at 1.6v at the most, but me (and a few others ive talked to) can easily go to 1.75v without frying the chip or having cooling issues
and then there are some who cant even get 1.5v


The only way to find the maximum is to set it at it, but then youll risk frying the chip (youll know you crossed the thresh hold when your CPU stops working :beer: )

 

[RichUK]

Originally posted by: BigCoolJesus

its like the Venices, most people run thiers at 1.6v at the most, but me (and a few others ive talked to) can easily go to 1.75v without frying the chip or having cooling issues
and then there are some who cant even get 1.5v

WTF are you talking about

 

[BigCoolJesus]

Originally posted by: RichUK

Originally posted by: BigCoolJesus

its like the Venices, most people run thiers at 1.6v at the most, but me (and a few others ive talked to) can easily go to 1.75v without frying the chip or having cooling issues
and then there are some who cant even get 1.5v

WTF are you talking about

Voltages, on Venice chips
and how most people who raise them only go to a maximum of around 1.6V, but there are some who can go higher without any damage or high temps (such as me, and theres also someone who runs his at 1.82v constantly)
And how there are also others who cant raise theres past 1.5v without heat problems, or instability (very few though)

Im trying to show that CPU's cores can be different from one to another (so one Barton may do 2.0 and another may only do 1.9, its all a matter of how lucky you are)

 

[RichUK]

Originally posted by: BigCoolJesus

Originally posted by: RichUK

Originally posted by: BigCoolJesus

its like the Venices, most people run thiers at 1.6v at the most, but me (and a few others ive talked to) can easily go to 1.75v without frying the chip or having cooling issues
and then there are some who cant even get 1.5v

WTF are you talking about

Voltages, on Venice chips
and how most people who raise them only go to a maximum of around 1.6V, but there are some who can go higher without any damage or high temps (such as me, and theres also someone who runs his at 1.82v constantly)
And how there are also others who cant raise theres past 1.5v without heat problems, or instability (very few though)

Im trying to show that CPU's cores can be different from one to another (so one Barton may do 2.0 and another may only do 1.9, its all a matter of how lucky you are)

Im trying to show that CPU's cores can be different from one to another (so one Barton may do 2.0 and another may only do 1.9, its all a matter of how lucky you are)

The motherboard/BIOS denotes what voltage is used not the CPU, and if you are referring to an overclock, then what you mean to say is you are lucky if you have a good performing core.

but me (and a few others ive talked to) can easily go to 1.75v without frying the chip

And the above was with regards to the venice .. and no you wont fry a chip at 1.75 nor at 1.82

 

[Fern]

Originally posted by: bupkus
I set the vcore voltage to 1.65 for an OC to 2.2GHz, but the bios reads it as 1.71.
Question: how high can I safely take the voltage?

The stock voltage for a (desktop) Barton IS 1.65v.

How much vcore you can apply is directly related to cooling and ambient temps. I don't like mine above 60c. Some of these chip exhibit instability above 60c.

Fern

 

[Mogadon]

The stock voltage is 1.65, but the max it should be taken to, according to AMD is 1.85, i am under the impression that when people are talking about the 10% safe voltage above the norm, they are talking about 10% above the max recommended. I've been running my Barton at 1.95 for at least a year.

 

[RichUK]

Originally posted by: Mogadon
The stock voltage is 1.65, but the max it should be taken to, according to AMD is 1.85, i am under the impression that when people are talking about the 10% safe voltage above the norm, they are talking about 10% above the max recommended. I've been running my Barton at 1.95 for at least a year.

within the AMD whitepapers it actually states that a less than 10% increase in voltage is perceived as acceptable . this is to accommodate for the different stock voltage produced by different motherboards, as some are higher and some are lower then what they should be.

 

[BigCoolJesus]

Originally posted by: RichUK

Originally posted by: BigCoolJesus

Originally posted by: RichUK

Originally posted by: BigCoolJesus

its like the Venices, most people run thiers at 1.6v at the most, but me (and a few others ive talked to) can easily go to 1.75v without frying the chip or having cooling issues
and then there are some who cant even get 1.5v

WTF are you talking about

Voltages, on Venice chips
and how most people who raise them only go to a maximum of around 1.6V, but there are some who can go higher without any damage or high temps (such as me, and theres also someone who runs his at 1.82v constantly)
And how there are also others who cant raise theres past 1.5v without heat problems, or instability (very few though)

Im trying to show that CPU's cores can be different from one to another (so one Barton may do 2.0 and another may only do 1.9, its all a matter of how lucky you are)

Im trying to show that CPU's cores can be different from one to another (so one Barton may do 2.0 and another may only do 1.9, its all a matter of how lucky you are)

The motherboard/BIOS denotes what voltage is used not the CPU, and if you are referring to an overclock, then what you mean to say is you are lucky if you have a good performing core.

but me (and a few others ive talked to) can easily go to 1.75v without frying the chip

And the above was with regards to the venice .. and no you wont fry a chip at 1.75 nor at 1.82

im not referring to an overclock, im referring to the voltage. Even though the mobo sets the voltage (and by the way im not dumb, so spare me the simple things you state), but that doesnt mean the CPU will be able to run on that voltage.
Some people can set the voltage to 1.5 and have crazy ass temps of 55C or whatever, some can set it at 1.6 and fry thier cpu, and some can go even redicuously high without seeing any noticeable temp increase


all im trying to stress is the MAXIMUM vcore that a cpu (any core) will support is not a constant

 

[Mogadon]

Originally posted by: RichUK

Originally posted by: Mogadon
The stock voltage is 1.65, but the max it should be taken to, according to AMD is 1.85, i am under the impression that when people are talking about the 10% safe voltage above the norm, they are talking about 10% above the max recommended. I've been running my Barton at 1.95 for at least a year.

within the AMD whitepapers it actually states that a less than 10% increase in voltage is perceived as acceptable . this is to accommodate for the different stock voltage produced by different motherboards, as some are higher and some are lower then what they should be.

Thanks for the clarification Rich, I guess I don't read as much as I should.

 

[RichUK]

ok originally i quoted you because what you had stated did not make sense, and was quite incorrect.

for example:

Some people can set the voltage to 1.5 and have crazy ass temps of 55C or whatever, some can set it at 1.6 and fry thier cpu, and some can go even redicuously high without seeing any noticeable temp increase

i mean seriously WTF is that statement supposed to mean, i am not trying to ridicule you it just doesn?t make sense its just an highly inaccurate statement on many levels, let me tell you something 1.6v will not fry a CPU .. even if you take off the heatsink.

and this one as well

all im trying to stress is the MAXIMUM vcore that a cpu (any core) will support is not a constant

 

[RichUK]

Originally posted by: Mogadon

Originally posted by: RichUK

Originally posted by: Mogadon
The stock voltage is 1.65, but the max it should be taken to, according to AMD is 1.85, i am under the impression that when people are talking about the 10% safe voltage above the norm, they are talking about 10% above the max recommended. I've been running my Barton at 1.95 for at least a year.

within the AMD whitepapers it actually states that a less than 10% increase in voltage is perceived as acceptable . this is to accommodate for the different stock voltage produced by different motherboards, as some are higher and some are lower then what they should be.

Thanks for the clarification Rich, I guess I don't read as much as I should.

yeah no worries, you were right as well, thats just what AMD officially state are the safe parameters, although it is safe to increase it more temperature permitting

 

[BigCoolJesus]

Originally posted by: RichUK
ok originally i quoted you because what you had stated did not make sense, and was quite incorrect.

for example:

Some people can set the voltage to 1.5 and have crazy ass temps of 55C or whatever, some can set it at 1.6 and fry thier cpu, and some can go even redicuously high without seeing any noticeable temp increase

i mean seriously WTF is that statement supposed to mean, i am not trying to ridicule you it just doesn?t make sense its just an highly inaccurate statement on many levels, let me tell you something 1.6v will not fry a CPU .. even if you take off the heatsink.

and this one aswel

all im trying to stress is the MAXIMUM vcore that a cpu (any core) will support is not a constant

your not getting it are you?

first off, yes, 1.6v can (freak accident, but it has happened) fry a cpu if you got a crappy CPU......not all CPU's can handle the same vcore (just like not all CPU's can handle the same clock speeds, each is built "differently")


I dont see what so confusing about my last statement, that vcore is not a constant among CPU's..........because it isnt (you know what a constant is, right? ie, vcore isnt the same for every single cpu)
There are some people on here with overclocks of 2.6GHz on a 1.58vcore and they have temps of around 50-52C.......and then there are some who run 1.64v and have temps of 45-48C
IT ALL DEPENDS ON THE LUCK OF THE DRAW

1.6v will not fry a CPU .. even if you take off the heatsink.

Also, you seem a little confused yourself, as the heatsink has nothing to do with whether the Vcore will fry (ie, electrically kill) your CPU

 

[RichUK]

Originally posted by: BigCoolJesus

Originally posted by: RichUK
ok originally i quoted you because what you had stated did not make sense, and was quite incorrect.

for example:

Some people can set the voltage to 1.5 and have crazy ass temps of 55C or whatever, some can set it at 1.6 and fry thier cpu, and some can go even redicuously high without seeing any noticeable temp increase

i mean seriously WTF is that statement supposed to mean, i am not trying to ridicule you it just doesn?t make sense its just an highly inaccurate statement on many levels, let me tell you something 1.6v will not fry a CPU .. even if you take off the heatsink.

and this one aswel

all im trying to stress is the MAXIMUM vcore that a cpu (any core) will support is not a constant

your not getting it are you?

first off, yes, 1.6v can (freak accident, but it has happened) fry a cpu if you got a crappy CPU......not all CPU's can handle the same vcore (just like not all CPU's can handle the same clock speeds, each is built "differently")


I dont see what so confusing about my last statement, that vcore is not a constant among CPU's..........because it isnt (you know what a constant is, right? ie, vcore isnt the same for every single cpu)
There are some people on here with overclocks of 2.6GHz on a 1.58vcore and they have temps of around 50-52C.......and then there are some who run 1.64v and have temps of 45-48C
IT ALL DEPENDS ON THE LUCK OF THE DRAW

Ok .. you are talking about different CPU's, as in different cores and different tolerances right?

if so you should state that in your text, as you are not very clear on some points, that you are infact talking about differrent cores. Ok so you are talking about different cores thats a given ..

But once again this statement:

first off, yes, 1.6v can (freak accident, but it has happened) fry a cpu if you got a crappy CPU......not all CPU's can handle the same vcore (just like not all CPU's can handle the same clock speeds, each is built "differently")

that is just incorrect, yes different cores built on different manufacturing processes do operate in a different voltage range, due to the current draw on the smaller/larger transistors and higher/lower gate lengths of the 90nm and 130nm architectures, but the smallest manufacturing process at the moment is 90nm, and 1.6v will NOT FRY THE CPU !!!

Do you understand me

 

[BigCoolJesus]

Originally posted by: RichUK

Originally posted by: BigCoolJesus

Originally posted by: RichUK
ok originally i quoted you because what you had stated did not make sense, and was quite incorrect.

for example:

Some people can set the voltage to 1.5 and have crazy ass temps of 55C or whatever, some can set it at 1.6 and fry thier cpu, and some can go even redicuously high without seeing any noticeable temp increase

i mean seriously WTF is that statement supposed to mean, i am not trying to ridicule you it just doesn?t make sense its just an highly inaccurate statement on many levels, let me tell you something 1.6v will not fry a CPU .. even if you take off the heatsink.

and this one aswel

all im trying to stress is the MAXIMUM vcore that a cpu (any core) will support is not a constant

your not getting it are you?

first off, yes, 1.6v can (freak accident, but it has happened) fry a cpu if you got a crappy CPU......not all CPU's can handle the same vcore (just like not all CPU's can handle the same clock speeds, each is built "differently")


I dont see what so confusing about my last statement, that vcore is not a constant among CPU's..........because it isnt (you know what a constant is, right? ie, vcore isnt the same for every single cpu)
There are some people on here with overclocks of 2.6GHz on a 1.58vcore and they have temps of around 50-52C.......and then there are some who run 1.64v and have temps of 45-48C
IT ALL DEPENDS ON THE LUCK OF THE DRAW

Ok .. you are talking about different CPU's, as in different cores and different tolerances right?

if so you should state that in your text, as you are not very clear on some points, that you are infact talking about differrent cores. Ok so you are talking about different cores thats a given ..

But once again this statement:

first off, yes, 1.6v can (freak accident, but it has happened) fry a cpu if you got a crappy CPU......not all CPU's can handle the same vcore (just like not all CPU's can handle the same clock speeds, each is built "differently")

that is just incorrect, yes different cores built on different manufacturing processes do operate in a different voltage range, due to the current draw on the smaller/larger transistors and higher/lower gate lengths of the 90nm and 130nm architectures, but the smallest manufacturing process at the moment is 90nm, and 1.6v will NOT FRY THE CPU !!!

Do you understand me

no, the whole time through i have been talking about the SAM CORES


and no, i dont understand you because i have had at least 2 people tell me about how thier 3200+ Venice's were killed when they tried going to 1.62V (they set it at 1.62v, restarted, and as soon as computer came back on, vrap, it turned off, and the cpu was dead)


You make it seem as if every single CPU (same core were still talking about) is exactly the same.......thier not, every now and then youll find ones that can do a hell of a lot better then others, and ones that cant do crap and will die when forced to

 

[RichUK]

Originally posted by: BigCoolJesus
no, the whole time through i have been talking about the SAM CORES


and no, i dont understand you because i have had at least 2 people tell me about how thier 3200+ Venice's were killed when they tried going to 1.62V (they set it at 1.62v, restarted, and as soon as computer came back on, vrap, it turned off, and the cpu was dead)

If the whole time you have been talking about the same cores then you are even more confused then i thought !!

for example:

Some people can set the voltage to 1.5 and have crazy ass temps of 55C or whatever, some can set it at 1.6 and fry thier cpu, and some can go even redicuously high without seeing any noticeable temp increase

Then my previous statement still stands, that THIS (ABOVE) statement by you is still ill educated and highly incorrect! Going by your theory if i can try and understand it (as we are still talking about the same cores here wink wink:

UNO Some people can set the voltage to 1.5 and have crazy ass temps of 55C or whatever :- They might have genuine problems e.g. incorrectly seated HSF.. what ever, or they just installed it upside down

DEUX some can set it at 1.6 and fry thier cpu:- People just leave it outside to operate in the rain as a water-cooling substitute

TRES and some can go even redicuously high without seeing any noticeable temp increase :- some people just stick their whole comp on a big ice cube to eradicate the heat output

You make it seem as if every single CPU (same core were still talking about) is exactly the same.......thier not, every now and then youll find ones that can do a hell of a lot better then others, and ones that cant do crap and will die when forced to

And OMG as your above statement is still with regards to voltage and not frequency, then yes you are still wrong, every CPU with the same core (I.E San Diego), will react the same with regards to voltage given the same circumstances. Only the CPU's frequency is variable to voltage (for example the operating frequency of the 3700+ and FX57, same core same voltage different stock frequencies, due to different die yeilds). Only on the extreme rare occasion does the CPU fail and require an RMA (this is usually very unheard off, very rare)

i cant be bothered to explain any further, maybe this thread might help enlighten you a bit more http://forums.anandtech.com/messageview...atid=28&threadid=1690294&enterthread=y

 

[BigCoolJesus]

haha, im not confused


and no, not every CPU will react the same (same core still)..... yes, 99.9% of them will, but there are those few (which i have been talking about) that will either do better then the majority, or worse

and within those few, there were 2 instances were at 1.62V t he cpu was fried electrically


im not gonna argue it anymore though, as you seem to believe everything is set in stone, that there are no "loopholes" (which there are)

 

[bupkus]

I may have solved my overclock problems. I guessed that either the problem was the ambient temp or the northbridge may be too warm.
I tried to glue a 50x5mm fan to the nbhs but it didn't take. While looking for some drywall screws I recalled some 60x25mm Y.S. TECH fans I had. I screwed one down and I'm sure it's cooler than without. I boosted the vcore from 1.650 to 1.675 and then to 1.700 and there it stays and seeming to be stable(for now). If it stays good I will be very pleased as it will serve as a good gaming platform for a Radeon 9800 Pro my nephew has. Knowing that I still have plenty of room to spare if I need to bump up the core voltage is reassuring.
Thanks to all even if you guys can't see that you're just having a simple "[Captain, Road Prison 36: What we've got here is]... failure to communicate."
BTW, I'm using a Thermalright SLK-900U heatsink with a 92mm fan on the cpu.

 

[bupkus]

Which voltage should I believe? My bios supposedly set it to 1.700v, the bios reports 1.72v, and cpu-z v1.30 calles it at 1.600v!
Which should I believe?

 

[BigCoolJesus]

Originally posted by: bupkus
Which voltage should I believe? My bios supposedly set it to 1.700v, the bios reports 1.72v, and cpu-z v1.30 calles it at 1.600v!
Which should I believe?

The BIOS, always believe the BIOS (its the closest thing you can get)

 

[RichUK]

Originally posted by: bupkus
Which voltage should I believe? My bios supposedly set it to 1.700v, the bios reports 1.72v, and cpu-z v1.30 calles it at 1.600v!
Which should I believe?

Go by CPU-z it is more accurate, as some motherboards under volt .. like mine .. if i set it to 1.55v in the BIOS which is max .. it is actually producing about 1.42v. Only when I use the % over VID can I produce higher voltage for the CPU. My board also over volts on the RAM, when i set it to 2.85v (which is max) it will actually produce 2.9v.

You could also try other applications that will also report the voltage to the processor as well, just to give you a bit more of an idea of what the processor is receiving voltage wise.

 

[BigCoolJesus]

Originally posted by: RichUK

Originally posted by: bupkus
Which voltage should I believe? My bios supposedly set it to 1.700v, the bios reports 1.72v, and cpu-z v1.30 calles it at 1.600v!
Which should I believe?

Go by CPU-z it is more accurate, as some motherboards under volt .. like mine .. if i set it to 1.55v in the BIOS which is max .. it is actually producing about 1.42v. Only when I use the % over VID can I produce higher voltage for the CPU. My board also over volts on the RAM, when i set it to 2.85v (which is max) it will actually produce 2.9v.

You could also try other applications that will also report the voltage to the processor as well, just to give you a bit more of an idea of what the processor is receiving voltage wise.

the only quips i have with relying on applications, is a lot of the time thier off (CPU-z is on par with my BIOS readings for vcore, but on my last rig, it was ridicuously high (2.7 or something way way off))

but yes, CPU-z is trustworthy most of the time, but so are the BIOS




for some reason everest reports my 12v line as 10.72 and my vcore as 3.93