• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Overclocking and Vcore...... how high can i go

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Within reason I suppose you should keep your temps. One must think though. The transistors in the procs are getting smaller and smaller each time, so the wires (or whatever) are getting smaller too. Smalller wires cannot carry as much current with out damage. Certainly though, alot of people here are awfully concervitive with their vcore. I'm getting 54C loads with my winnie at 1.65vcore on stock cooling, so I figure I'm fine.
 
Originally posted by: KoolDrew
Here is even a link that shows what I am saying is true.
http://www.xtremesystems.org/forums/showthread.php?t=64881&page=2&pp=25

Read the second post. He even said "Oh temps were always good 'cos I have an excellent watercooling rig.." So again, Good temps do not make a high vcore any safer!
Click to expand...



Yes, but his died right away when he bumped up the Vcore, mines still alive and kicking

what im saying is if im running at 1.72 right now, with well under safe temps, whats wrong with it?



Im still confused, you keep saying temps dont matter, but then you said that 1.7 should only be attempted by phase change cooling?
WHY???

If im able to hit 1.7 and still keep the temp under the max safe, why is it any different then someone running 1.7 with phase change...... either way the CPU is still operating within safe temperatures
 
I agree with the voltage issue..too high and you will kill a cpu..but you make a strong statement on "safe vcores"...while I tend to agree I do think 1.65v is probably safe on air with good temps...note I am taking both into account

But my problem is the reading..bios, winxp differ..I do not have voltmeter...so I am at 1.57v in xp, 1.59v/1.6v in bios..so I foigure both are wrong and I will stay here..I know if I go to 1.65v in bios/1.616 winxp I can hit 2800...but my real vcore maybe higher???
 
Originally posted by: nealh

BTW there is a program yo ucan find at xtremesystems.com..that shows different newer A64 have a different max temp..some are 55-57-60C
all winnys report 65C..but the Venice chips are reporting primarily 55-57C....some as high as 63C
Click to expand...

Link to program by any chance?
 
Another vote for lowering the VCore.

The best option when overclocking, IMO, is to find a balance.

My 3000+ Venice runs 2.6GHZ 1.47v (1.49v BIOS reported) Prime stable for days upon days until I get that itch to play BF2.

If I try to bump it to 2.7GHZ I need a Vcore of 1.6v to get it Prime stable (1.55v is game stable).

The extra 100mhz to me is not worth the extreme bump in Vcore.

When you have to bump the Vcore so dramatically to get an extra 100mhz, that's telling me that you have taken the CPU outside of it's "comfort zone" and have entered the "danger zone".

What Vcore is required for you to hit 2.5 or 2.6GHZ? I bet you'll find either of those options will require much less voltage, lower your temps significantly and unless you're doing synthetic benchmarks, won't make a difference at all in games.

That is of course unless you're just trying to go for the MAX OC of your chip just for the sake of doing so.

EDIT: typo
 
Originally posted by: Matt2
Another vote for lowering the VCore.

The best option when overclocking, IMO, is to find a balance.

My 3000+ Venice runs 2.6GHZ 1.47v (1.49v BIOS reported) Prime stable for days upon days until I get that itch to play BF2.

If I try to bump it to 2.7GHZ I need a Vcore of 1.6v to get it Prime stable (1.55v is game stable).

The extra 100mhz to me is not worth the extreme bump in Vcore.

When you have to bump the Vcore so dramatically to get an extra 100mhz, that's telling me that you have taken the CPU outside of it's "comfort zone" and have entered the "danger zone".

What Vcore is required for you to hit 2.5 or 2.6GHZ? I bet you'll find either of those options will require much less voltage, lower your temps significantly and unless you're doing synthetic benchmarks, won't make a difference at all in games.

That is of course unless you're just trying to go for the MAX OC of your chip just for the sake of doing so.

EDIT: typo
Click to expand...



Actually, to do 2.6GHz my Vcore had to be 1.68 and
and 2.5GHz was 1.64........

and yes, I AM TRYING TO GET THE MAXIMUM I POSSIBLY CAN

just to clarify that, i bought the 3500+ over t he 3200+ so ensure i had a better chance of getting high..... and 2.7GHz is around where i was aiming at


Again, the only danger there is now that my CPU lifespan has been shortened significantly, which is fine as ill have upgraded by the time it dies (if it didnt fry when i first bumped it up, then i know it will last some time)


and just for proof that Vcore voltages over 1.6 are (sometimes) safe for venice's, ive been talking to someone with a 3500+ who runs it at 1.78 (1.82 when he feels like hitting 3.1GHz) and has been for the last 4 months....

just because 1.6 is the excepted maximum doesnt mean its not safe to go over in some instances.... some CPU's work better then others (hence why some people can hit 2.7GHz on a 3000+ and others cant get past 2.4GHz, it all depends on the luck of the draw)
 
Originally posted by: blckgrffn
That is why you have to keep drawing sometimes 😉
Click to expand...

😕

i........m

s.......l.............o.......................................w



Also, just wondering when my system will crash or my games will freeze, as a lot of people guarantee either of those will happen since i didnt pass Prime95 (aka, THE DECIDER ON OVERCLOCKS :roll: )?

Because after 6 straight days of running benchmarks after benchmarks, havnt shut off comp once, and just played BF2 for 4 hours, so when will it happen

dun dun duh......
 
Originally posted by: BigCoolJesus
Originally posted by: blckgrffn
That is why you have to keep drawing sometimes 😉
Click to expand...

😕

i........m

s.......l.............o.......................................w



Also, just wondering when my system will crash or my games will freeze, as a lot of people guarantee either of those will happen since i didnt pass Prime95 (aka, THE DECIDER ON OVERCLOCKS :roll: )?

Because after 6 straight days of running benchmarks after benchmarks, havnt shut off comp once, and just played BF2 for 4 hours, so when will it happen

dun dun duh......
Click to expand...

Prime crash doesnt mean you're going to crash, but Prime is a maximum stress test. If you crash Prime, that means that your CPU made an error and your hardware failed.

No matter how stable your system is in games, the fact that it errored out in P95 is never good. THat's not to say that Prime95 is the final say on if your system is stable or not, but if it errors out in Prime, the problem MAY rear its ugly head again down the road.

There's a reason why it's called a "torture test".
 
Originally posted by: BigCoolJesus
Originally posted by: blckgrffn
That is why you have to keep drawing sometimes 😉
Click to expand...

😕

i........m

s.......l.............o.......................................w



Also, just wondering when my system will crash or my games will freeze, as a lot of people guarantee either of those will happen since i didnt pass Prime95 (aka, THE DECIDER ON OVERCLOCKS :roll: )?

Because after 6 straight days of running benchmarks after benchmarks, havnt shut off comp once, and just played BF2 for 4 hours, so when will it happen

dun dun duh......
Click to expand...

I was trying to say that it may be time to sell your 3500+ and buy another similar processor and see if you have better luck. I have been through 2x 3000+ winnies, 1x3000+ venice and 1x3200+venice searching for a 2.6ghz+ chip @ less than ~1.65V.
 
you fvcking idiot !!!

AMD chip cannot stand high vcores like Intel !!!

So why not give it 3 volts and make yourself a keychain LOL !!!
 
Originally posted by: Dothan
you fvcking idiot !!!

AMD chip cannot stand high vcores like Intel !!!

So why not give it 3 volts and make yourself a keychain LOL !!!
Click to expand...

umm, ok 😕


and blckgrffn, i see what you mean

i will be upgrading my processor sometime soon (earliest would be around thanksgiving) but until then, ill just stick with whats working for me, regardless of the Vcore (it works, so whats wrong with it?)


The one thing that does seem common around here is that everything has to be set in stone, and if you do anything other then whats in the stone, its 100% bad and will end the world :roll:
 
Originally posted by: BigCoolJesus
Originally posted by: n7
Originally posted by: Kensai
1.7.. 1.72.. Close enough.. How do you have 47C load? :shocked:
Click to expand...


He likely doesn't.

Most likely his temp sensor is off by about 5-10C 😉
Click to expand...


again (responded in general hardware also) my temps are very accurate, i am not only going by Epox's readings, but by readings from an actual temp probe, which has never gone past 34C (its taped right on the heatink, out of the way of the fan, so air being rushed by it is not a variable)
Click to expand...

Simply taping a temp probe to the outside of a CPU's heatsink is the worst possible way to measure CPU temps and is in no way even close to accurate. You're only measuring some of the heat that is being dissipated and probably getting more of a reading of the internal case temp rather than the CPU's core.

If you insist on using an external temp probe and you want more accurate readings ( they still won't be as good as what the board is giving you) then there's a way to modify the heatsink. I've seen other overclockers drill a small hole through the side of the heatsink as close to the bottom of it as they can get. Just large enough to squeeze a temp probe through it so that it contacts the sides. Make the hole deep enough so that the end of the probe can reach the center of the heatsink and over over the CPU. Then the temps will be closer to being accurate.
 
Originally posted by: icepik
Originally posted by: BigCoolJesus
Originally posted by: n7
Originally posted by: Kensai
1.7.. 1.72.. Close enough.. How do you have 47C load? :shocked:
Click to expand...


He likely doesn't.

Most likely his temp sensor is off by about 5-10C 😉
Click to expand...


again (responded in general hardware also) my temps are very accurate, i am not only going by Epox's readings, but by readings from an actual temp probe, which has never gone past 34C (its taped right on the heatink, out of the way of the fan, so air being rushed by it is not a variable)
Click to expand...

Simply taping a temp probe to the outside of a CPU's heatsink is the worst possible way to measure CPU temps and is in no way even close to accurate. You're only measuring some of the heat that is being dissipated and probably getting more of a reading of the internal case temp rather than the CPU's core.

If you insist on using an external temp probe and you want more accurate readings ( they still won't be as good as what the board is giving you) then there's a way to modify the heatsink. I've seen other overclockers drill a small hole through the side of the heatsink as close to the bottom of it as they can get. Just large enough to squeeze a temp probe through it so that it contacts the sides. Make the hole deep enough so that the end of the probe can reach the center of the heatsink and over over the CPU. Then the temps will be closer to being accurate.
Click to expand...



Im not basing my temp findings on the external probe, im using the actual probe that Epox installed....... im only using the probe i installed for ambient temperatures around/close to the CPU
 
Originally posted by: icepikIf you insist on using an external temp probe and you want more accurate readings ( they still won't be as good as what the board is giving you) then there's a way to modify the heatsink. I've seen other overclockers drill a small hole through the side of the heatsink as close to the bottom of it as they can get. Just large enough to squeeze a temp probe through it so that it contacts the sides. Make the hole deep enough so that the end of the probe can reach the center of the heatsink and over over the CPU. Then the temps will be closer to being accurate.
Click to expand...

my gosh, that is a great idea! i'm going to do that next weekend when i go home from school... well, maybe... i suppose it'll be the matter of making myself take the heatsink and fan off again...
 
1.8 is outrageous..

1.7v is way to high for my liking..

My Venice @ 2.8 did it on 1.50-1.51 loaded and anymore typically had dimishiing returns..

I really don't want to bring my X2 over 1.5v..

I wouldn't even consider 1.7v
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top