Is lower the CPU voltage harmful the CPU

According to basic theories, lower voltage will cause higher currents when the POWER is the same. Can that kind of theories apply to CPU?

 

Well, the idea behind undervolting your processor is to reduce power consumption. Thus, the current remains the same while the voltage is lowered.

If anything, undervolting will likely increase CPU life expectancy, but they last for so long anyways that it's not a real benefit.

 

No.
Check here:
http://www.thegamebooks.com/undervolting-intel-pentium-m---how-to-keep-your-gaming-cool-t78.html

Cheers,

 

Don't do it. If anything, increasing your v-core would be your only option for overclocking. We are talking about laptops here, leave it alone. If you want to experiment with under volting or overclocking, build a desktop. As thinkpad owners I would think it's in our best interest to get the best battery life and best performance. Therefore, I stick with what Intel already provides in my Pentium M 1.7ghz.

I dream of the day AMD makes a Turion processor that is equal to or better than what Intel offers. Down with the evil empire of Intel, but that is neither here nor there...

 

What are you on about? Nobody is talking about overclocking or increasing the vcore. The question was about undervolting, and despite what you seem to think, it makes a lot of sense to do to a laptop processor - much more so than most desktop processors.

As I've said already, lowering the voltage reduces power consumption, which increases battery life and lowers heat production, all with no performance or stability loss if done correctly. It's win-win; there's no reason not to undervolt.

 

Under volting your V-Core will always lead to instability issues. Instability does not mean your laptop will crash. This could happen. By undervolting your laptop may lag by not perfoming tasks as quickly. Just because you can under volt your laptop doesn't mean that this is a good thing.

My advice is to leave the V-core alone. If you want more battery life by an extra 8 cell battery.

Note, I'm not a special olympian or one of Jerry's kids. Stability is very important. If you listen to the previous poster, why not reduce the voltage on your ram as well, lol...

 

...what?

I don't even know where to begin. It boggles the mind...

Core voltage does not directly affect performance! As long as it is high enough for the processor to avoid making errors when executing the code (which results in instability), an undervolted processor will run exactly the same as one at stock voltages. My processor running at 800MHz on 0.716V will perform exactly like yours does at 800MHz on 0.988V. That's a fact.

Stability is important to me too. That's why my voltage setting for each multiplier is one increment above the lowest stable voltage, just in case. And guess what? I routinely have a 250+ hour uptime on my machine, with no crashes.

 

Hi guys,

I must say that I agree with Trippytiger. There are problems and issues with undervolting, and I described them in above article link. If you read that article, maybe you change your mind Karl Gruber . No one says you must do it. It is a good thing to do if you want and know how to do it.

Personally I think that CPU manufacturers just set a bit higher voltage to be extra sure that voltage isn't the issue when problems happen.

Even a small voltage decrease greatly affects the temperature. And temp is killer number one for any electronic device inside the notebook.

It is really a good thing. At least that is my opinion, and I didn't see any negative side effects.

Cheers,

 

Because Trippytiger knows what he/she is talking about. I must abstain from commenting on the other poster.

 

Just because you can do it doesn't mean it right. I'm not saying that it won't work. I'm saying that instability is more likely going to happen when you significantly undervolt your CPU. Maybe 1/10 of a volt won't make a difference, but cutting half of volt will most surely result in errors eventually.

Let's consider the T42 with DDR memory vs. T43 using DDR2 memory. DDR memory has 2.5v or 2.6v compared to the minimal 1.8v on DDR2 memory. This is a huge difference in power consumption yet T42 owners claim superior battery life. If you have 2 sticks of Ram, that's almost 2v of energy.

I do realize my intelligence is far superior to the many egocentric posters on this forum. I pity the special olympians who post here posing and knowledgable.

Why not undervolt the ram as well? Instability will creep up on you sooner or later. Get a larger capacity battery or a second battery.

 

Please, don't comment on undervolting until you actually understand to do it. .5 V is a huge change and will obviously result in instabilities.

Again, please understand undervolting before commenting. Undervolting isn't something like depriving the brain of oxygen, rather it is like teaching the body to be more efficient. Secondly, "I pity the special olympians who post here posing and knowledgeable." is grammatically incorrect.

 

Thank you for the pity. Unfortunately, your grammar teacher taught you to substitute the word "and" for the correct word "as." I could be incorrectly assuming you know the word "as," though.

 

Now you have to figure out where the .8v of extra energy consumed by DDR 2.6v memory in a T42 still manages excellent battery life compared to the 1.8v of energy consumed by the T43. With 2 sticks of ram, that's almost 2v alone. I don't think the Pentium M's run anything close to 2v.

Undervolting the Pentium M chip really doesn't make sense considering my argument. Stop being ego-centric. You may not be a special olympian but you certainly rode the short bus to school.

You forgot one important factor, I'm a golden god!

 

You know, I don't think I ever got an answer from you when I asked if you were for real. Are you actually serious, or are you just a regular troll?

Anyways, the lower voltage requirements of DDR2 RAM still don't go very far to offset the increased power draw of 533MHz FSB Pentium M processors and the Sonoma chipset.

And since you clearly don't know this, power draw is not affected by voltage alone. It is a function of both voltage and current; the formula is given by P=IV, where P is power (in Joules), I is current (in Amperes) and V is voltage (in Volts). This is what the OP was referencing in his post, of course.

Now, with this in mind, it's an obvious inference that - since we know that RAM has nowhere near the power draw or heat production of a modern processor - despite the higher voltages needed by the memory, the current draw is also much lower. From this, it's pretty clear that undervolting the processor makes much more sense than undervolting the RAM - although, if there were an option to do so, I would certainly consider it. The only reason undervolting the processor doesn't make sense by your argument is because your argument doesn't make sense.

FYI, the largest voltage drop that a Pentium M can take while still running stable at a given multiplier is about .300V or less, not 0.500V like you suggested.

 

Let's try this without being snappy. My understanding - and I don't claim to have a lot of it on the topic - is that, if done properly, undervolting can increase lifespan and decrease temperatures, both of which are desirable.

By the way ... yo mama! ;-)

Best,
Howie

 

Haha, Nice.

Anyways, I didn't join this forum to debate. But the general consensus is that undervolting is better for the laptop as mentioned.

 

just as an example, there was a guy on here who was load-testing his CPU with Prime95 and had a single error after 12 hours with the 6x multiplier at 0.700v, so he bumped the voltage up a notch and went for 18 hours without an error and stopped testing. I dunno, but for my application, that seems plenty stable. I think I've left my laptop on for 18 hours idling once, but never with a constant full CPU utilization.

Personally, my ASUS z63a 1.73 Pentium M runs about 6-7 degrees celsius cooler. With the brightness on my monitor turned down and Wi-Fi off before and after the undervolting, I went from about 3.5-4 hours to between 4.5 and 5 hours run-time on battery while taking notes in class - the variation mainly coming from the harddrive.

 

wow, I didn't expect drunk posts on this forum

 

I was not implying that you should undervolt the ram. I was kidding. I also didn't say you couldn't undervolt the CPU. I simply said you would be opening yourself up to instability.

With regards to FSB being a drain. Totally untrue. The FSB is dictated by the ram. Because DDR2 runs at 533mhz and a lower voltage 1.8v vs. 2.6v there is no increased power consumption with a higher FSB. MHZ measures speed not power consumption. Speed does not translate into increased power or increased voltage. There are other factors with the Sonoma chipset that cause heat increase. This has more to do with the design of the chip.

 

Ok, I will lay out the simple facts for everyone here.

* Undervolting may cause system crashes, therefore stability testing is required. If you don’t know what you are doing then it is best to increase your knowledge before proceeding.

* Undervolting will NOT cause damage to your hardware.

* Undervolting could corrupt your OS install if a crash occurs while writing a critical system file, requiring you to reinstall your OS. I don’t know the probability of this but I think it is likely VERY low. None the less, I would recommend backing up critical data before proceeding.

 

Good notes, TheBeck. I never said undervolting would damage your ThinkPad. My concern is only with the instability and the irrationality of some posters. If you want more battery life buy a larger cell battery or an additional battery. The Pentium M chip is already an energy saving design. I think about 25w.

 

~21 for Banias and Dothan, 27 for Sonoma