No sense in arguing with someone with a closed mind.
I'm coming from an overclocking background where perf/W is the biggest factor ; smartphones.
Currently I'm using a Galaxy S III.At stock,it ran at 1.4GHz with 1.275/1.3mV,depending on the binning of the chip.Mine was 1.3mV,and is now running at 1.6GHz with 1.225mV.The result?A nice bump in performance with actually LESS wattage drawn than at stock.I can easily do 1.7GHz but it exceeds stock voltage and destroys battery life.
An even better example was the Galaxy S 2 with its initial overheating problems.It ran at 1.2GHz with a rather high voltage (I think 1.3mV again).And believe me when I say that at times it would almost burn my finger when I touched the area around the CPU.After spending weeks tinkering with it,I had finely tuned voltages for each frequency and modified frequency scaling.The result was a phone running at 1.5/1.6GHz (I always tried to stay at the more sane 1.5GHz but again went back to 1.6GHz after a few days) with no overheating problems.
My point?If you really take the time to find the sweet spot for your particular CPU,you can make it work wonders.Having the optimal voltage for each frequency will help clock higher and consume less energy,which means lower temperatures too.Don't stick to the "+1GHz with +0.5V" methodology.Do it right or don't do it at all.
How can you compare overclocking on a cell phone to doing the same thing on a full desktop computer??
Your coming from a cell phone overclocking background while most of the people on site have been overclocking desktop processes since the Celeron 300A.
Who cares?People could have been overclocking freezers for all I care.The basic principles I was referring to apply to both PCs and phones.Power usage goes up linearly with clockspeed and exponentially with the voltage.Increase clockspeed,decrease the voltage and do the math.Also,luck of the draw is always a factor and spending enough time to know the tolerances of a particular CPU always yields the best results.These could apply to GPUs as well.I can't see what your argument is.
I'm saying that having overclocked quite a few PCs,just not to extreme levels and without spending too much time.My most recent and extensive experiences,however,come from the mobile space.And I can assure you,things are quite similar.Just a lot simpler.
While i understand your point it does matter.
Overclocking a phone is not the same as a desktop processor and last I checked this is a cpu and overclocking forum not the mobile one.
For me, you're the guy with the closed mind. You're so bent on thinking that overclocking is "free" that you forget that power consumption represents money in electricity, heat that gets transferred to the ambient, Fan noise, extra stress on some components, and some of those things are measurable and important. At stock Voltage it may be minor, but still there is, and as I stated in my previous Post, you're skipping the part that the reason why you can overclock at stock Voltage is because you have more of it that you really need, as you can also undervolt at stock Frequency if you prefer to save on power consumption instead of getting more performance.Since you 2 continue to ignore my comments, and change the rules, I am done with this thread...................No sense in arguing with someone with a closed mind.
BTW, I also noticed that you edited a previous Post explaining what the PPD bonus is, which wasn't there when I saw it. Also that you edited some of my Posts when I quoted you and your statement that I must return under a rock I crawled from or something like that. I'm happy you noticed how offensive your comment was, I'l take that as an apology.
When all else fails, argue semantics! :whiste:
For me, you're the guy with the closed mind. You're so bent on thinking that overclocking is "free" that you forget that power consumption represents money in electricity, heat that gets transferred to the ambient, Fan noise, extra stress on some components, and some of those things are measurable and important. At stock Voltage it may be minor, but still there is, and as I stated in my previous Post, you're skipping the part that the reason why you can overclock at stock Voltage is because you have more of it that you really need, as you can also undervolt at stock Frequency if you prefer to save on power consumption instead of getting more performance.
You're also forgetting that in order to overclock you have to spend more, because during the last three generations Intel charges you for doing so by forcing you to get specific Processors and Chipsets at a premium for that feature. Plus the aftermarket heatsink, as for Haswell overclocking the retail one is totally undersized. In comparison, I recall Core 2 Duo overclocked very well with stock heatsinks, so you didn't even had to spend on one and still get decent overclock margins. It is not free, neither to get the parts, nor the actual results.
In *YOUR* use case, you say that overclocking is justified because you have a big timed bonus goal. Fine. However, for MY use case, when I upgraded from my 2006 Athlon 64 3000+ Venice to a 2010 Athlon II X4 620, I had soo much CPU performance that during an entire year that I spend playing with emulators and old games I was happy running 1 GHz @ 0.8V with the Fan at minimum, making it a nearly silent machine even on Full Load. And as the machine was running 24/7 anyways, I didn't really care about system power consumption (Which should have been rather low) as a whole but just the Processor.
With an unbloated WXP SP3 installation (14 Processes at boot), I didn't notice any meaningful difference in daily usage between 1 GHz and 2.6 GHz. Actually, on some newer games even 2 GHz was overkill as I was bottlenecked by the integrated Radeon 4200 (AMD 785G), so it was pointless to run it at its fullest, yet most of those old games love to make a Core get to Full Load even when its noticeable than that level of performance is unneeded.
Its the user the one that knows what he values the most. How much performance you actually need? Do you need 4.2 GHz? Do you need 3.4? Can you live with 2 or 1 as I did (On a K10, a Haswell would do better)? Do you want to have an airship turbine as a Fan to get to 5 GHz, or would sacrifice performance in exchange of silence? If benchmarks didn't tell you a performance number, would you notice it?
The point is that from the Core 2 Duo onwards, there seems to be consensus among people and reviews that CPU performance for Joe Average is getting good enough to the point where more isn't easily noticeable (Didn't Anand himself say so if I recall correctly?), and even we had a performance regression to 2002-2003 levels during the Netbook craze with those worthless Atoms, yet people somehow managed to adapt to those slugs. And today, I see people quite often building low end Celerons quite happy with the everyday performance that they offer. Yet at the other side of the fence there are guys that can never get enough.
So, I postulate three scenarios where you may want more performance than what suffices for Joe Average in his bloated Windows installation:
1 - Real time, like games, where performance is quite variable with ups and lows, and the target is to be over 30 or 60 FPS under a worst case CPU bound scenario
2 - Applications whose workload is continuous and infinite, like your folding thing (I could also say cryptocurrency mining, for that matter). This is a "all what you can eat"
3 - Applications whose workload is continuous but has start and finish, like a render, unzipping a file, etc. You may have a deadline to act as time limit, which I could say, my sleeping hours, and that decides how fast the CPU should have to work to make sure it finish by then. Indeed, if I need to wait for it to finish before I can do something else, I would want it to be as fast as possible
For every other scenario, I take undervolting @ stock to minimize noise, power consumption, and component stress, instead of overclocking. Your mileage may vary, this is mine. But this is also what I see that is the best option for any Joe Average that wants to tweak either for fun or efficiency, as it is overally less risky, and assuming that you feel you have enough performance, the best way, too.
BTW, I also noticed that you edited a previous Post explaining what the PPD bonus is, which wasn't there when I saw it. Also that you edited some of my Posts when I quoted you and your statement that I must return under a rock I crawled from or something like that. I'm happy you noticed how offensive your comment was, I'l take that as an apology.
I didn't really care about system power consumption (Which should have been rather low) as a whole but just the Processor.
Yep, my computer is always on 24/7 because I always leave all the instant message applications so if I'm awake I can hear if someone is talking to me, BitTorrent, my previous work/gaming session open at the exact same spot I left it in case I felt sleepy, etc. So I take system power consumption as granted, the only variable is the Processor, that as I explained earlier, as an standalone device is always more efficient at a lower Frequency. And even then, the computer at Idle should be competing with a 40 or 60W light bulb in power consumption, is relatively little.I dont even know how to respond to this, that's like saying you dont care how much MPG the car gets, only how much the leather seats take away from the MPG.
In one breath you care about power efficiency, the next you say the PC was on 24x7 which is horribly inefficient.
the only variable is the Processor, that as I explained earlier, as an standalone device is always more efficient at a lower Frequency. And even then, the computer at Idle should be competing with a 40 or 60W light bulb in power consumption, is relatively little.
Re-check Slomo4shO Posts, he is arguing the very same thing I do about undervolting @ stock.Point being, you're way on the wrong side of this topic. Youre first clue should have been when no one agreed with you.
1 - Not everyone has access to cheap used parts markets. I live in a place where the used market is around 70-80% of the price of the part when it was new. Purchasing something used here is usually a complete ripoff. And you don't see Server parts of recent generations often.Last I checked, a $75 chip on a $100 board was the very definition of budget. Of course if you read the damned thread you'd know that.
Re-check Slomo4shO Posts, he is arguing the very same thing I do about undervolting @ stock.
1 - Not everyone has access to cheap used parts markets. I live in a place where the used market is around 70-80% of the price of the part when it was new. Purchasing something used here is usually a complete ripoff. And you don't see Server parts of recent generations often.
2 - Yes, eBay exist, but I need to pay 30-40 U$D per Processor shipping or 70-100 U$D for a Motherboard shipping cost, custom tax (50%) and expect that the seller wants to do international shipping. This put it in the range of a brand new parts adquired locally, and a lower amount of possible suppliers.
Quote from this Post:You realize that the undervolting@stock can be interpreted as free money, instead of a free overclock ? Using less electricity and producing less heat than Intel (Or AMD) indended@stock.
So, is it free money or free overclock you are arguing for ? make up your mind.
You never mentioned undervolting @ stock as viable until now. I say that for anyone that knows how to do overclocking @ stock, they can't compare it to stock because I'm assuming they know how to undervolt, too.You can take two paths: Highest possible overclock at stock Voltage (More performance for slighty higher power consumption, as the default Voltage is already higher than needed), or lowest possible Voltage at stock Frequency. Comparing overclocked vs stock is unfair, I said that earlier.
By that Thread results, things are like this:
Stock: 3.4 GHz @ 1.164V = 84W
Lowest Voltage: 3.4 GHz @ 1.038V = 65W
Highest Frequency: 4.2 GHz @ 1.163V = 99W
It is NOT FREE. Is simply that you're looking at it unidimensionally. The actual jump is 34W, a whole 50%. And considering that its pretty much the sweet spot in power efficiency if you take into account system power consumption, I would prefer to run at 3.4 GHz @ 1.038V and put the Fan at minimum to reduce noise.
Wouldn't it be safe to assume that debating over pennies in electricity in a CPU/Overclocking forum is silly? This is an enthusiast board, not Greenpeace.
Besides, you can't ignore the fact you can still OC and leave Speedstep and other power-saving features enabled, so that you are not clocked higher when not using your machine.....
There is nothing wrong with wanting an efficient machine, but you shouldn't be buying i7 or 4+ GHZ AMD chips if that is the case.
Wouldn't it be safe to assume that debating over pennies in electricity in a CPU/Overclocking forum is silly? This is an enthusiast board, not Greenpeace.
.
Wouldn't it be safe to assume that debating over pennies in electricity in a CPU/Overclocking forum is silly? This is an enthusiast board, not Greenpeace.
Besides, you can't ignore the fact you can still OC and leave Speedstep and other power-saving features enabled, so that you are not clocked higher when not using your machine.....
There is nothing wrong with wanting an efficient machine, but you shouldn't be buying i7 or 4+ GHZ AMD chips if that is the case.
The biggest cost when overclocking is going to be the additional money you spend on cooling. As stated, the extra costs of electricity, especially if you keep all the power savings features enabled is pretty small unless you're PC is under load 24/7, in which case it could be a reasonable chunk of change by the end of the computers life. Particularly with AMD processors as they consume a reasonable amount of power when OC'd
For me personally, overclocking with stock volts makes me uncomfortable. Whenever I OC, I find the maximum overclock I can get while keeping the temps reasonable, then I drop that OC by 100MHz. At the same time, I find the voltage that passes all the tests I throw at it, and raise it by a couple notches.
I like to have that stability cushion which wouldn't exist by OCing under default volts, and if it did exist, it means a pretty mild OC that probably isn't even worth the trouble.