Slashdot Mirror
The Plusses And Perils of Overclocking
mblase writes "This C|Net article, published this morning, covers some of the advantages and many of the drawbacks involved for those who want to seriously overclock their PC hardware -- and why."
← Back to Stories (view on slashdot.org)
in order to reduce noise & electrical consumption. For example, I have a baby-AT system with a Pentium 200MMX processor. It needs to run in a quiet environment.
... faster ain't always better.
So, I underclock the CPU to 100MHz, and have removed the CPU fan. The power-supply fan is disconnected, because the case interior simply does not get hot. There is no disk drive in the machine (it netboots) and the box is totally silent.
Just goes to show
Everytime there is a story about overclocking on Slashdot the naysayers flood the forums with comments about how terrible an idea it is to overclock. They say things like "You only save a little bit and spend more on cooling..." or "Your chip will be unstable and then have a shorter life."
I would like to clear some things up about overcloking for the uninformed people.
Firstly, the stability issue. Overclockers hate instability. Most of us go way out of our way to make sure that the system we are running is not going to be acting all flaky when we overclock. The whole point of overclocking is to get the most out of the system, and if your system is freezing all of the time you aren't getting much out of it are you? We accomplish this by running benchmarks and torture tests to make sure that the overclock isn't adversely affecting performance or stability, if it is then we step it off.
Second, with the exception of a few extreme instances, most overclockers save money for the same performance. We don't all go out and buy peltiers and liquid cooled heatsinks. Most of us spend more on cooling than the average person but not by much, and our cooling system usually lasts through several cpus. Compare the $50 hsf I'm using now with your $10, so I spent $40 more than average, big deal, I saved $300 on the cpu and I'll use this cooler with my next upgrade too.
Which brings me to the savings. We save a lot of money for the performance. When I purchased my Celeron300A I spent $109 for it and after I overclocked it, the performance I got out of the chip in games at the time was almost identicle to a P2-450 which was selling for well over $600.
Now thats about as good as overclocking gets, but there are many other examples of chips since then that have done almost as well.
That celeron300a I spoke of is still running at the same overclocked speed as the day I put it in, and it's rock solid. You want stability, there you go.
On top of all of this, overclocking is fun! No really. It's an enjoyable experience, you learn a lot about hardware, and at the end of the day you can be happy that you have a screaming fast system for a fraction of the price you could have spent.
If you want to think about overclocking try checking out some of the sites around the net:
www.overclockers.com
www.hardocp.com
www.anandtech.com
www.tomshardware.com
Try it, you might like it.
Sigs are awesome huh?
Am I the only one here that thinks that a C|Net article on OCing has no bearing to most of the slashdot community? C|Net has a reputation for barely technical articles, glossing over subjects and missing out on all the important details. The real details are available (and have been for a long time) on sites like AnandTech, Tom's Hardware and others. Besides, most of the /. crowd already knows about overclocking and is not going to benefit from a story like this. This is not news for nerds. This is not stuff that matters. If I could moderate this story, it would definately get a -1, Redundant.
Using your sig line to advertise for friends is lame.
Almost every industry does this. You overengineer a device, so that flaws that occur during the manufacturing process don't break down the guarantee.
If I see a sign that says `2 ton limit' on a bridge, it better take 2 tons. In fact, I'd be unhappy if it had problems with 2.5 tons.
I don't understand why they're upset about over-clocking (and trying to make it more difficult). Yes, you can buy a 1 GHz processor and over clock it to 1.2 GHz (or whatever the number is...I don't OC, so don't use these numbers as a basis for OC planning). Alternatively, you could pay for that 1.3 GHz processor and OC it to 1.5 GHz. In my mind, the manufacturers should be out there challenging people to OC their processors, so that they learn the actual limits of their processors, find new ideas on cooling, encourage OC'ers to buy the top-of-the-line and make it run better.
The article missed one of the coolest elements of the overclocking phenomena. Overclockers today are like the hot rodders of the 50's and 60's. Ask them why they spend hours and dollars to crank out that extra 2-5 percent and they will look at you like you just don't get it.
There will always be a base of people who want to push their toys/hobbies to the limit. We've moved from cars to bikes to boats and now PC's.
Hobbyist underclocking is starting to take off. The MP3mobile, a home-built MP3 player for autos, involved a Pentium 166 underclocked to 125MHz. Automotive hardware is often underclocked; the thermal and power environment of the auto is fierce.
As for the gamers, they're probably better off getting a graphics card upgrade. The current generation of graphics boards are essentially overclocked already; current NVidia products have heat sinks on the RAM.
George W Bush had discovered that Tom Leufkens had struck a secret deal with NASA to ship his 500MHz Celeron to Pluto, where he believed the surface temperature would allow him to run the machine at 3.3GHz.
Bush, the current champion overclocker (his Hillary-Clinton-cooled P75 benchmarked 900MHz last Friday) foiled the world-record attempt just in time.
In a statement, Bush was reported as saying "Ain't no pissin' on the presedential PC."
Asikaa
Asikaa
Come in, twenty-seventy-seventy, your time is up.
I put it in a Frigidaire freezer side.
Go on...
Which I tried to bury in a snowdrift.
Which was because...?
The Frigidaire was smoking.
I see... and what did you hope to gain from all this
33 more MHz.
to...
1367MHz.
That is quite impressive for an 80286.
It's like an addiction, doctor, I just can't stop!
There, there. We'll break this addiction, it just takes time.
Thanks, Doc, do you think my girlfriend will ever come back?
Perhaps, but in the meantime, I have this old Pentium II, what do you think you could do with it?
--
A feeling of having made the same mistake before: Deja Foobar
When you build a new chip you build software models of the entire chip - right down to the gate and polygon level, you do a LOT of timing analysis - these days we extract the polygons and do 3d parasitic extraction - this takes a long time (days for a big chip) - but the results are by their nature statistical - because the results of building a particular chip are somewhat staistical (depends on etch rates, temp, etc etc at the fab) so we calculate the worst case fast and worst case slow process corners and use the timing tools to check all the potential timing paths (thing combinatorial explosion here). After we think we have something that will make timing we build some - at the fab at the begeinning we get the fab guys to explicitly vary the process to push some wafers into each 'corner' of the process - then we bring those die into the lab and use them to make sure that they will work at speed within the various temp ranges the chip is supposed to work at.
One of the problems with making chips is that testing them is VERY expensive - the testing machines that do die and chip sorting cost millions of $$ and the number of seconds a die spends on one effects the final cost - so you design your tests to uncover raw defects (via scan and maybe functional tests) and speed problems by using the results of your original timing imulations to identify the timing paths that are so close to the edge that they are likely to fail first - because the testers don't have access to most of the internal nodes you have to do things like overclocking by say 10% and then hoping the internal logic will fail in some manner that you can catch (you also use the previous lab work to validate this approach by identifying known bad chips and making sure they fail on the tester).
One thing you can do is 'bin' chips - test them at different frequencies and sell the ones that happen to be faster for more - because binning is a more expensive process its usually only done for CPUs and other expensive sorts of chips.
What commonly happens over the lifetime of a chip is that as the process improves the number of die that fall into the faster bins increases - however for marketting reasons a company may wish to continue to sell the 'faster' ones at a premium so it will label some fast chips as 'slower' so to keep their product mix in the market (a fancy way of saying 'so they can make more money'). I'm told the same thing happens with olive oil :-)
Now the chips are vey carefully screened and carefull spec sheets are written for them - you buy an 850MHz chip from AMD or Intel and it will work at 850 within the appropriate voltage/temp range specified on the data sheet (if not as, we've seen, Intel will recall chips that don't) - it's not in the chip manufacturer's interest to sell chips that don't work - they get soldered on to expensive boards and expensive system which have to be trashed at the OEM if they don't work - those data sheets make sure that to parts in a million those chips work as advertised.
Having said that - some chips do run faster if overclocked - you can always tell which ones - because you don't know which process corner the die was fabbed at - or what the binning policies were the day it was manufactured etc etc - even worse yet - and here's my traditional warning - WARNING - your overclocked CPU may work perfectly for months because what you're doing may not exercise the slowest timing path(s) in the design (remember combinatorial explosion!) - you might play quake for months on end without a problem .... then silently drop $1000 off your tax refund ....
If you buy a 1000 Mhz machine for a mission-critical system and simply cannot afford any downtime, you'll almost certainly want to use better cooling (the same cooling overclockers use) on the CPU, chipset, et al. You'll also want to underclock your system because by doing so, you'll reduce heat still further.
So you could still find overclocking techniques useful even if you do not overclock.
--
Oceania has always been at war with Eastasia.
Let's see what happens when I try this liquid oxyge...
134340: I am not a number. I am a free planet!
Pro:
Save $100 on a chip.
Con:
Spend $300 on cooling gear.
Trolls throughout history:
Jonathan Swift
Right now I have a 600MHZ athlon and it sounds like an airconditioner is running in my study. So if I got a 1.2 GHZ cpu and ran it at say 700-850 MHZ, I could theoretically take the fan out of my box (but keep the heatsink), use less electricity and it would last longer (besides being faster than my current cpu). Is there a downside to my logic that I am not seeing? My view of cpu's needing any/better cooling technology is to me an indication that cpu manufacturers ARE over-clocking thier cpus... just not as much as enthusiast do. Is there a reason why Intel/AMD/Cyrix etc... need better cooling besides the HZ war:
if (MyCPU.MHZ > Other.MHZ) {
BankAccount = (BankAccount + BetterSales)
}
I miss the Karma Whores.
...until you've seen a 386 boot at 400 Mhz. It won't run, but it'll boot.
God, I love the smell of silicon in the morning!
This