Slashdot Mirror
Intel Puts the Lock on Overclocking
Patrick Schmid writes "Intel included an overclocking-prevention mechanism into the 915/925 chipsets. So far, only Asus and Gigabyte know how to override it. You can start from the beginning or jump to where we discuss the overclocking lockout."
I was not very impressed with the ratings of the new processor. While it has a little bit of improvement - you really need to get a completely revamped system (mother board, memory at the minimum - but you want a video card to be able to utilize the new speed). I believe even SATA hard drives had some problems.
Not to mention that the fastest P4 runs at 35 degrees centigrade, while this processor runs at 70 degrees. That is a major problem, imho, for a marginal increase.
Overall, this is not impressive technology. They rushed the material out. I referenced my information from Toms Hardware.
I mod down so you can mod up. Your welcome.
Remember when they locked the multiplier on their CPUs? The only reason they did this was to sell 'faster' chips. Celeron 300A anyone?
Overclocking isn't overrated, every so often.
Why? Every so often, Intel releases a clocked-down chip from a better batch or ships new tech that's nowhere near pushing its limits.
My two favorite examples:
#1. Celeron 300A - This was the first batch of Celeron chips with L2 cache, clocked at the chip speed. The 300A was 66 x 4.5, Intel had just released the BX chipset with 100 MHz FSB. 300A's ran at 100 x 4.5, often with no stability concerns.
#2. Pentium III 700 Coppermine Core - This was a relabeled 933 MHz chip. Literally, they took chips from P3-933 batches and sold them as 700's. The multipliers were the same (7 x 100 vs 7 x 133). Another chip that ran a lot faster than it was labelled.
I had PC's built on tight budgets that used these chips. I was happy with both, and both are still stable to this day. YMMV.
With such a fragile socket and obscene thermal properties, who here is going to overclock a prescott P4 to begin with? 115W rated thermal dissipation on the 560 (3.6GHz) model. 115W!
------- "From bored to fanboy in 3.8 asian girls" ----------
Several years ago some white box OEMs were selling overclocked systems as though they contianed the rated chip. I don't know how common it was, but that was the offical reason that Intel clamped down hard on the practice. When it was just geeks in their houses saving a few bucks it was a minor loss (probably good advertising--Intel generally has the better manufacturing process and most overclocking headroom), once frad was diluting their brand and really reducing revenue they stepped in and put a stop to the practice.
Degaussing scares the bad magnetism out of the monitor and fills it with good karma.
Within the 915/925 chipset is a function that halts the function of the cpu should the cpu-chipset bus speed deviate outside of the acceptable tolerance (5%) of the rated bus speed. This affects both overclocking and underclocking the bus. The "fix" mainboard makers are considering is not much more than a small hack that allows for another 10-15% deviation from the nominal bus speed.
Analysis: this is nothing new. Intel retail mainboards have *never* allowed overclocking, and their processors have been multiplier-locked since 1998.bottom line: the lock is there for stability concerns. If you want to overclock a socket-T processor, use a mainboard with a different chipset.I'm sure Ali, SiS, VIA, and Ati would love your business.
------- "From bored to fanboy in 3.8 asian girls" ----------
If anything, the overclocker group is *shrinking*, not growing. Back when there was a huge difference with OCing, it had a following. Nowadays it isn't nearly as popular or as common a practice not necessarily out of technological restriction so much as a lack of need for an insignificant gain in speed.
-Jem
Try not to sound like Intel beat ur dog and poured sugar down your gas tank next time.
This lockout is built into their new chipset, not the processor. Last I looked there were motherboards out there using non-Intel chipsets to run their processors.
No sig for you!!
I really don't care if they stop me from overclocking because I don't do it. I don't do intensive CPU tasks, so I have no need for it.
/. community.
But, what I do is UNDERclock my CPU so that it runs cooler, thus needing fewer/quieter fans. I don't need all 2.4ghz, so I send the FSB and voltage down a bit. And then, instead of hitting 45 celsius when idle and fans blazing, I get around 40 degrees, and can't hear the fan at all. If I need to do something intensive, I just reboot, change to default and flip the switch that turns the fan on "high" mode.
If they are locking the FSB, voltages, multipliers, and everything else, this doesn't just prohibit overclocking, it stops tweaking at all. Which, in some form or another, is a fundamental need for most of the
Intel doesn't care for ordinary Joe to liquid cool it's P4 and overclock it to 4+GHZ. They care about motherboard makers making a "high end" product out of a "low/mid range" product. When some motherboard makers transformed the 865 (if I recall correctly) into a much expensive 875 (IIRC) chipset motherboard by simply flipping a switch, it probably got Intel pissed off. Almost same thing can be applied to Radeon X800 Pro and XT Platinum (or whatever the highest end it's called), with a price difference of nearly 100$.
IMHO, in this case, Intel is to blame. If the difference between a low/mid range and high end product is just a switch, or a simple transistor, they shouldn't ask extra money for it, because the features in the low end are only "hidden".
They don't have to do any extra work to allow overclocking; it's a natural artifact of how the chipmaking process works. All chips, at the start, are capable of being top-speed. After manufacture, they're tested and sorted as to what their actual maximum stable speed is, individually. Because there's no actual die-level differences on these chips, external factors are what tell it how fast to run. You feed it clock and a multiplier number, and it tries to run at that speed, whether or not it actually can.
Basically, it takes no extra work to allow overclocking, but significant extra work to prevent it.
---
Mod me down, you fucking twits. Go ahead. I dare you.
(I read with sigs off.)
Well IANAL either, but considering I think the language in the DMCA specifically says *copy* protection, I think an attorney would have a hard time arguing that it should apply here.
"The problem with internet quotations is that many are not genuine" -Abraham Lincoln
Your heat problems are almost certainly related to the S2K bus disconnect problem. Either run fvcool (if your chipset is supported) or buy a newer AMD-certified motherboard that says it comes with S2K bus disconnect enabled.
You will see a massive decrease in heat, and a more than 50% drop in electricity consumed by your CPU. Read my most recent journal entry for more information. In addition, you should probably invest a few dollars in thermal paste, a decent heatsink and 80mm fan. My 2GHz AMD processor doesn't even reach 130 F degrees, despite 90 F degree ambient tempuratures, a demand for silent fans, and a motherboard that isn't supported with FVCOOL.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Default speed: 1.86 ghz
Current speed: 2.5 ghz
It's completely stable with a quiet, cheap air cooler.
In fact, the pentium 4 northwood/athlon xp thoroughbred and barton generation is (was? not quite yet) one of the best overclocking opportunities around. P4 2.4C running at 3.2ghz and above, athlon 2500+ going to 3200+ without so much as a second thought, the overclockability of this generation is excellent.
They can put in what ever they want..
Might piss off a small part of their customers, but in the long run, the average consumer will just buy the faster model outright.
Us in the 'minority' don't count. Never have, never will. Its just how the world works.
---- Booth was a patriot ----