Video Card Overclocking Software
Overclocking your computer
There are two schools of thought as to why it might, or would like to overclock most CPUs and GPUs. One of them takes the peace, love and understanding of route, ie the manufacturing process is never 100 percent reliable, so every card that falls from the same production line born equal. Those with the better-brighter and brighter eyes (bred Pedigree, presumably) are ready for high-end components, but with a bit of a strabismus and a runny nose can have a fun twist if too much strain.
Therefore, some chips are hit with a minor official Clockspeed and cereals sold for less than their brothers more robust. The potential for its intended glory remains, however. Overclocking techniques can unlock at least part of that potential, albeit with the risk of frying the chip completely.
The tinfoil hat / Angry Internet Men theory is based on the same concept, but runs into a bit of paranoia. In this scenario, all the same processor in the series are created equal, but the man artificially neutralizes most of them and slapping logos differ on what are essentially the same chips. Overclocking, then, is simply a way to recover what is rightfully theirs.
The truth probably lies somewhere between the two. Mass production certainly has more financial sense than dozens of separate lines, and certainly one of low-end CPU or GPU can punch well above its weight, but its stability is not guaranteed as a chip that is officially capable of running at a higher speed. No manufacturer wants to deal with a steady trickle of returned parts, after all. But it does mean overclocking house almost always productive – and seemingly more with each new hardware generation.
It is also increasingly easy. The first of overclocking in the 4 to 10MHz CPU 8088 based on 1983 involved desoldering a watch glass base plate and its replacement by a third one, with results only partially successful. Ouch. However, the precedent was set: a dedicated man at home could overcome its official specification chip. IBM, then much the dog on the land of PC, was not entirely happy with this, what the tracking hardware including blocks wired overclock.
Over this time welding a BIOS chip, was able to overcome this difficulty. Strangulation IBM 1986 was broken, resulting in a series of "clone" systems – and a wealth of choice. Intel 286 and 386 processors became standard chips de facto, and the bus speed and voltage controls began to change physical switches and bridges to the BIOS settings and configuration.
This was the 486 that really changed everything though. This means that this chip was more common during the era that gave birth to the first-person shooter as we know it: in 1993 Doom very popular PC games performance for system upgrades driving the same way a half-life of 2 or Crysis has in recent days. At the same time, the 486 introduced two concepts absolutely crucial for overclocking both then and now. First, divide became popular product lines, was no longer a matter simply to buy a processor, but the processor. The 486SX and DX offers some serious performance differentials, particularly the SXS is limping / DXS not, leading to the current practice to assign names to different speeds and they were the same chip.
For a while, too, the 25MHz SXes could be overclocked to 33MHz by adjusting a jumper on the motherboard, retailers took advantage of something less healthy. Second, introduced the multiplier means conducting more clocks for each meeting by the bus frontal system. The 2x multiplier 486 is doubled as the bus frequency. This was something overclockers would make the best of successive generations of processors – is met with the multiplier was the simplest and often most effective to increase the speed of the CPU. Today (since the Pentium II, actually), the multiplier is locked to prevent than this, except for high-end chips such as Intel Extreme Edition series. For a time, had complicated forms of defeating the multiplier lock: the solder a PCB for third earlier chips, accessories and the infamous practice of drawing a line in certain AMD CPUs with a pencil. No other manufacturer of the CPU is likely to make that mistake again.
Around this, the weather became a place of RAM overclocking more common, as were endorsed memory speeds, and with that came more tuning of the front-side bus to compensate for multiple blocks. Overclocking BIOS shifted farther away from the bridges, which in turn led to overclocking software.
The first was in 1998 SoftFSB, allowing adjust bus into Windows first. With the era Pentium III wine coolers aftermarket, such as processors so hot now cast a cooling block and fan standard was not sufficient to against an overclocked chip. And so it went, overclocking largely becoming easier and more common place with each generation of processors. This leads the Core 2 chips today, and current Intel terribly undisputed market dominance of the CPU. In general, drawing at least half of the power of Pentium 4 that preceded them, most of the range offers a large amount of free space from overclocking, to the extent that a low-end Core 2 Duo you can almost go toe to toe with the top of the line.
So how is it done? The key to overclocking the processor front side bus (FSB). In simpler terms, this is the connection between the CPU and the rest of the PC, and its speed defines the speed of the processor to a significant extent. Intel CPU speed is the time final FSB multiplier – so if you have a 266MHz FSB and a multiplier of 9, the chip will run at about 2.4. While the multiplier is normally closed – though some chips allow at least lower than, to conserve energy and reduce heat – the FSB is not. Bump the FSB and bump up the chip. In our example, take the bus to 290MHz gives a 2.6GHz processor. This is not a random example, by the way, is what we run the Intel Core 2 Quad Q6600 in one of our test systems in the office, giving a healthy impulse to 200MHz makes a noticeable difference in the CPU intesive games and hi-def video re-encoding.
What prevents us from going higher? Not much, in the case of this chip in particular. We're playing it safe for the desk job, because we are in an office all sweaty. When we are playing with high-end tasks, we can operate stably more than 33GHz (with an FSB of 370 or so) in a decentish, the third-party cooler air. That's more or less trade blows with the best Intel has to offered $ 200 chip. But while going to 280MHz on the FSB took a pinch of BIOS, restart Microsoft BOB and his uncle, goes much higher means more complications.
First up, when our Q6600 is 33GHz, which is also near 70 ° C when under full load (and about 50 ° C when idle). It is perfectly stable, but it could damage the long run, and over which the fan is making enough noise to wake the deaf pensioner in street next door. Digital cameras, an elegant cold air or even just a speck of dust cleaning will bring the heat down, but can not reach a point where this becomes more expensive and hassle than simply buying a better processor.
The second obstacle is the motherboard. Raising the FSB did not
only affect the CPU, but also to the motherboard and, in many cases, RAM
and PCI-e slot to boot. In our case, we are using a motherboard that supports FSB monstrously high. When shopping for a motherboard, its highest FSB generally refers to four times the actual speed because of how the processor actually gets the data. Thus that when we've got the FSB to 266MHz, in effect, which is 1,066 MHz. When it's up to 372MHz, we need a motherboard that is happy at about 1500 Mhz. That is simply not a given, especially on cheaper boards, so shop carefully. And that, if you have a motherboard with a BIOS stingy, you may not be able to alter the times of the RAM and PCI regardless of the FSB, which can lead to the fall. Ours does, and our strong wall gigahertz overclock Q6600, we have to under clock speed RAM a little to offset the pressure they put on it the FSB – have to be very 893MHz. Conveniently could increase, but the real benefits (as opposed to the Willy-waving benefits, which are an entirely different matter) would be so tiny it's not worth putting extra pressure on the RAM.
Just Thus, the faster and probably more expensive RAM will better cope with the speed of their actions with a massive FSB, the payoff is often so minor that the value of RAM, running at a lower clock speed may be sufficient for overclocking its master plan a huge success. Even the best memory you report something in the region only five percent performance increase – if it is worth keeping every little helps, but it is the FSB that makes the difference. And for that, the motherboard is crucial.
Thirdly, there is the issue of stress. The faster the chip runs, the more power you need for feeding. As the FSB goes up, you'll find on the motherboard's North Bridge and RAM also more hungry.
Unfortunately, the hardware automatically report the revised energy requirements, so miserable trial and error needed to find the sweet spot. Adjust Volt is a danger, more cumbersome and full of business.
Some overclocking friendly boards can automatically adjust the voltage for you, but understandably conservative about it, so for very large overclock you will have to configure yourself. This should be done by smaller increments possible, establishing how many volts reboot reset the needs embiggened CPU, as low as possible, essentially, as too many shots it can fry it. Establish in advance what your chip out-of-the-box are volts and, through a mixture of common sense and google, determine the number who are not going to risk higher. We have increased our Q6600 13 to 1.4V, which is a big enough increase modding volts going. It is not just a question of so-called core CPU or – as you go for large overclock, you'll find is having to play with the likes of the arcana of the CPU PLL and FSB termination. Once again, as long as you raise things in small increments of risk of killing your RAM chip or motherboard is quite minimal.
It is a different matter with AMD, that some time ago have had an integrated memory controller, which allows the chip to communicate more directly with the RAM, which means it does not have an FSB as such. Instead, you are overclocking something known as the HyperTransport bus, which is achieved in much the same way, but may require the reduction of NT stability multiplier itself to retain the package of speed. If you went to one of the last AMD Phenom Black Editions, you will find that comes with the unlocked multiplier, making overclocking easier matter.
By contrast, overclocking a graphics card is dead simple. As more self-contained piece of hardware, there is nothing of this confusing business multiplier or FSB, only overclocking the card itself, find the appropriate speeds for both GPU and memory card. Free software – some of them officers NVIDIA / ATI driver Plug-ins – will do the trick from within Windows, and built-in safety and stability testing cuts make it extremely difficult to damage the card, although course goes beyond the warranty. It also has grown a little more complicated than the afternoon you may need to overclock the shader clock as well as of the GPU and RAM for the better increases. For NVIDIA cards, which used to be that it was twins at the speed of the GPU, which means an increase in one had a synchronous effect on the other, but not for long as they have been able to change independently. So, if you reach the speed barrier in the GPU, may still be possible to gain more performance from the card by pressing the shader clock a little more.
While the current situation is that can overclock everything and be very sure it will work, the future of the form is harder to call. One thing seems certain: there is a dirty little secret nerdy, but a practice increasingly common, especially with Core 2 chips. There is a great cooler industry aftermarket support, and even economic plates can handle a bit of a lift off. If no overclocking will be easier, with more and better software to do so from within Windows, rather than from the BIOS, and possibly more in the way of automatic volt modding. But much depends on the future of desktop processing. There is a great war brewing between Intel and NVIDIA as to whether the CPU or the GPU be the principal on the PC of the future.
Intel are pushing ray-tracing using a multi-core CPU to graphics game, while that allows your recent NVIDIA GeForce cards for parallel processing, like video encoding and physics of the game, much faster than a CPU can handle. If any of these beds, overclocking will have to take into account. At the same time, slowly becoming more cores potentially reduces the need for overclocking conventional, such as raw clock speed continues to be a minor concern for multi-threading, and in the case of 3D cards, the number of stream processors and texture units. That will not stop anyone from trying, of course. Even if their effects are minimal, overclocking, always going to be a safer way to make a system to feel as its own rather than simply a collection of mass produced parts.
Modding the case is one thing, but what makes a PC's performance. When has been carefully adjusted to yield something to suit their own purposes, and has become something that feels like you've gone beyond what you paid for it, the scheme would be felt most unique of all green neon tube in the world could hope to achieve.
How to Overclock Your Graphics Card
