Crank that S#!t Up
Perfect World Beta
Life After DX 11
AMD Contest Group
Final Round FiringSquad-Intel Edito...
Round 3 Editors Challenge Sponsored...
S.T.A.L.K.E.R. Screenshot Contest
Top 10 Challenge Round Sponsored by...
Editors Challenge Sponsored by Inte...
FiringSquad Editors Challenge Round...
Synchronous Failure (1147)
||23 entry(ies) in this category
| Intel Core 2 Duo E6600 Review (11 comments )|
by: jacobvandy (1636) | Posted in cluster Round 3 Editors Challenge Sponsored by Intel
Posted 74 months ago ( edited 74 months ago ) in category DEFAULT
When the first benchmarks for the Core 2 Duo line began to appear, the internet was wildly abuzz. Intel's new chips, codenamed "Conroe," simply obliterated competing parts in every price range. In fact, the E6600 model held its own against AMD processors that cost twice as much. Overclocking enthusiasts were reaching incredible speeds with little difficulty, while gamers everywhere began switching to the LGA 775 form factor. Any inquiry as to which CPU should be used for a new build was (and still is) answered with a resounding "C2D!" With nothing new in the immediate future to answer the challenge, AMD was forced to cut prices drastically and hope for the best. The tables had turned, and the performance crown was returned to the largest chip manufacturer in the world.
|» MEDIA (7)|
The Intel Core 2 Duo Processors
Figure 1: 3DMark06 Results
Figure 2: Supreme Commander Results
Figure 3: Company of Heroes Results
Figure 4: F.E.A.R. Results
Figure 5: WinRAR and Nero 7 Results
Figure 6: DivX and SuperPi Results
Thanks to the 65 nanometer manufacturing process, Core 2 Duo CPUs contain about twice as many transistors as the Athlon 64 X2 series and on a smaller die, to boot. Smaller die-size means a chip that is cheaper to produce, and that is beneficial to producer and consumer, alike. The E6600 contains 4 megabytes of L2 cache, which is 2-4 times the amount present in the X2 line, depending on model; In this regard, the simple idea that more is better has a sizable effect on performance. You see, the cache on a processor is like the desk in your office. Everything that you use frequently - pen, paper, calculator, and stapler - you keep right at hand. From time to time, you may need something else, perhaps a book from a shelf on the wall. Before you can work with that book, you need to get up, walk to the shelf, pick it up, and bring it back to the desk. In a CPU, frequently used instructions are kept in high-speed, embedded memory called the cache. Compared to how fast the CPU works, accessing the main memory on a RAM module is incredibly slow. The more cache you have, the more items you can store in your desk, and the less often you have to fetch something.
As evidenced by the Core 2 architecture, Intel has kicked its habit of implementing brute force to achieve greater computing power. The Pentium 4 line of processors continued to grow in not only clock speed, but power consumption and heat output. Eventually, the progression began to approach a glass ceiling of sorts. Initial predictions that we would be seeing a P4 running at frequencies upwards of 10GHz were obviously incorrect. As such, Intel never finished producing Pentium 4 processors rated at 4GHz and above. On the consumer side, many were perplexed as to exactly why their brand new 3.2GHz CPU was somehow worse than an Athlon 64 rated at a lowly 2.4GHz. Tearing a page from the AMD handbook, a C2D processor can perform more operations per clock cycle, resulting in greater efficiency at lower clock speeds.
Multi-core processors have nearly become the mainstream in the past couple of years. In the meantime, newer games and applications have become increasingly likely to support the extra cores. While dual-core performance isn't quite like having two processors instead of one, the ability to process two threads at once helps considerably when multi-tasking. You can watch high-resolution video while encoding music or play a game while running an anti-virus scan. More often than not, dual-core can mean the difference between a slight drop in frame rate and a freeze that lasts many seconds. Beyond multi-tasking, several aspects of a next-gen game can be distributed among each independent core. While one is crunching physics calculations, another can be handling artificial intelligence routines. Supreme Commander is a prime example of this, with significant advantages being reported by those running it on dual- and quad-core processors.
» The Benchmarks
Lacking several processors to compare, I've evaluated my own E6600 at stock and overclocked frequencies. The test system consists of the following:
• Intel Core 2 Duo E6600 (2.4 and 3.0GHz)
• Gigabyte GA-965P-DS3 Revision 2.0 (BIOS version F7)
• 2GB OCZ Platinum PC2-6400 (OCZ2P8002GK)
• ATi Radeon X1900XT 512MB (Catalyst 7.3)
• Maxtor 6L300S0 (7200 RPM, SATA, 16MB Cache)
• Windows Vista Ultimate 32-bit
3DMark06 - Basic Edition
One of the more popular benchmarking applications, 3DMark stresses your machine and assigns it a set of scores that can be easily compared to others. Controversy erupted when graphics vendors were accused of cheap optimizations implemented specifically for 3DMark tests, but they remain a highly regarded index of system performance. The free version of the software allows you to run the basic tests at the default settings, including four "game" tests and two tests that use solely the CPU to render a scene. Overclocking yielded unsurprising results: the CPU score was the only to increase significantly, bolstering the overall score by a few hundred points. See Figure 1. Undeniably, the game test results were limited by the GPU due to the extreme load placed upon it, thus the increase in processor capability had little effect.
Heralded by many as the new king of real-time strategy, the sheer scale of warfare in this game can take a serious toll on your system. Adding "/perf" to the command line of a desktop shortcut will run an automated performance test, simulating a match between several computer opponents. It consists of a series of scripted events, such as a large-scale naval battle, some ground skirmishes, and a whole lot of nuclear missiles decimating everything. Performance is monitored throughout and a detailed results log and composite score is provided. Overclocking afforded substantial improvements across the board, reinforcing Supreme Commander's reputation as a very CPU-intensive game. For a comparison between the low, medium, and high graphics presets, see Figure 2.
Company of Heroes
One of my personal favorites, Company of Heroes is another classic RTS from Relic. In the video settings tab of the options, there is a performance test that renders an in-engine cinematic using the options you have selected. Since there are no presets aside from the auto-configuration feature, I tested using the lowest- and highest-possible settings. Well, highest settings except for the elusive "DX10" shader option. Grayed out and unable to be selected, it taunts me... but I digress. See Figure 3 for minimum, average, and maximum frame rate results. Judging by the difference in the impact of extra CPU horsepower between low and high settings, I think it is safe to say that CoH is fairly GPU-dependent.
Though it no longer brings even the best gaming rigs to a grinding halt with its dynamic lighting and soft shadows, this eerie FPS still aims to please your visual palate. Like CoH, it provides a performance test in the graphics options menu to allow you to preview the impact your selections will make. Each of the five fidelity presets made trial runs through the flyby over firefight, fancy water effects, and explosion. See the average reported frame rates in Figure 4. Some sort of anomaly occurs when running the "minimum" option: the performance is actually worse than "low," the next higher preset! Upon further investigation, I determined that DirectX 8 shaders were the culprit. I am not quite sure exactly why this is; perhaps it would make an interesting subject for another time.
Tasks like file compression are where a Core 2 Duo chip really shines. Raw computing power is the name of this game, and the E6600 does not disappoint. The time taken to compress several image files using the "normal" and "best" settings was recorded. See Figure 5. As expected, increased CPU throughput was very beneficial. While a difference of a few seconds doesn't seem like much, it is actually a %10 decrease. So, if you needed to compress much more data, say a few gigabytes, the slight overclock could save you several minutes.
A single WAV file of about 100MB in size was encoded using various bit-rates. The recorded times are displayed in Figure 5. A consistent reduction of about %20 is brought about by the overclock.
A single 1.63GB AVI file, recorded with FRAPS, was encoded using the Home Theatre profile with a bit-rate of 780Kbps. See Figure 6 for the results.
Though it may be the simplest of all widely-used CPU benchmarks, SuperPi is popular among the overclocker enthusiast crowd. The best of the best wear the ability to calculate pi to a million decimal places in less than ten seconds like a badge of honor. SuperPi Mod is a specially modified version that includes higher precision in the timer results, cheat-protection, checksum support, and bug fixes. I compared the results of calculations to one million, two million, and four million decimal places. See Figure 6.
• unprecedented performance
• low power consumption and heat output
• great bang for your buck
• insane overclocking ability
» The Verdict: 97%
Intel has certainly delivered an outstanding performer in the E6600. It's about time, too. My career as a system builder, though admittedly short, has seen nothing but AMD dominance in the mainstream market until Core 2 Duo convinced me to make the switch. I can tell you from plenty of personal experience that each of the games I tested runs extraordinarily well, on the highest settings, with the E6600.
In addition, Core 2 Duo chips consume less power than their Pentium 4 and Athlon X2 counterparts. Though not necessarily a major consideration when deciding on a processor, it can make a difference where it counts: your wallet. Saving a few bucks a month by choosing a C2D over a P4 may not sound like much, but for a large office or school, that can add up quick. Not to mention, less power consumed means less heat is created, resulting in lower overall cooling requirements for your system.
Though the E6600 defaults to a clock rate of 2.4GHz, I've been running it overclocked to 3GHz on the stock heat sink and fan for the past few months. Since the final clock speed is the multiplier - 9 on the E6600 - times the front side bus speed, increasing the FSB from 266 to 333MHz accomplishes this. I didn't need to increase the voltage for this relatively mild overclock, so the extra heat output isn't much. I could easily drop $40 on a decent aftermarket cooling solution, but it simply is not required. This ultimately equates to a free increase in CPU speed with little to no consequence. Essentially, since the only difference between the E6600 and E6700 or even the X6800 is a few hundred megahertz, you can save yourself hundreds of dollars and enjoy the same or better performance! This, combined with the fact that the E6600 is currently the cheapest way to get that 4MB of L2 cache, is very strong evidence of its position as the sweet spot of price-to-performance ratio.
The only apparent downside to the Core 2 Duo E6600 is the price tag. $300 may be just out of reach for those concerned with keeping a budget. However, no matter which model of C2D you choose, your satisfaction is guaranteed. Sub-$200 solutions are available now in the form of the E6300 and E4300, but keep an eye out for future revisions at a similar price point.
|11 User Comment(s) • 8 root comment(s)|
» Note: You need to be logged in to write a comment!Login here, or if you don't have an account with FiringSquad, register here, it's FREE!
My Media-Blog categories
No categories created yet.