As it stands, the most powerful single-GPU graphics card money can buy is without doubt the Radeon HD 5870. Although the dual-GPU Radeon HD 5970 is AMD’s flagship graphics card, a pair of Radeon HD 5870 graphics cards operating in Crossfire mode are actually quite a bit faster thanks to their 17% core clock speed advantage.

Throwing a pair of Radeon HD 5870 graphics cards into a gaming system is serious business for a few reasons. First and foremost this is one of the most powerful dual-card combination's possible, and secondly at over $800 US it is also one of the most expensive. With all this GPU power many readers often ask us what kind of CPU they would require to fully utilize such a setup.

In the past we have done our best to address this question by testing a range of processors using the latest and greatest graphics cards. Today’s article is no different, as we set out to test a wide range of CPUs with a pair of Radeon HD 5870 graphics cards. However when compared to recent CPU scaling articles featuring the Radeon HD 5970, this version is a little different.

The next page titled “Testing Methodology” better explains what I mean, so make sure you read this page before proceeding to the test results. Also, please be aware that the results from this article can in no way be compared to previous articles such as those featuring the Radeon HD 5970, as the test conditions are very different this time round.

Already we have tested half a dozen processors since testing began in January of this year, with another half a dozen in the works. This article will feature four processors, two high-end models and two low-end models. The processor series in question are the AMD Phenom II X4/Phenom II X2 and Intel Core i7 9xx/Core i3 5xx.

In total this article alone features 576 unique results. However in total 3456 test were conducted, focusing on the average and minimum frame rates each test was carried out a total of six times allowing us to recorded the average result. This has been a massive undertaking and as I just mentioned, has been months in the making.

Each series will be tested at six different frequencies ranging from 2.0GHz through to 4.0GHz in 400MHz steps. For more information on how the testing has been conducted please read the next page...

The idea was to determine if a Core 2 Duo processor could deliver the same gaming performance with the Radeon HD 5970 graphics card as the mighty Core i7, and if so, at what operating frequency would the two meet. Such an article is useful for readers who are looking to upgrade their graphics card but wish to know if their current CPU is powerful enough to take full advantage of the upgrade.

However this time around we are testing a pair of Radeon HD 5870 graphics cards without enabling any AA/AF quality settings. Furthermore, we are not just testing at 2560x1600, and will instead include a low resolution test using 1680x1050. This will allow the high-end processors to really stretch their legs, and will uncover which processor would provide the best performance assuming no limits.

That said, these results will not necessarily reveal which processor is required to power a pair of Radeon HD 5870 graphics cards running in Crossfire mode. This is because the cards are not being pushed as hard as they could be given that AA/AF is disabled. Therefore, while our previous testing methodology may not have shown much of a difference in performance between a Core i7 processor operating at 3.6GHz and 4.0GHz for example, these non-GPU limited results that you are about to see likely will show quite a significant difference.

Something else worth noting is the fact that we will not only be including the results for the average frame rate, but also the minimum recorded frame rate. This is not something we have done before when running so many tests, as it is quite difficult to get accurate minimum frame rates.

For example, when testing with Company of Heroes with the Core i7 processor at 4.0GHz, we saw a consistent average frame rate of 155fps at 2560x1600. This frame rate would never deviate more than 1-2fps up or down. However the minimum frame rate would fluctuate between 40–80fps, making it difficult to accurately record this data.

Occasionally we would get a result as low as 2fps, which we decided to ignore as this quick drop in frame rate was likely caused by something else happening in the background. In the end we decided that the only way to accurately show the minimum frame rate was to take the average of six runs. Typically we take the average from three runs, which is quite time consuming in itself. Still, this was not enough to accurately record the minimum frame rate and we were forced to double our work load.

Finally, please note that the flagship processor from each processor series was used to record all data. For frequencies that required us to tweak any settings past their defaults we did so in a manner that would keep them as close to the default specifications as possible, for the fairest possible comparison. Also please note features such as Turbo Mode for the Intel processors has been disabled to keep the processors locked at the intended frequency.

Hardware - Intel Core i7 965 Extreme Edition - x3 2GB G.Skill DDR3 PC3-12800 (CAS 8-8-8-24) - Seagate 500GB 7200-RPM (Serial ATA300) - Radeon HD 5870 (1GB) Crossfire - Asus P6T Deluxe (Intel X58) Software - Microsoft Windows 7 Ultimate (64-bit) - ATI Catalyst 10.1

Hardware - Intel Core i3 540 - x2 2GB G.Skill DDR3 PC3-12800 (CAS 8-8-8-24) - Seagate 500GB 7200-RPM (Serial ATA300) - Radeon HD 5870 (1GB) Crossfire - Asus P7P55D-E Premium (Intel P55) Software - Microsoft Windows 7 Ultimate (64-bit) - ATI Catalyst 10.1

Hardware - Phenom II X4 965 Black Edition - Phenom II X2 555 Black Edition - x2 2GB G.Skill DDR3 PC3-12800 (CAS 8-8-8-24) - Seagate 500GB 7200-RPM (Serial ATA300) - Radeon HD 5870 (1GB) Crossfire - Asus M4A79T Deluxe (AMD 790FX) Software - Microsoft Windows 7 Ultimate (64-bit) - ATI Catalyst 10.1

The Crysis Warhead results are interesting, particularly given that this game is notoriously GPU bound. However with a pair of Radeon HD 5870 graphics cards at 1680x1050, quite a difference can be seen between the various configurations. Something that you will notice right away is that although the Phenom II X4 (4.0GHz) is just 5fps slower than the Core i7 (4.0GHz) processor when measuring the average frame rate, when looking at the minimum frame rate the Core i7 is 10fps faster than the Phenom II X4, which works out to be a huge 29.5% performance increase.

When looking at minimum frame rates the Phenom II X4 processor clocked at 4.0GHz is only able to match the Core i7 once it is clocked down to 2.8GHz. However the average frame rate of the Phenom II X4 at 4.0GHz is 55fps, which matches that of the 3.6GHz Core i7 processor. When both processors are clocked at just 2.40GHz the Core i7 more than doubles the minimum frame rate of the Core i7, while the average frame rate is just 17% greater.

Now at 2560x1600, where owners of a Radeon HD 5870 Corssfire configuration are more likely to game, we see that Crysis Warhead becomes far more GPU bound, at least as far as average frame rates are concerned. In fact, when looking at average frame rates the Phenom II X4 actually comes out on top and the 4GHz configuration favors AMD by 2fps. However the same 4GHz configurations also sees the Phenom II X4 render 5fps less when measuring the minimum frame rate.

While we only see a 6fps difference when looking at the average frame rate between the fastest and slowest configurations, there is a much more significant 14fps separating the fastest and slowest configurations when looking at the minimum frame rate. Clock for clock the Phenom II X4 tends to deliver the best average frame rates, while the Core i7 provides the best minimum results.

Although the Phenom II X2 series is generally much cheaper than processors from the Core i3 series, we are going to make the comparison anyway, just as we did with the Phenom II X4 and Core i7. As you can see this is a much more evenly contested match up, though Intel does still appear to have the upper hand. Although the Phenom II X2 is based on the same architecture as the Phenom II X4, the Core i3 has a number of technological advantages, such as its support for 4 threads.

When testing with Crysis Warhead at 1680x1050 the Core i3 displayed a massive performance advantage when clocked at 4.0GHz, rendering 24% more frames when measuring the average frame rate. However at 3.60GHz the Core i3 was just 4% faster when measuring the average frame rate. However once again it was the minimum frame rate that upset AMD and it was here that the Core i3 was a staggering 45% faster.

Now at 2560x1600 we start to see a different story. The processors are quite evenly matched, primarily because we are limited to 35fps. Still the Core i3 does provide a performance advantage over the Phenom II X2 when looking at the minimum frame rate.

Comparing the Core i3 to the Core i7 is interesting, especially at 4.0GHz where the Core i3 delivered 7fps more. This could have been due to the fact that at this extreme frequency the Core i3 runs much cooler and perhaps the Core i7 was suffering from some kind of throttling feature. In any case, at 3.6GHz and below this was not see again, as the Core i7 leads the Core i3.

Clock for clock the average frame rates of the Core i3 and Core i7 processors are surprisingly close. However it is the minimum frame rate that plays to the advantage of the Core i7 processor and at 4.0GHz we see that the Core i7 is 29% faster.

The Phenom II processors do not fare as well and clock for clock the Phenom II X4 is only able to match the minimum frame rate of the Core i3 processor. The Phenom II X4 and Phenom II X2 deliver similar results and it appears that the quad-core version only provides a very slight performance advantage.

Now at 2560x1600 the average frame rate of the Phenom II processors matches that of the Core i3 and Core i7 at frequencies above 2.4GHz. Even the minimum frame rates are very close, though the Intel processors do have an advantage here once again.

The Far Cry 2 results strongly favor the Core i7, which outperformed a 4.0GHz Phenom II X4 at just 2.4GHz! In fact it was not until we down-clocked the Core i7 to just 2.0GHz that the Phenom II X4 clocked at 4.0GHz was able to overtake it, and even then only by a small margin. Clocked at 3.2GHz the Core i7 was able to find the limits of the Radeon HD 5870 Corssfire graphics cards at 1680x1050 in Far Cry 2, which appeared to be around 155fps.

The best results the Phenom II X4 achieved was 108fps, which was only 3fps greater than the minimum frame rate recorded by the Core i7 at 3.2GHz, which is surprising. Clocked at just 2.0GHz the Phenom II X4 managed 62fps, while at the same clock speed the Core i7 produced 103fps, making it 75.5% faster!

Now at 2560x1600 things do not look much better for the Phenom II X4, as the 4.0GHz configuration was only able to beat the Core i7 when clocked at 2.4GHz and even then only by a very slim margin. Clock for clock at 4.0GHz the average frame rate favored the Core i7 by 15% and the minimum frame rate by 18%. However as the clock speed was gradually wound down the Phenom II X4 looked even worse, and at 2.0GHz it was on average 41% slower than the Core i7.

Far Cry heavily favored the Core i7 when compared to the Phenom II X4, and we are finding much the same when comparing the Core i3 and Phenom II X2 processors. When comparing the 4.0GHz configurations the Core i3 is 32% faster when comparing average frame rates and 38% faster when comparing minimum frame rates. Clocking the Phenom II X2 at 4.0GHz produced about the same level of performance as the Core i3 at 2.8GHz.

Increasing the resolution to 2560x1600 does very little to aid the Phenom II X2 processor. This time the 4.0GHz data shows the Core i3 to have a 35% performance advantage when comparing the average frame rate and a 41% advantage when measuring the minimum frame rate performance. It is also worth pointing out that the Core i3 at 4.0GHz produced a minimum frame rate of 68.6fps, which was higher than the average frame rate of the Phenom II X2 at 4.0GHz!

At 4.0GHz the Phenom II X4 just managed to out muscle the Core i3. Although the Phenom II X4 was 19% faster when comparing the average frame rate data, it was just 3.5% faster when looking at the minimum frame rate. Far Cry 2 is obviously optimized for more than 2 threads, as the Phenom II X2 was considerably slower than the Phenom II X4. With both processors clocked at 4.0GHz, the average frame rate of the Phenom II X4 was 57% greater, while the minimum frame rate was 30% higher.

Similar performance trends were seen at 2560x1600, though with the Radeon HD 5870 Corssfire headroom reduced the Phenom II X4 was able to close in on the Core i7. The Phenom II X4 was able to maintain its lead over the Core i3, while it was still considerably faster than the Phenom II X2.

The S.T.A.L.K.E.R.: Call of Pripyat results are more competitive, though it would seem that the Phenom II X4 still needs around a 400–800MHz clock speed advantage to beat the Core i7. When clocked at 4.0GHz the Core i7 was 9.5% faster when measuring the average frame rate, though it was 24% faster when looking at the minimum frame rate. Similar performance trends were seen scaling all the way down to 2.0GHz!

Interestingly at 2560x1600 the Phenom II X4 overtakes the Core i7 when clocked at 4.0GHz, at least when looking at the average frames per second data. With 113fps the Phenom II X4 outgunned the Core i7 by just 2fps, while it was 2fps slower when measuring the minimum frame rate. Similar performance trends were seen right down to 2.8GHz, while the Core i7 was able to match the average frame rate of the Phenom II X4 at 2.4GHz and overtook it at just 2.0GHz.

Again the S.T.A.L.K.E.R.: Call of Pripyat performance is quiet competitive, and again we notice that while the Phenom II X2 provides better average frame rates, the Core i3 is faster when comparing minimum frame rates.

This same trend is also seen when testing at 2560x1600, though this time with the exception of the 4.0GHz configurations. When clocked at 4.0GHz the Phenom II X2 was found to provide a greater average and minimum frame rate when compared to the Core i3, albeit by a very small margin.

S.T.A.L.K.E.R.: Call of Pripyat is clearly not designed to utilize more than two cores, as the Phenom II X4 and Phenom II X2 processors delivered virtually the same performance. The Phenom II X4 and Core i3 processor delivered similar average frame rates, while the minimum frame rate strongly favored the Core i3 processor.

Now at 2560x1600 the margins become even smaller, and the Phenom II processors are able to take charge. The Core i3 and Core i7 do appear to maintain an advantage when looking at the minimum frame rates, while AMD was slightly faster when measuring average frame rate performance.

World in Conflict is a very demanding game for both the GPU and CPU, for this reason it has been featured in a number of our CPU scaling articles. As you can see, like Far Cry 2 this game really tortures the Phenom II X4. When clocked at 4.0GHz the Phenom II X4 was out right slower than the Core i7 at just 2.4GHz. Clock for clock at 4.0GHz the Core i7 was 44% faster when measuring the average frame rate, and 67% faster for the minimum frame rate. Similar performance trends were seen all the way down to 2.0GHz!

Despite increasing the resolution to 2560x1600, the Phenom II X4 still struggles against the Core i7. Clock for clock at 4.0GHz the Core i7 is 32% faster when measuring average frame rate performance. However it is the minimum frame rates that kill the Phenom II X4, as the Core i7 is 89% faster here.

Again we find that the Phenom II X2 struggles with World in Conflict, and at 1680x1050 the 4.0GHz configuration was only able to match the Core i3 when clocked at 3.2GHz. The Phenom II X2 appeared to need a 400–800MHz clock speed advantage to beat the Core i3 in this game.

The 2560x1600 results are no different and here we again see the 4.0GHz Phenom II X2 delivering similar performance to that of the Core i3 clocked at just 3.2GHz.

When testing with World in Conflict we again see that the quad-core processors do in fact have an advantage, as the 4GHz clock for clock data shows the Phenom II X4 enjoying an 18% performance increase over the Phenom II X2. However it is the Core i7 that performs best in this game and although the Core i3 was able to beat the Phenom II X4, it was still considerably slower than the Core i7 on a clock for clock basis.

Even at 2560x1600 the Phenom II X4 was unable to match the Core i3 processor and at each frequency it was seen trailing by a few frames per second. The Core i7 remained top dog and was considerably faster than the Core i3 in this game.

The Company of Heroes results also strongly favor the Core i7. When comparing the clock for clock data at 4.0GHz we see that the Core i7 was 37% faster when measuring the average frame rate, and 85% faster when looking at the minimum frame rates. As we clock the processors down the minimum frame rate data strongly favors the Core i7, and in some cases it is 2x and even 3x faster than the Phenom II X4.

Increasing the resolution does little to help the Phenom II X4 out. At 4.0GHz the Core i7 is now just 6% faster when measuring the average frame rate, though it is still 55% faster when looking at the minimum frame rate. In fact the Phenom II X4 clocked at 4.0GHz was still unable to match the minimum recorded frame rate of the Core i7 at just 2.8GHz.

When measuring the average frames per second the Phenom II X2 looks quite impressive, delivering roughly the same performance as the Core i3 on a clock for clock basis. However it is once again the minimum frame rate that lets the Phenom II X2 down, and at 4.0GHz the Phenom II X2 is only able to match the Core i3 at 3.2GHz. As the clock speed is reduced the situation becomes worse for AMD and at 2.4GHz the minimum frame rate of the Phenom II X2 is only half that of the Core i3.

The situation only slightly improves for the Phenom II X2 at 2560x1600, and again while the average frame rates are much the same, it is the minimum frame rate that lets the Phenom II X2 down.

Although Company of Heroes does not appear to be all that well optimized for more than two cores, the Core i7 is considerably faster than the Core i3 when comparing clock for clock data. Then on the other hand, the Phenom II X4 and Phenom II X2 processors delivered very much the same performance on a clock for clock basis. However the quad-core processor did appear to have an advantage when measuring minimum frame rate performance.

Now at 2560x1600 the Core i3 is much slower than the Phenom II X4 when looking at the average frame rate. However the 4GHz data shows that the minimum frame rate of the Core i3 is 25% greater, despite the Phenom II X4 having a 14% performance advantage when measuring average frame rate performance.

Interestingly, while the Phenom II X4 and Phenom II X2 processors displayed similar average frame rates at 1680x1050, the Phenom II X4 provided a significant advantage at 2560x1600. This was largely due to the fact that the average frame rate of the Phenom II X4 increased at 2560x1600, though this is not uncommon when using high-end multi-GPU setups.

Supreme Commander is much more evenly contested and here we see the 4.0GHz clock for clock performance to be very close. The Core i7 is just 1.5% faster than the Phenom II X4 when looking at the average frame rate performance. However the Core i7 is still 24% faster when measuring the minimum frame rate.

Similar performance trends are seen at 2560x1600, though the margin has been reduced slightly as now the Core i7 is 24% faster when measuring the minimum frame rate performance. Again, although there was not a huge difference in performance when measuring average frame rates, the Core i7 was considerably faster when looking at the minimum recorded frame rate.

The Supreme Commander performance at 1680x1050 is extremely competitive, and we were surprised to find that for the most part the Phenom II X2 lead the Core i3. However at 4.0GHz the results were even, which was also the case at 2.0GHz.

Again the Supreme Commander results are very even, and we see this once again at 2560x1600. The performance gains were steady from 2.0GHz through to 4.0GHz, and both processors saw similar gains every step of the way.

Supreme Commander is certainly not optimized for more than two cores, as the Phenom II X4 and Phenom II X2 configurations deliver virtually the same results. Both were on par with the Core i3, while the Core i7 provided a slight performance advantage when looking at the 4GHz data.

Now at 2560x1600 very little changes, as the performance trends remain very much the same. The Phenom II and Core i3 processors all deliver similar results at each frequency, while the Core i7 provides a slight performance advantage.

There are a few ways in which these results can be taken. For example, in almost every test the Core i7 was found to be far superior to the Phenom II X4, and while these processors do not compete directly in the market place, they are flagship products.

The most expensive Phenom II X4 processor is the 965, which costs just under $200 US. The cheapest Core i7 processor is the 920, and at $290 it is considerably more expensive. Then of course you have the associated motherboards and memory kits, all of which are much more expensive for the Core i7 platform.

Something worth considering is that although the Core i7 clearly is the superior processor, can all that extra horse power be converted into better gaming performance? When we conducted our CPU scaling articles with the Radeon HD 5970 we found that the extra power of the Core i7 was of no benefit. This was due to the heavy use of AA/AF, which placed limitations of the GPU’s before the CPU.

Therefore it could be said that when enabling the kind of quality settings that owners of Radeon HD 5970 or Radeon HD 5870 Corssfire configuration would generally play with, there is almost no difference in performance between the Phenom II X4 and Core i7 processors.

However what the data recorded in this article suggests is when future games became more demanding on the CPU, the Core i7 is going to provide a higher level of performance. While this is a very likely scenario, the Core i7 has been available for well over a year now and in that time we have not seen games become most CPU dependent. Therefore in terms of value, the Phenom II X4 still makes more sense for gaming purposes.

Getting away from the Phenom II X4 vs. Core i7 comparison for a moment, we found the Core i3 vs. Core i7 comparison to be just as interesting. Now keep in mind that the Core i3 540 processor costs half as much as the cheapest Core i7 processor. For the most part the Core i3 was not a great deal slower than the Core i7, and more often than not provided higher minimum frame rates than the Phenom II X4.

In fact, we often found that the Core i3 and Phenom II X4 would deliver very similar clock for clock performance when comparing the average frame rate. It is almost guaranteed that if we were making heavy use of AA/AF quality settings, that there would be almost no difference in performance between the Core i7 and Core i3 processors. Still, when conditions are right, like what we saw when testing Far Cry 2, the Core i7 can be over 70% faster than the Core i3 when comparing clock for clock data.

Overall we have certainly come across some interesting results that were very different to those of our Radeon HD 5970 article, primarily because much less stress was being placed on the GPUs. Still to come we have the Core i5, Core 2 Quad, Core 2 Duo, Pentium G, Pentium E6xxx, Phenom II X3, Athlon II X2, Athlon II X4 and Sempron 1xx processors, so stay tuned...