Sign in with
Sign up | Sign in
AMD FX Vs. Intel Core i3: Exploring Game Performance With Cheap GPUs
By ,
1. FX-4100 And Core i3-2100 Go Head-To-Head

At the end of January, we published our analysis of the sub-$200 gaming processor market called Picking A Sub-$200 Gaming CPU: FX, An APU, Or A Pentium?. We were surprised to find that Intel’s budget-oriented LGA 1155 offerings are surprisingly capable when it comes to handling modern titles. In fact, the $125 Core i3-2100 beat out AMD's entire line-up including top-tier Phenom IIs, Athlon IIs, APUs, and even the new FX models. Although they're easier to overclock, AMD’s best efforts could only achieve parity with the Core i3-2100, and Intel's Core i5 was so far ahead of the sub-$200 pack that it sat in a league of its own.

Now, we used a very high-end Radeon HD 7970 graphics card in that article because we wanted to isolate CPU performance. You can't draw conclusions about a CPU's potential when you're faced with a graphics card bottleneck, after all. But some of our readers rightly pointed out that, when it comes to building an inexpensive machine, our combination is unrealistic. A $110 CPU would never accompany a $550 graphics card. And if we used an entry-level GPU, the resulting bottleneck would have masked the differences between processors to a greater extent. The counter, of course, is that a cheaper graphics card would have also imposed lower resolutions and detail settings, shifting load back in the direction of the CPU.

As you know, though, we put a big emphasis on addressing your feedback, so we went back to the lab to run some follow-up data on two of the most interesting $120 options from our previous story. Intel's Core i3-2100 is the low-cost processor to beat, so we made sure to include it. On the other hand, with AMD's Phenom II and Athlon II lines disappearing from store shelves, the $110 FX-4100 represents that company’s best low-priced option.

Every game's workload is different, but Intel’s i3-2100, on average, achieved 18% higher minimum frame rates and 11% higher average frame rates compared to the FX-4100 in our previous story. As we said, though, that was with a Radeon HD 7970. This time around, we’re using a broader range of graphics cards ranging from the Radeon HD 5570 up to the Radeon HD 6950 to see if the bottleneck situation changes.


AMD FX-4100Intel Core i3-2100
Codename: ZambeziSandy Bridge
Process: 32 nm 32 nm
Cores (Threads): 4 (4)
2 (4)
Clock Speed (Turbo): 3.6 (3.8) GHz
3.1 GHz
Interface: Socket AM3+LGA 1155
L3 Cache: 8 MB3 MB
Thermal Envelope:
95 W
65 W
Online Price:
$110
$125


We also received some feedback on our test platform's memory configuration; it was suggested that AMD's FX might perform better complemented by higher memory data rates. So, this time we're using 8 GB (2 x 4 GB) of Corsair Vengeance DDR3-1600 at 8-8-8-24 timings.

2. Test System And Benchmarks

One of the challenges we face every time we analyze game performance is that most titles offer a large number of customizable graphics settings. Typically, we apply the same settings to every contender. The result is often that low-end GPUs hold performance back when they try to push the same options as the high-end models. Those results end up being unrealistically-low. Nobody plays a game on a Radeon HD 5570 at 12 FPS using the same settings as a Radeon HD 7970.

In order to keep our scores in tune with reality, we're going to try using different settings for each graphics card. How do we choose the right options? Some folks find that, in some games, 30 FPS is perfectly playable, while others won’t take any less than 60 FPS across all of the titles they enjoy. There’s no objective way to choose settings, so we test each game twice, each time using a different performance target.

Our first performance target is a 30 FPS minimum. The video game industry traditionally considers this to be the goal for smooth game play. So, we choose settings that keep the game's minimum frame rate around 30 FPS, but at the highest detail and resolution possible. If you value resolution and visual fidelity over all else, this is the kind of performance target you’re probably shooting for.

The second target is for gamers who want a more fluid experience. In this case, we’re looking for about 60 FPS average, and we don’t want to dip below 40 FPS. This imparts a much smoother feel than 30 FPS minimum, even though we’ll probably have to drop visual settings and resolution to achieve it with lower-end hardware. Gamers who enjoy first-person shooters often favor responsiveness over visual fidelity, especially in competitive environments.

Our FX-4100 is actually an FX-8120 with two of its Bulldozer modules turned off. We've run exhaustive testing against an actual FX-4100 to confirm that the performance is similar. The only complication is a lack of granularity in setting its Turbo Core multiplier. So, the CPU itself runs at 3.8 GHz (rather than ranging between 3.6 and 3.8 GHz). With that established, a 100 or 200 MHz difference isn't going to impact our results noticeably.

Finally, we want to point out that the $125 Core i3-2100 is only $3 less than the Core i3-2120, a CPU that is 200 MHz faster. Although the -2120 would be our recommended buy, we're using the -2100 because that's what we have on-hand.


Socket AM3+
LGA 1155
CPU

AMD FX-4100 (Zambezi), 3.6 GHz Base, 3.8 GHz Turbo Core

Intel Core i3-2100 (Sandy Bridge), 3.1 GHz, Hyper-Threading enabled
Motherboard

Biostar TA990FXE
Socket AM3+, Chipset: AMD 990FX

Asus P8P67 Pro
LGA 1155, Chipset: Intel P67 Express
Networking
On-Board Gigabit LAN controller
Memory

Corsair Vengeance LP PC3-16000, 2 x 4 GB, 1600 MT/s, CL 8-8-8-24-2T

Graphics

AMD Radeon HD 5570
650 MHz GPU, 1 GB DDR3 at 900 MHz

AMD Radeon HD 6770

850 MHz GPU, 1 GB GDDR5 at 1200 MHz

AMD Radeon HD 6850

775 MHz GPU, 1 GB GDDR5 at 1000 MHz

AMD Radeon HD 6950
800 MHz GPU, 1 GB GDDR5 at 1250 MHz

Hard Drive

Western Digital Caviar Black 750 GB
7200 RPM, 32 MB Cache, SATA 3Gb/s

Power

ePower EP-1200E10-T2 1200 W
ATX12V, EPS12V

Software and Drivers
Operating System
Microsoft Windows 7 x6, Service Pack 1, KB2645594 and KB2646060 installed
DirectX
DirectX 11
Graphics Drivers

AMD Catalyst 12.1

Benchmark Configuration
3D Games
Metro 2033
Version 1.0.0.1, Built-In Benchmark
Battlefield 3
Version 1.0.0.0, Operation Swordbreaker, FRAPS runs
Elder Scrolls V: Skyrim
Version 1.4.21.04, FRAPS runs
DiRT 3
Version 1.2.0.0, Built-In Benchmark
Just Cause 2
Version 1.0.0.2, Concrete Jungle Benchmark
StarCraft 2
Version: 1.4.2.20141, Tom's Hardware Guide Benchmark
3. Benchmark Results: Battlefield 3

Performance Target: 30 FPS Minimum

Battlefield 3 was obviously limited by the performance of our Radeon HD 7970 the last time we tested it. As such, we don’t expect our choice in processors to have much effect on the frame rates of our configurations. In the first round, our target is a 30 FPS minimum frame rate:

Only AMD's Radeon HD 5570 has to be forced under 1080p to hit our minimum target; we set it to run at 1280x720. Every other graphics card achieves our goal at 1920x1080.

And, as expected, there is little to no difference between CPUs as we step up from entry-level to mid-range graphics setups. The Core i3-2100 does manage a marginal lead with AMD's Radeon HD 6770, but nothing you'd notice during game play.

Unfortunately, while we see minimum frame rates that look good, the averages are very close to 30 FPS, too. So, let’s aim for a 40 FPS minimum and closer to 60 FPS average and see what happens.

Performance Target: 40 FPS Minimum, 60 FPS Average

This time we had to drop the Radeon HD 5570 to 800x600 to achieve our target. The Radeon HD 6770 dips to 1680x1050. Both the Radeon HD 6850 and 6950 are able to handle 1080p.

Even at our lower settings, the choice of CPU doesn't affect the outcome in an appreciable way. In fact, the FX-4100 scores marginally better using the Radeon HD 6850, although the advantage is not significant. Overall, we're not surprised by the outcome of our Battlefield 3 charts. After all, even AMD's powerful Radeon HD 7970 couldn't shift enough of the game's workload to our CPUs to expose a weak link.

4. Benchmark Results: DiRT 3

Performance Target: 30 FPS Minimum

DiRT 3 is another title that presents a moderate graphics workload, and consequently isn't really held back by our choice in processor. Only the dual-core A4-3400 APU lagged behind. Again, we don't expect to see much difference here.

All of our graphics cards manage playable performance at 1920x1080, and DiRT 3 deserves its reputation as a game that scales well with graphics hardware. None of the results are particularly telling of a trend one way or the other. Intel's CPU performs slightly better with the Radeon HD 6950, but again, that's nothing you'd pick up on while playing the game itself.

Performance Target: 40 FPS Minimum, 60 FPS Average

AMD's Radeon HD 6950 achieved our second goal at DiRT 3's highest detail settings (and with 8x MSAA applied) in the last test, so we won't re-test that board here.

We had to lower the resolution to 1440x900 in order for AMD's Radeon HD 5570 to make the cut, but its 6770 and 6850 handle 1080p without a problem. And although the bottom chart appears empty, all of the scores are above 60 FPS on both processors.

According to the minimum and average frame rates, Intel's Core i3-2100 has a slight advantage with the Radeon HD 5570 and 6850. Faced with such high frame rates, however, the benefit ends up less significant.  

5. Benchmark Results: Just Cause 2

Performance Target: 30 FPS Minimum

Unlike the prior two titles, Just Cause 2 did help demonstrate the difference between processors in our previous analysis. Last time around, the Core i3-2100 managed to achieve a 28% higher minimum frame rate and a 15% higher average frame rate compared to the FX-4100.

This game is hard on AMD's Radeon HD 5570; we had to drop to 1366x768 for a 30 FPS minimum frame rate.

Despite our previous benchmark results, CPU choice makes very little difference when we step down to lower-end graphics cards. Average frame rates are notably better than the other games we’ve tested, too (showing 40 FPS+ for all configurations). Maybe the load will shift to our CPUs when we raise the performance target to 40 FPS minimum and 60 FPS average?

Performance Target: 40 FPS Minimum, 60 FPS Average

Intel's Core i3-2100 has a very slight advantage paired up to AMD's Radeon HD 6950. But the 4 FPS spread isn't something you'd identify while playing through this game. This does support what we saw from the Radeon HD 7970, though.

We had to drop the Radeon HD 5570-based configuration to 720p and the 6770 to 1680x1050 in order to hit our performance target. As we've seen several times already, the Radeon HD 6850 and 6950 are able to handle 1080p without any trouble.

6. Benchmark Results: Metro 2033

Performance Target: 30 FPS Minimum

We know Metro to be a graphics-limited title capable of dropping very high-end boards to their knees. In our last story, however, we used the Very High quality preset and barely achieved a 16 FPS minimum frame rate. Now that our target is 30 FPS, we have to drop our settings and see what happens.

The results are fairly close across the board, although the Core i3-2100 demonstrates a lead in some parts of the benchmark when the Radeon HD 5570 is used. This is interesting because low-end GPUs most often limit graphics performance. But the results make more sense when you remember that a low-end card must necessarily be used in conjunction with lower resolutions and detail settings. The Radeon HD 5570 was the only card that couldn’t handle 1080p, so it was tested at 1366x768.

This is the first time we've seen a sizable gap between processors, so we're curious to see what will happen when we smooth out performance by targeting higher frame rates.

Performance Target: 40 FPS Minimum, 60 FPS Average

Faced with higher frame rate targets, we see noteworthy differences in performance. Finally, we see evidence of processor choice affecting performance in a definitive way.

We had to drop our Radeon HD 5570-based setup to 1024x768 to hit these numbers. And while the Core i3-2100 capitalizes on the lesser workload, the FX-4100 is unable to push frame rates much higher than it did in our prior analysis. The story is similar across all graphics cards.

The differences are too large to ignore. Interestingly, the FX-4100’s minimum frame rate never exceeds 34 FPS, suggesting a particular bottleneck that didn't affect the Core i3, which manages to hit a 45 FPS minimum.

7. Benchmark Results: The Elder Scrolls V: Skyrim

Performance Target: 30 FPS Minimum

Between our first sub-$200 gaming CPU article and now, Bethesda released a Skyrim patch that greatly improves performance. This game was particularly processor-dependent last time. That's not so much the case now.

There’s almost no differentiation between CPUs. Average frame rates are dismal, though, so let’s see how these competing platforms fare when the settings are lowered to facilitate better performance.

Performance Target: 40 FPS Minimum, 60 FPS Average

Little changes until we install AMD's Radeon HD 6950. Suddenly, Intel's Core i3-2100 takes off with a notable lead. The FX-4100 delivers playable frame rates, but the Sandy Bridge-based CPU maintains a 10 FPS spread.

8. Benchmark Results: StarCraft II

Performance Target: 30 FPS Minimum

Finally, let’s consider StarCraft II, a real-time strategy game notorious for its processor dependency. We start with the 30 FPS minimum frame rate target.

Our minimum goal of achieving 30 FPS proves irrelevant, as the demanding benchmark we use pushes both of these CPUs to their limits. The Core i3's advantage is clear. There’s no point to running a second set of benchmarks targeting 40 FPS when we can’t keep these platforms over 16 FPS, regardless of settings or resolutions.

To the FX-4100’s credit, this is a very demanding benchmark, and most StarCraft games won't push a machine quite as hard. There are some custom maps that tax processor performance, though, so the benchmark's validity stands. If you're wondering why the frame rate climbs over time, that's because units are destroyed and removed, alleviating the load. Regardless of the caveats, Intel's Core i3-2100 shows itself superior in this title.

9. Your Game And Performance Target Matter Most

Not only are we happy to address reader feedback, but we also take great pleasure in exploring areas of performance that might otherwise get ignored. Challenging dogma is part of what we do here, and every chart based on data gives us a little more information. So, what did we learn after today's experiments?

Despite the fact that Intel’s Core i3-2100 achieves 18% higher minimum frame rates (on average) and 11% higher average frame rates than AMD's FX-4100 when it's matched up to a very fast single-GPU graphics card, it was much rarer to observe an advantage in the same tests when we set a specific performance target. With a goal of achieving 30 FPS minimum frame rates, only one out of six tested games definitively favored Intel's budget-oriented chip.

But while a 30 FPS minimum is playable, it doesn't translate to a completely smooth experience, especially when the average frame rate hovers close to 40 FPS, as we saw in our tests. Competitive gamers looking for responsiveness want minimums in the 40 FPS range, with averages at or above 60 FPS. With this target in mind, we can add Metro to the list of titles that favor the Core i3, demonstrating between 27% to 32% higher minimums. In addition, Skyrim biases toward Intel's chip once we drop in a Radeon HD 6950.

The important message here is that, if you're concerned about a processor bottleneck, your favorite games and the performance you want to see from them are more influential than the price of your graphics card. At least up to a Radeon HD 6870 or GeForce GTX 560, we'd expect that to be the case.

What conclusions can we draw from all this? First of all, AMD’s FX-4100 isn't necessarily the disappointment it appeared to be in our sub-$200 gaming processor comparison if you match it up to a comparably entry-level graphics card. Equipped with anything slower than a Radeon HD 6950, you can set your resolution and detail settings as high as possible to maintain a 30 FPS minimum, and in most cases, the graphics card will emerge as your bottleneck. With a higher-end GPU installed (or a CrossFire/SLI arrangement), the CPU's limitations are more likely to be exposed. Oh, and take advantage of AMD's unlocked multiplier ratio to crank the clocks up as high as possible.

The good news is that AMD fans can still enjoy games on a capable machine without spending a ton of cash. With that established, though, getting in the door with an LGA 1155-based platform costs about the same and yields a more consistently-good experience. We've seen enthusiasts throw blame all over the place: review sites aren't picking the right benchmarks, developers aren't spending enough time optimizing for AMD's architecture, and Intel is squelching innovation. But it comes down to this: when a new game you’ve been waiting for gets installed on your machine, finger-pointing won't help you enjoy it any more if it behaves like Metro 2033, demonstrating between 27% and 33% higher minimum frame rates on the Core i3-2100. Even a $200 FX-8120 won’t solve your problem; our tests show that chip acts just like the FX-4100 in gaming environments.

Today, Intel's LGA 1155 platform remains the best bet for a gaming rig. And not only for its budget-oriented performance, which is great, but also for its potential. Start with a cheap Core i3 and an inexpensive discrete GPU. Then, upgrade later to an Ivy Bridge-based chip and a faster graphics card without imposing any sort of bottleneck. SLI and CrossFire are both viable with a fast-enough CPU (even splitting PCI Express connectivity between two x8 slots), and the $180 Core i5-2400 is a gaming beast that AMD's overclocked processors cannot touch.

AMD simply cannot counter those advantages right now. We must look to the Piledriver architecture and hope that our current assessment can be reevaluated later this year.