AMD took the lead by announcing the first quad-core CPU to be offered on the USD 100 range, Athlon II X4 620. Thus this CPU promises to deliver a terrific performance to entry-level PCs. Will this promise hold true? Let’s see.

Athlon II X4 620 runs internally at 2.6 GHz and has four processing cores, each one with a 64 KB + 64 KB L1 memory cache and a 512 KB L2 memory cache. No L3 memory cache is provided, being the main difference between Athlon II and Phenom II. Like Phenom II, Athlon II is a socket AM3 CPU, and thus can be installed on socket AM2+ or socket AM3 motherboards, depending on whether you want DDR2 or DDR3 memories on your computer, respectively.

Athlon II X4 620Figure 1: Athlon II X4 620.

AMD is targeting Athlon II X4 620 to entry-level PCs with motherboards based on the AMD785G chipset.

We have several different ways to select a competitor to Athlon II X4 620. The most obvious would be to select a CPU from Intel with the same price tag. The problem with this methodology is that the total price of the two computers would be different, as motherboards for Intel CPUs tend to be more expensive than the ones targeted to AMD processors. Since no one can run a CPU outside a computer, we have to think about the total price of the system, especially on entry-level PCs, where every little bit counts.

This way we decided to compare two systems with a similar price for the CPU + motherboard combo, quoted on the same store ( With Athlon II X4 620 costing USD 107.99 and the motherboard we picked (Biostar TA785GE 128 M) costing USD 80, we had a budget of USD 187.99 to buy an Intel counterpart. Since we were using a motherboard with integrated graphics, we had to pick a similar product on the Intel side, and we ended up choosing Intel DG45ID motherboard (based on the Intel G45 chipset), that costs USD 99.99, leaving us with USD 88 to choose a CPU. The CPU from Intel that most matched this budget was Pentium E6300, which we bought for USD 84.

Now we had two systems with comparable costs, since memories, hard drive, etc were the same. But there was still one important detail. For several of the tests – especially gaming – we were going to rely on the motherboard integrated video, meaning that in fact we were going to measure the motherboard video performance, not the CPU performance.

To overcome this issue, we decided to perform two tests with each CPU: first with the on-board video, and then disabling the on-board video and installing a mid-range video card (a GeForce 9600 GT was chosen for this task). This way we could easily simulate the scenario where the user installed these CPUs with a “real” video card, allowing us to compare the performance exclusively from the CPU, not putting the motherboard performance into the equation. It is important to note that under this scenario we were still comparing two systems with practically the same price.

This way this will be a very interesting review were we will cover both scenarios: users using on-board video and users using a “real” video card.

Before going to our results, let’s compare the main specs from the two CPUs included in this review.


Gabriel Torres is a Brazilian best-selling ICT expert, with 24 books published. He started his online career in 1996, when he launched Clube do Hardware, which is one of the oldest and largest websites about technology in Brazil. He created Hardware Secrets in 1999 to expand his knowledge outside his home country.