Original Link: http://www.anandtech.com/show/1201

With new technology constantly being developed and released into the high end market, it is sometimes easy to overlook the slightly less glamorous world of budget microprocessors. It's been a while since we've taken a look at what AMD and Intel have to offer in the area of low cost computing, and our curiosity recently got the better of us.

We were particularly curious about what you could get for $100, and it turns out that there are quite a few CPUs that you can get for less than the price of a motherboard. Currently, the budget market is made up of low end Athlon XP, Celeron, and Duron processors. There aren't any Pentium 4 processors that come in under our $100 price point, but we've included the Pentium 4 1.8A (Northwood) as a reference point for the Celeron processors.

Performance is always being pushed in the high end market, but it is arguably even more important in the low end systems. If we are trying to save money on a computer system, we want our dollar to go as far as possible, so price/performance is the most important factor when determining components to fill a budget box. Just because we want to save money doesn't mean we want to suffer a huge performance loss. With the price of PCs that perform well dropping all the time, it becomes easier for those who haven't yet entered the digital realm to join the party. Of course, the last thing someone wants when they first start up their new computer is to be frustrated by lackluster performance. Hopefully this article will serve to help people make the best possible decision when it comes to budget computing.

These Sub-$100 CPUs serve as decent upgrades for aging systems (e.g. the P3-800 that is barely chugging along) when combined with a new motherboard, but they are also the heart and soul of many of today's sub-$1000 PCs that you'd find in the retail market. Walk into any Best Buy or CompUSA and you'll find tons of PCs selling from $400 - $600. The OEMs making these systems are cutting corners in every way possible, so you had better believe that one of these CPUs we're comparing today will be under the hood. Retail customers should pay close attention to the results of this roundup — they may be even more shocking than expected.

When looking to get the absolute maximum performance out of every dollar spent, overclocking should be considered. We are hoping to address the overclockability of these budget processors in an upcoming article, but for now, we will only be looking at stock speeds.

Before we get to the tests, let's take a look at the processors.

The Contenders

Since this is a budget roundup, price is a very important factor in decision making. To get an idea of how current pricing is shaping up, here are the prices of these processors at the time of publishing, sorted from the most expensive to the least (pulled from our DealTime engine).

 Processor  Price
Intel Pentium 4 1.8A $120
AMD Athlon XP 2600+ (2083MHz) $88
AMD Athlon XP (Barton) 2500+ (1833MHz) $86
Intel Celeron 2.6GHz $85
AMD Athlon XP 2400+ (2000MHz) $68
Intel Celeron 2.4GHz $68
Intel Celeron 2.2GHz $67
Intel Celeron 2.0GHz $65
AMD Athlon XP 2200+ (1800MHz) $63
AMD Athlon XP 1700+ (1466MHz) $56
AMD Duron 1.6GHz $41

The prices fall where we would expect. Intel processors are priced near AMD CPUs with similar model numbers. That makes a price–to-performance comparison fairly simple, as the only factor we really need to consider is performance.

The Athlon XP processor has been in the spotlight for quite some time. Over the years, what used to be high end processors are given new life as budget products. The technology behind the Athlon XP and Pentium 4 1.8A are very well documented, so we'll spend some time speaking about the other players in this review.

Intel Celeron

The main difference between a Pentium 4 processor and a Celeron is cache (high speed memory on the processor core) size. The Celeron takes a cut in L2 cache from 512KB down to 128KB. The L1 cache in the Celeron remains unchanged from that of its big brother. Cutting down the L2 cache's size will increase cache misses (number of times when the information that the processor needs is not located in the cache), which will slow down the processor while it has to wait for its data.

The Celeron processors are also limited to a 400MHz system bus, which, in turn, limits RAM speeds on the system to 133MHz (DDR266) when used on 865 or 875 based motherboards. Aside from these, the only other difference between Celeron and Pentium 4 is that none of the Celerons offer HyperThreading.

Celeron processors are available in many speed grades between 1.7GHz and 2.8GHz. For this comparison, the fastest Celeron under our $100 price point runs at 2.6GHz.

AMD Duron

Like the Celeron, the Duron is basically a stripped down version of a mainstream processor. In this case, we drop to a 64KB L2 cache. The L1 cache on the Duron remains at 128KB, giving the AMD budget line a larger overall cache than the Celeron. The Duron also operates on a 133MHz FSB, and there isn't a limit on RAM speed as there is with the Celeron line when used on any Socket-A platform.

The Duron processor is currently only available in three speed grades: 1.4GHz, 1.6GHz and 1.8GHz. For this review, we tested with the 1.6GHz model.

Test Setup

As many factors as possible were kept the same on both systems in order to help give a fair comparison. Of course, the exception is the motherboard. For the Intel system, we used an Intel D865PERL (865PE) board, while we employed an ASUS A7N8X (nForce2) for the AMD system. The common elements in the systems were:

ATI Radeon 9800Pro 256MB
2 x 256MB DDR400 at 2-3-3-6 (frequency chosen by the BIOS)
2 x Western Digital Special Edition Hard Disk Drive

Our systems were configured with on drive set up specifically for the Business and Content Creation Winstone tests, in order to minimize the impact of other system files on the performance of the benchmark.

For our AMD system, we installed the 3.13 nForce drivers. Intel's 5.00.1009 chipset drivers were used on the 865 board. Both systems ran with ATI's Catalyst 3.9 drivers.

In the BIOS of each system, we disabled USB, IEEE1394, onboard audio, and just about everything, but the parallel port. Our goal in doing this was to eliminate factors other than the processor in performance.

Business Winstone 2004

PC Magazine and VeriTest have just recently released a new version of their Winstone tests. The goal of the test is to model normal computer usage patterns and compare how well systems do across the board. For the Business version of the test, this involves surfing the internet, using Office, and running a handful of other common utilities and programs.

As we can see by the graph, all the AMD processors performed better at this test than any of the Intel CPUs. What's also important to note is that a score of a 10.0 here is about the performance level of a 1GHz Pentium III, which should help you put the Celeron's performance in perspective.

Content Creation Winstone 2004

This Winstone runs programs from Photoshop and Premiere to LightWave.

We can see again that every single AMD CPU is ahead of all the Intel CPUs.

DivX Encoding

As we have done before, we used Gordian Knot to encode Chapter 9 of the movie, The Sum of all Fears, with the DivX 5.1 codec. Again, we did a 2-pass encoding and averaged the frame rates of the two passes to get our score.

Generally, encoding is an area where Intel's processors shine, but this time, it's a different story. Only the fastest Celerons that we tested were able to keep up with the slowest AMD processors. The 2.6GHz Celeron only beats the 1.6GHz Duron by about half a frame per second. When we saw this test stacking up the way it has, we lost hope in any Celeron keeping up with even the least expensive AMD processors in any test.

3D Rendering

To test 3D Studio Max rendering performance, we rendered the SinglePipe2.max benchmark file and used the render time as the score.

In an area where Intel has traditionally done very well, AMD leads across the board. Only our P4 1.8A is able to edge out the XP 1700+ in this test.

Development Workstation Performance

As introduced with our Athlon 64 article, we are looking at compile times for the Quake III Arena source code. In running this test, we compiled the source for both Q3A and Q3TA in both debug and release modes using the Batch Build feature of Visual C++ 6.0. The build was run three consecutive times for each processor (in order to try to reduce file read time impact), and we used the third compile time.

Once again, the trend continues with AMD leading. We expected this test to favor AMD anyway, as compilation naturally favors a shorter pipeline (the impact of mispredicted branches is decreased).

Aquamark3 Performance

For Aquamark3 (and all our subsequent game tests), we followed the same procedure that we used in our recent video card articles. For this test, we only looked at the free version of Aquamark3, as that is what our readers use. Aquamark uses DX8 and DX9 pixel and vertex shaders.

The P4 1.8A is able to nudge out the 1.6GHz Duron processor in this benchmark, but everything else falls in favor of AMD.

C & C Generals: Zero Hour Performance

AMD also takes this benchmark, with the exception of the P4 1.8A leading the Duron.

GunMetal Performance

This game is a DX8 game with DX9 vertex shaders. It isn't possible to disable Antialiasing or Anisotropic filtering in this benchmark, so we ran at the same 4 x AA settings we used in our graphics card benchmarks.

For all these tests, AMD's CPUs outperformed Intel's.

Halo Performance

For the Halo benchmark, we used the PS/VS 2.0 rendering mode (DX9), and ran the timedemo with vsync off at 75Hz.

Under Halo, the only competitor with AMD is, again, the P4 1.8A (this time beating out the 1.6GHz and 1700+).

SimCity 4 Performance

In this almost disturbing test, we can see that AMD's fastest Athlon XP has nearly twice the performance in SimCity as the fastest Celeron. In fact, even the 1.6GHz Duron was 30.9% faster than the 2.6GHz Celeron.

Unreal Tournament 2003 Performance

Whether looking at the botmatch or the flyby benchmark, Intel can't touch AMD here. The performance trend continues very strong.

Warcraft 3: Frozen Throne Performance

This time we have the P4 1.8A saving some face for Intel again. But still, none of the Celerons can keep up.

Wolfenstein: Enemy Territory Performance

This game is based on an enhanced version of the Quake 3 engine and uses OpenGL.

Once again, the fastest XP is almost twice as fast as the fastest Celeron. This time, the Duron is an incredible 45.9% faster than the 2.6GHz Celeron.

Quake III Arena Performance

Finally, we have the definitive in CPU game tests: Quake III Arena. It's been around for a long time, but there are a great number of games based on the engine, so (like it or not) it is still relevant.

Here, we have another greater than 30% lead for the Duron over the fastest Celeron that we tested. The 2600+ outscores the 2.6GHz Celeron by over 69%. This agrees with the results that we've seen from Wolfenstein.

Final Words

It is very obvious from these tests which line of budget processors is worth the money. When we can find a 1.6GHz Duron for just over half the price of a 2.6GHz Celeron and get better performance consistently in almost every test we ran, the choice is clear.

It's obvious that the long pipeline of the Pentium 4 just can't handle the crippled cache of the Celeron. With more cache misses and pipeline stalls, the processor isn't getting as much useful work done as it is trying constantly to refill the pipeline. We are seeing these results for the same reason we saw the performance gains from the P4 Extreme Edition with its 2MB L3 cache: the pipeline needs to stay full for the P4 to really shine.

The Pentium III based Celerons offered, at one time, acceptable performance. However, it is clear that in the value segment today, Intel has nothing to offer but a high clock speed. AnandTech readers will know to stay away from the Celeron at all costs; however, what is troublesome are the number of retail customers who are faced with the decision between a higher priced 2.6GHz Celeron system and an Athlon XP 2200+. We would highly encourage system vendors like Compaq and eMachines to shift their low-end focus to AMD if their customers are of any importance at all. As we've seen through our extensive benchmarking, the Celeron's performance is truly dismal; so while Intel is quite competitive in the mid-range and high-end segments, their value processors are inexcusably slow compared to AMD.

This review really isn't complete without taking a look at overclocking performance. For enthusiasts who want a lot of performance for a small amount of cash (cache?), pushing a cheap processor beyond its limits is the way to go. Every overclocker remembers the original Celeron processor and its amazing ability to run incredibly fast because of its lack of cache. At this point (however unlikely), such a feature would be the only saving grace of the Celeron line. Of course, even if the Celeron is a good overclocker, it will be very interesting to see how high the Duron can be pushed with its cut cache as well.

The conclusion we can make from all this is that the Duron processor is a solid purchase. If you have the extra 40 to 50 dollars to spend, a Barton processor would be a nice addition to any system for that added dimension of performance to a tightly budgeted system. Hopefully, system builders will take note and start offering better performing systems for an even lower price based on the Duron processor rather than the Celeron. For those who want the cheapest possible system, AMD will give you the best performance every time.

Log in

Don't have an account? Sign up now