"Bucket" Compression
An old army joke about a compression bucket that the "experienced" senior comrades are instructed to bring to a young soldier, checking his quick wits, is more relevant than ever…

Continue reading →

Who needs these 64 bits?
On September 23, the new 64-bit processors from AMD, Athlon 64 3200+ and Athlon 64 FX-51 were officially announced. However, Internet users have long been familiar with these processors, since…

Continue reading →

MAX + plus II: FPGA Integrated Digital Device Development Environment
The general tendency for the development of the element base of digital circuitry, starting with the appearance of the first integrated circuits in the early 60s to the present, is…

Continue reading →

What is a model number?

Before comparing with competing processors, let’s look at the performance of Athlon XP processors with different CPUIDs.

The model number is simply the number of a specific model in the Athlon processor family. This is AMD’s official position and investing any other meaning in this figure is hardly a sign of special literacy. The model number is calculated from the set of tests. The test suite consists of three sections: Productivity, Visual Computing, and Gaming. The procedure for determining the performance of this set of tests is called the True Performance Initiative. We already spoke out negatively about the informativeness of testing in a large group of different tests. However, we are not opposed to such an approach if processors of the same structure are compared (in this case, Athlon XP family processors are compared). It is only important to understand that the result of such a procedure is a comparative assessment of AVERAGE. In some particular cases, the performance ratio may vary significantly.

In particular, we compared the performance of various processors of the Athlon XP family for a specific task: non-linear editing in Adobe Premiere 6.5.

The results presented on the graph are selected from a large set of measurements we carried out according to a single criterion – for these filters, the execution time is approximately the same, which allows us to better visually compare the results shown in the figure.

It can be seen from the figure that indeed, on average, a processor, for example, an Athlon XP 2500+ is faster than an Athlon XP 1500 processor in proportion to the model number value. However, in some cases (in Fig. 1 the values ​​for the Basic 3D filter, Athlon XP 2800+ and Athlon XP 2600+ processors) this proportionality is violated. This is due to the fact that the Athlon XP 2800+ and Athlon XP 2600+ processors have the same core clock speed. However, the Athlon XP 2800+ processor (on the new Barton core) has a larger cache size. On average, according to a large set of tests, it really has better performance than the Athlon XP 2600+, but if a particular task is insensitive to the cache size, the performance is the same. But on other tasks the reverse situation is also possible. In particular, Fig. 2 shows that when applying the Color Balance filter, the Athlon XP 2500+ processor on the Barton core, which has a larger cache volume, turns out to be faster than the Athlon XP 2600+ processor.

Two important conclusions follow from the foregoing:

The model number in the Athlon XP processor designation correctly characterizes the relative performance of various models, but characterizes it “on average.”
The model number is determined by the results of a special testing procedure (True Performance Initiative). The connection with the real clock speed is very indirect. However, improvements to the processor core (increasing the cache size, accelerating the system bus) that lead to increased performance are reflected in the test results. Therefore, in the vast majority of cases, Athlon XP 2800+ processors with CPUID = 6А0 (Barton core) and Athlon XP 2800+ with CPUID = 681 (Thoroughbred core) will show the same performance on average. Therefore, in particular, it makes no sense to go into the technical specifications of processors. Athlon XP 2800+ processors have the same performance, regardless of the core on which this processor is based. And the Athlon XP 3000+ is faster than the Athlon XP 2800+, despite the fact that some Athlon XP 2800+ models have a higher clock speed than the Athlon XP 3000+.
After we figured out the relative performance of the Athlon XP family of processors, we can proceed to compare them with Pentium 4 processors.

True, there is a certain difficulty: which particular models will we compare? Indeed, Intel Pentium 4 processors, unlike AMD Athlon XP processors, indicate the real clock frequency in the designation. We solved this issue simply. We understand that, despite such a long digression on the topic “what is the model number,” most readers will not be able to overcome the “magic of numbers.” For them, the number in the designation of AMD processors will remain rated in relation to Intel processors. We will not persuade them, just see how much this rating weighs in relation to the nonlinear editing task. We will start with the most current processor, Athlon XP 2200+.

Power line interference
Power quality issues play an important role in ensuring the sustainability of computers and peripherals and their longevity. Computers, like any electronic equipment that is powered by AC power, are…

...

Overview of Graphic Solutions for Laptops
Over the past year, there have been plenty of events in the graphics market. For quite some time, S3 was the undisputed leader in the production of graphic chips and…

...

The victory of the losers, or the story of the creation of FLASH memory (part4)
Our main goal was not to change anything during the production process, and to solve the problem of how to place electrons on a floating shutter. Initially, we abandoned the…

...

Athlon XP: not by number, but by reduction!
When this issue of the magazine was being prepared for publication, a remarkable event occurred: the next Microprocessor Forum. Probably, when you read this article, you already know that AMD…

...