Albatron Medusa Series
GeForce 4 Ti4280PV
Reviewed by Shadrach 12.17.2002

...and now, a word from our sponsor...

Before moving on to the benchmarks, there is a topic that is rarely brought to the attention of the reader. Initially, most reviews will post the system that is being used to benchmark a certain item.  True benchmarks are meaningless without being performed in a testing environment.  In college it is taught that scientific testing can be nullified unless the environment remains closely monitored. For benchmarking computer hardware, this entails using systems that are identical except for the piece being tested. 

What about drivers?  Personal experience shows that drivers can have as diverse effect on computer peripheral performance as hardware. This is the reason hardware vendors ask for driver versions in support scenario's. In some situations, certain drivers can be as effective as extreme overclocking! On indistinguishable systems, a GeForce 3 Ti200 could theoretically out-perform a GeForce 3 Ti500 with drivers alone.

What does it all mean?

What this all means is that when reading benchmark reviews on products, the consumer must pay strict attention to all the systems that are being used to perform the testing. It is also recommended that you find a system close to yours before judging the benchmarks. An ATI Radeon 7500 card will perform entirely different in an Intel PII 500 than it will in an AMD Athlon XP 2200.  Not to say that processor is the only difference.  Two systems with identical processors could throw different results due to motherboard differences.  Yes, architecture plays a role as well. Don't believe for a minute that buying a screaming video card will the be the end to all of your woes.

Memory now runs anywhere from 266 MHz to 433 MHz. That alone, is huge when using generic benchmarks to do cross comparisons. If you are comparing benchmarks from two different sites, first find out if the systems they were tested on are closely related.  Operating systems, hardware differences will give different results, as stated previously, drivers will also vary results widely.

Just be careful if you use benchmarks to consider what hardware you should buy. This is not to say that strict fundamentalism should be used, but look out for wild speculation from sites that are just happy to get a nice piece of hardware.

The system specs being used for this review comparison will be as follows:

Soyo DRAGON Ultra
333 MHz memory (166 bus speed x 2)
AMD XP 2000 (1.667 GHz Palomino architecture)
20 GB 7200 RPM Seagate Cheetah ATA133
60 GB 7200 RPM  Maxtor ATA133