Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: CNet always does the most useless things... (Read 4214 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

CNet always does the most useless things...

CNet never ceases to amaze me!! They always perform stupid benchmarks, post useless reviews that don't tell you anything and once in a while, do the ultimately ambitious things-the wrong way!!

Check this out to find out what I am mean.

CNet always does the most useless things...

Reply #1 –
Well, it does say that they didn't tell these 8 people what codec they were listening to, although they did tell them which was the reference sample (the WAV). They used iTunes to encode the MP3's, which is clearly silly albeit valid if you assume they were testing iTunes against Windows Media Player. It doesn't appear to have been double blind (they don't say), although if the guy doing the actual playing was isolated from the people doing the listening, it could still be.

But yes, on the whole, very silly, and their results show the inconsistencies pretty clearly. The lower bitrate files are often rated higher than the higher bitrate ones on the same samples. And all their numbers fall within about a 2 point sample range, roughly, so there's not a lot of variation there either.

In other words, a wholly useless test.

CNet always does the most useless things...

Reply #2 –
Quote
To calculate our results, the jurors were asked to rate each 20-second sound clip on a scale of one to ten, assuming the WAV file was a perfect ten. One decimal point was allowed if needed.


This scale is psychometrically problematic. Each person can attribute different meanings to the scale, what is undesirable on experiment. The scale used in ABCHR has only five points, and each point has a determinate meaning based on the characteristics of heard encoding artifacts. And, if the later is not always perfect (no psychometrical scale conveys perfect results), figure out the consequences of CNET's scale!

CNet always does the most useless things...

Reply #3 –
Quote
CNet never ceases to amaze me!! They always perform stupid benchmarks, post useless reviews that don't tell you anything and once in a while, do the ultimately ambitious things-the wrong way!!


No kidding  they aren't the only ones. I love reading useless reviews from sites like  Amazon too with the exception of the people who know what they are talking about. I use the blatent bashing of the Rio Kharma 24/7 as a example 
budding I.T professional

CNet always does the most useless things...

Reply #4 –
Quote
although it's odd we thought the 128Kbps file sounded significantly better than the higher-quality clip.


? ummmhhh? next time we should use 128kbps instead of 192kbps

CNet always does the most useless things...

Reply #5 –
Quote
No kidding   they aren't the only ones. I love reading useless reviews from sites like  Amazon too with the exception of the people who know what they are talking about. I use the blatent bashing of the Rio Kharma 24/7 as a example 
[a href="index.php?act=findpost&pid=301034"][{POST_SNAPBACK}][/a]


Yeah. I know what you are talking about (especially about the Karma since I own one and I know that all the reviews about it are BS).