The dishonesty of overclocking...

[kmrivers]

I don't know if you guys have read this: http://scientiasblog.blogspot....y-of-overclocking.html

I found it interesting, but overall I didn't agree with it at all. Consumers don't come to anandtech. If you read the first paragraph of any review and don't understand, you quickly will know you don't belong here.

I don't see how sites "blurring the line of what consumer products are able to get off the shelf and adding overclocking into it WITHOUT warning the consumer of the drawbacks... and leading to completely different conclusion."

Not to mention, most sites give you stock results as well, then they share the overclockability of the chip in question. Honestly I thought the site was an AMD fanboy site, especially if you read the comments. Where someone posted benchmarks showing the Athlon spanking some other chips.

Anyhow, CNET, Consumer Reports, etc. Those are the placed REAL consumers go to for what to buy. They aren't reading technical reviews on the processor they are going to possibly have in their Dell.

What do you think?

 

[RichUK]

All your overclocks are belong to us.

 

[Anubis]

most consumers arnt reading review sites at all, they simply go i have xxxx$ to spend and then find a Dell/HP/whatever that fits their budget and purchase it,

 

[SampSon]

The article is a load of garbage with a seemlingly AMD/"underdog" bias. For example, go to his main blog page for today and look at the AMD fellating post he made.
When they started comparing cars with computer chips I just started skimming the rest.

Most CPU articles I've read on here and other sites give you equal setups with stock and overclocked results. The articles are written by tech heads, not "tech-saavy" consumer advocates. This article wants techie sites to be like Consumer Reports, when the whole idea is that they aren't supposed to be that.

The idea that a run of the mill consumer is going to be shopping for a specific chip, read up on these sites, then make a decision from biased information is ridiculous.

 

[SlitheryDee]

Anandtech's readership is a mix of enthusiasts and average consumers. Should they ignore the enthusiasts because the average consumers might try something they aren't familiar with? They're not here to baby us IMO. If you want to try overclocking, then do it and accept the consequences. If the consequences are potentially too great then don't do it.

I've always seen overclocking ability as a necessary part of any processor review. Even if I don't overclock at all I'd want to avoid a processor that's close to it's clockspeed ceiling straight from the factory.

 

[Conky]

Somebody posted that worthless blog a few months ago.

Everyone basically agreed it was written by an AMD fanboy who couldn't deal with the fact that the new Intel Conroe(new when the blog was written) overclocks very well. Without looking, I'm sure that same fanboy has a new blog claiming that Intel's quad-core is dishonest compared to the upcoming Barcelona. Whatever.

 

[kmrivers]

Exactly. His fallacy is in the beginning where he attempts to put magazines and "mainstream" review sites in the same category. Which is ridiculous. I think the problem is in the word "mainstream," it is so general one assumes "THE mainstream," but Anandtech and others are clearly mainstream in their target markets. Like, computer enthusiasts. He also ignores the fact that magazines serve niche markets, there is very rarely a magazine that covers every facet barring Consumer Reports.

I don't see anywhere that Anandtech claims unbiased reviewing. They are not obligated, just like a magazine about Honda cars isn't obligated to talk up Toyotas. If you think it is biased, don't go to it if you want an unbiased perspective.

And his most grand assumption, he is assuming consumers lack media competence. That they can't look at a source and determine whether it is credible and reliable. He thinks they will be swayed to Intel and of course they will overclock the snot out of their chips until they explode causing a great fire which engulfs the earth and from the Heaven he will say "We should have used AMD, I told you."
"

 

[kmrivers]

Speaking of that bloggers post today, is this math right?

"So to figure out how it might compare with Penryn we have, 1.4 X 2.66Ghz = 3.724Ghz for Clovertown. However, Intel has stated that Penryn is 9% faster than Clovertown. Therefore, 3.724Ghz / 1.09 = 3.42Ghz for Penryn. Interestingly, this is only 2.5% faster than Intel's expected 3.33Ghz speed for Penryn. Recalling that since 3% is the margin of error in testing, a 2.5% difference is not significant."

First off he says "with Penryn we have..." then after the numbers he ends with Clovertown. Those numbers being 3.724. Then he says, Intel states Penryn is 9% faster than Clovertown. He then divides 3.724 by 9%, bringing the speed down to 3.33, when it should be about 4.06 right?

If the first numbers are for Clovertown, and Penryn is 9% FASTER, wouldn't Penryn be higher than the Clovertown number. But even if the first calculation is Penryn and how the heck does he end that sentence with Penryn being less?

 

[Aikouka]

The guy's an idiot... following his idea, why do "car reviews" test the 0-60 times of cars? It's the same concept of testing how fast a processor will go... testing of the peak performance.

Sites like he mentioned may not always cater to the overclockers, but the point is that some people do overclock and enjoy it very much. These sites are not really designed for the people that just go out and buy PCs as just because a C2D is faster than a K8 doesn't mean that a C2D PC is faster than a K8 PC (i.e. most reviews are single-part/type reviews). These sites cater to enthusiasts and hobbyists, not to standard PC users. Car enthusiasts and hobbyists want to know more about their cars... they're the type of people that might do some minor upgrades or tuning.

EDIT: Why even ask about his bias... he's "Scientia of AMDZone".

Also, the only dishonesty you can get is overclocking a processor and publishing the numbers without notifying the users that the numbers are not stock (or trying to pass them off as stock).

 

[0roo0roo]

who the hell is that scientia?
the overclocking aspect of such articles are bonus information on top of the standard review. they never omit the regular results unless its an overclocking only article and frankly most or all of those too have stock results as baseline