DDR500 Fakeries

Remember Anandtech reviews of DDR500? You know what is the problem there? Anandtech reviewers got their memory from manufacturers so the reviews are doctored. Here is the real stuff, the stuff you buy:
http://www.legitreviews.com/Reviews/ddr500_1.shtml
Notice how much less they overclock than on Anandtech review.

Good reading! Thank you.

Yes, when manufacturers provide the product for review, there is always a chance that they pre-tested the product to issure it was top quality. That does not necessarily mean the review is invalid. You just have to take the results with a grain of salt.

That's why I read lots of reviews and look at trends. The trouble with any review is that it's not scientific. You really need to test say 100 RAM modules for each brand to give you meaningful statistical data. I think it's a little harsh to say the Anandtech article was a fake and that the review was doctored. What proof do you have of this claim? It just means the data "might" be skewed a little.

Regardless, the article that you posted the link for was quite interesting. It's also nice to see that my top two picks of RAM that I was thinking of buying scored the best... that being the Corsair XMS PC4000 and the OCZ PC4000 EL (I had already picked those two from reading several other reviews). By coincidence, both Corsair and OCZ use the same chips and are priced about the same too.

Anyway PC4000 mem are rather supposed to run on high voltages. All mems tested would probably have far less errors on testing if run on 3V.

yea but why doesnt the el go over 2.9? that seems really weird i would think to put like 3.2v in those puppies

You would think that given the money spent on the sticks and the card that they would at least explore the voltage issue, disappointing. Still a good effort and I imagine with enough feedback they will probably go the distance and see if some of them sticks can go with some volts. I know that I would much rather spend $200 less for mem if i know that a vdimm mod will get me where I want to go.

Another interesting thing, and a disappointment is the failure to include the Mushkin 4000 module. I would like to see how it stacked up against the competition.

I think it's a little harsh to say the Anandtech article was a fake and that the review was doctored. What proof do you have of this claim?

Click to expand...

Common sense.
Read the reviews and compare the overclocking. I am only blaming Anandtech for being cheap and using provided samples. You will need a little more than a grain of salt, more like a whole table spoon.

I also have objection to a site using provided for samples, I don't think it rises to the level of deception, as they always report that what you buy may not be what we have, but it is irresponsible in that they KNOW that what they have is better than what the consumer will get, and yet they still report tests that have no validity in real application. The purpose of the review should be to provide the end user the best information they can get to make their buying decision. The fact is that the reviews are worthless, and really only reflect the best that the company can do. Sort of like comparing a race car sponsored by a car manufacturer, to the stock version sold oon the streets. Very irresponsible and goes to the credibility of the site.

Can I rely on reviews from Anandtech? Hell no, and therefore I will not use them in the future for purchases. Legitreviews, and OCTools will hopefully start a trend in looking to see whether the retial versions are the real deal, and maybe we will see less posts regarding issues with hardware.

On another note, the industry itself has been subject to misrepresentating their products for a long time. The review sites are the best way to expose the hype and they have taken on a certain responsiblity to report accurate information.

So I don't think the initial post was too harsh at all, and Anandtech, and the other sites better take note.

I also have objection to a site using provided for samples, I don't think it rises to the level of deception,

Click to expand...

???
Of course it does!
It is claimed to be a particular memory evaluated and that is not the memory you'll buy. What else than deception?
By manufacturer and assisted by dummies and cheapos from the site that does evaluation.

Ordinary manufacturer would be sued for such a thing no matter what they put in the "fine print," it happened to many, like Publisher's Clearinghouse for example. They had the "fine print" but it is still nevertheless deceptive.

Every single module of memory won't perform the same no matter if it's a manufacturer-sample or a store-bought module. So every single memory review has to be taken with a grain of salt and even LegitReviews data can be off the mainline.

The best memory review I've read have been from Austrailia Overclockers. They make it a point to purchase several (10+) modules and test each, averaging the data. It still doesn't matter, though. Memory chips and module construction methods don't bring about the best in consistancy. But the biggest inconsistancy is the fact that every one's computer systems differ, and just bringing it all together causes widely varying results.

good read.. all reviews should be like this with non cherry picked samples ;d

A cool idea is instead of companies shipping samples for review.. they should allow the reviewer to buy a sample of the product being reviewed from any online retailer and then the company should pick up the tabs on the purchase so the item can be reviewed with the least amount of bias ;d but we all know that aint gonna happen