Letter to the Editor
Copyright © 1997 to 7am News
Date: Thu, 24 Apr 1997 15:17:29 +1200|
From: Keith Newman <firstname.lastname@example.org>
Organization: PC Magazine
Subject: PC Magazine says get the facts right
Thursday, April 24, 1997
As regards the PC Direct letter which is now published at Aardvark please find a point by point rebuttal of the issues raised by PC Direct. The issues needed response and did not meet our May deadline. We have not refused to publish but rather given an undertaking we will publish in June subject to errors being challenged.
As regards the core of the issue, benchmarking there are some discrepancies which need to be addressed. Martin Taylor says "raw performance, as PC World's published methodology shows, accounts for only about a fifth of the total score given". In the US PC World, the actual figure is 25%, or a fourth (I don't know what the PCW NZ one is, but neither does Taylor, it seems). He also says that PCW NZ factors in warranty terms, service and support, reliability, price and features into the total score. In the US PCW, reliability accounts for 25% of the total score, features another 20%, price 15%, and support 15%. However, PC Direct complain that neither PCW nor PC Magazine conducts reliability and support surveys, so it appears that PCW has no method of estimating these two factors with any sort of accuracy. In other words, 40% of the total PCW review score is assigned... how? Arbitrarily? At least PC Magazine does not write about reliability and support without having the material to support what is written.
Taylor also makes a big noise about PC Magazine using "raw performance" as its only yardstick. This is nonsense. We factor in price, features and warranties as well. Further, performance isn't as unimportant as Taylor makes out. People want fast computers, end of story. If they didn't, Intel wouldn't be upping the CPU ante so often, for example. However, a fast computer is no good if it isn't at the same time well-designed and reliable. We can give an assessment as to the former, but not the latter, at least not without a comprehensive NZ user survey. If PCW NZ says a machine is reliable and has good support without a user survey, the magazine may be misleading its readers. Note that the US version of PCW conducts readership surveys to assess the quality of a vendor's support and the reliability of its machines.
The ZD benchmarks are also useful as a general compatibility test, as you can see from vendor feedback. They use Winstone and Winbench as part of their quality control, as the benchmarks test just about every component in a PC. Even PC Direct prefers to use ZD's benchmarks in their ads over PC World's. Microsoft has only accredited the ZD tests, none others. PC World does not make its PCWorldBench public. Why?
Martin Taylor should also be very careful when he publicly states PC Magazine sells the Editors' Choice award to the highest bidder.
Even if PC Magazine did contact Dell to say that it is running an already published feature, in which it won Editors' Choice, and asking Dell if it would like to buy some ad space, there's no way this can be seen as corrupt. The machine from Dell had already received Editors' Choice in the US. In fact Dell did not place an advertisement as claimed - this was a PC Magazine "competition" totally unrelated to advertising as alleged in both Mr Taylor's correspondence and the commentary by Bruce Simpson in Aardvark.
Mr Buckman the managing editor of PC World is correct when he points out the error in the pricing of the Dell review machine. We published the price for a standard configuration not the configuration reviewed. You will find a letter from Dell marketing manager John Bessey in the upcoming issue apologising for giving us the wrong pricing. However Dell is very pleased with the high response it has had from our readers following that review.
Back to Aardvark Daily...