We've taken issue with AMD's attempt to shape reviews of its desktop Trinity APUs by dictating which test results independent labs can publish when. Below is a statement we received from AMD explaining its rationale for creating a multi-stage information embargo for reviewers:
Allowing media outlets to post 'limited previews' prior to the product launch date (and news embargo lift) is not a new practice. AMD and our competitors have used previews for years (if not decades) to give our customers and fans an early 'sneak peek' at some of the new features and technologies in upcoming products.
For the second generation AMD A-Series desktop APU launch, we thought it would be reasonable and fair to let all reviewers run previews, in all regions around the world, rather than selecting a handful of outlets. As for the preview rules, we allowed press to run gaming benchmarks (any game, whether we win or lose), run power tests (including tests we both win and lose in), as well as run experiential tests (again, whether we win or lose), where the reviewer can discuss his impressions of the product based on the quality of the experience in any application they see fit.
The goal was to provide an opportunity to talk about the real-world experience with the product and also highlight the performance in key applications where we are targeting the product. Our latest APU is designed for gaming and entertainment enthusiasts, and we wanted to make sure that the design decisions we made to improve the experience for these users were clearly discussed by the global review press.
Many media feel this approach allows an unprecedented level of freedom to previewers; it includes more press, it includes more readers, it includes more countries and it includes more hard news. Of course, there is no requirement that press participate in this or any other preview; they are free to wait for the full embargo lift date.
As we said before, what AMD is doing here is quasi-clever, taking an existing form—the "limited product preview"—hollowing it out, and using it in an innovative fashion in order to achieve a previously unheard of measure of control over reviewers. In the past, "previews" have been short and breezy, and they've traditionally included a very small amount of test data, if any. Quite often, those results are provided directly by the product's manufacturer and are understood to be preliminary.
By contrast, well, look at the articles out there today. They sure look like reviews, and some are labeled as such.
AMD has "allowed" wide leeway in its first-stage release of information to report nearly all product information and a broad set of independently obtained test data in areas where its product is strong, such as integrated graphics. Yet reporting the independently obtained results of CPU-focused benchmarks must wait until the second stage, when the initial buzz surrounding the product has worn off.
By exercising this sort of control over the release of independent test data, AMD gets to highlight its product's strengths and downplay its weaknesses. Or, as the firm put it, "The goal was to provide an opportunity to talk about the real-world experience with the product and also highlight the performance in key applications where we are targeting the product." At least they're being upfront about it.
Trouble is, companies whose products are being reviewed do not and should not get this sort of control over the release of independent review data. Reviews are not advertising and they are not marketing vehicles; they are independent evaluations of products that should keep the consumer's interests paramount. Blurring the fine line between "preview" and "review" may create confusion in some quarters, but when the dust settles, the principle at stake isn't difficult to spot: it's editorial independence.
For those who see little harm in AMD's practices in this particular instance, I get it. It's just one product release, and not perhaps the most consequential one in your book. But the principle at stake is paramount to the continued production of reviews you can trust, and that is a much larger issue than this one incident. If you're an AMD fan, perhaps this case doesn't bother you, but the next one, where Intel or Nvidia or whoever takes a similar approach, may not sit so well. Whatever your allegiances, your interests as a consumer would be better served by a truly independent review press.
|Gigabyte SA-SBCAP3350 puts formidable power on a single board||9|
|Alphacool Eisblock HDX-2 and HDX-3 help M.2 SSDs beat the heat||8|
|Corsair Lighting Pro Expansion Kit lets builders turn up the lights||8|
|Adata D16750 power bank is tougher than the average juice pack||15|
|Deals of the week: fast memory, an AM4 motherboard, and more||14|
|Corsair RMx White Series PSUs take a walk on the snowy side||24|
|Intel crams 100 GFLOPS of neural-net inferencing onto a USB stick||41|
|Toshiba's XG5 1TB NVMe SSD reviewed||9|
|Microsoft and Johnson Controls put Cortana in a thermostat||25|