Cheating on Tests

Eurosoft Test Bulletin: Opinion/Editorial: Cheating on Tests

This month we discuss a controversial issue that has been in the news lately; cheating on tests. More specifically, the Volkswagen emission scandal and how this is similar to using freeware or perhaps even borrowed tests to support the computers you are responsible for.

Is it possible that vendor diagnostics for computer components are intentionally engineered to pass? Or rather that they are designed to test components only so far?

Detractors of these questions may ask why vendors would intentionally design weaker standards and tests for their components. One motivator could be to keep returns outside warranty claims.

To be clear, no one is accusing vendors of intentionally rigging tests to pass. Nor are we suggesting that reputable vendors want to offer anything but the best possible products.

But, the recent Volkswagen news reminds us that sometimes there is a very short distance between paranoia and vigilance.

What would happen if mediocre testing products were used to service the customers you have worked hard to get? What do computer manufacturers, servicers, and refurbishers do when weak testing products are used?

This issue may go away quickly, and neatly, at first…but only temporarily. After that, all kinds of exponential problems arise. Margins are soaked up by wasted field repair visits, product recalls, and loss of repeat business.

With this vicious circle in mind, it seems that vendors would engineer their diagnostics to fully and completely test every aspect of a component.

The one nagging factor that vexes this point is the fact that not all customers follow up on their warranties. Just like 41% of people forget to send in rebate coupons¹, there must be a percentage of customers who…

• Are not aware of their warranty coverage.
• Switch to a different/new component rather than replacing the failed component.
• Are in a rush to bring their PC back up and are unwilling to wait for warranty fulfillment.

There may be more reasons than this…can you think of any we missed?

Do vendor diagnostics constitute a conflict of interest? And if so, then wouldn’t using multiple, different vendor diagnostics versus a standard, consistent diagnostic create multiple conflicts of interest to manage?

We already know that manufacturer-specific diagnostics are often designed to support only specific component models. This alone can cause complications in your process by creating too many options and potentially spurious as well as false-positive test results from using the wrong tests or vendor specific diagnostics on various components.

Would it be better to use a single, consistent diagnostic with higher standards, broad non-brand component coverage, and standard outputs?

Give us your thoughts.

¹ (Masnick, Mike, 41% of People Forget to Send in Rebate Coupons, TechDirt.com, Feb 2^nd 2005, https://www.techdirt.com/articles/20050202/1311201.shtml)