On 09/04/97 08:33:11 you wrote:
Mistakes do happen. I've gotten motherboards with bad onboard ethernet, slots that didn't work, etc. But by working with Netapp these are put into the corrective feedback loop and prevented from happening again (at least in theory).
But didn't someone _just_ say that the parts you get are tested for reliability...
Yes (it was me). Of course, that process isn't perfect...
I'm sure that was speculation on the part of the speaker--but I guess this just goes to show you that they are _NOT_ testing their product before shipping it. We've sent numerous motherboards back.
One would would be curious to know the circumstances. Perhaps RMAs are not tested the same way as shipped product, or perhaps shipped product breaks during shipping (DOA). Neither of these negate what I said about the testing (which does occur normally), nor does it negate the assertion that these problems are resolved through corrective action. I agree, though, that as the quality of the testing process degrades, the trade-off for premium prices diminishes.
Keep in mind also that my point of testing was not that it prevents normal breakage, but that it ensures compatibility and realibility of the working part. Disk drive failures, memory simm failures (due to slighting off timing), and compatibility of difference card revisions (i.e. the card you buy off the shelf from DEC may not be the same as the one in your Netapp, even if it looks the same and seems like it should work).
Bruce