Quote:
Originally Posted by EricVanWyk
I'd like to take this opportunity to remind teams that the development engineers absolutely must have all failure mode data available in order to do their jobs well.
Scott quoted a 0.2% failure rate, which is acceptable by my standards considering the RMA policy. What he can't quote or analyze is unreported failures. Each unreported failure represents a disconnect in our community, because the people who create and the people who use see two different data sets.
For example, it is possible that 50% of my analog breakouts are going up in smoke, but teams aren't reporting it because they simply use the other one. If this is the case (unlikely), we need to know it so we can fix it.
|
Quote:
Originally Posted by Dave Flowerday
It's dangerous to look at this issue from a purely statistical point of view. First of all, there's a very, very good chance that the majority of those 13,000 Jaguars shipped have not yet been used. Many teams do not have a robot up and running for practice in week 4 yet. Many other teams (like 1114, apparently) are avoiding them in the first place. When I hear teams like 71, 188, 11, and others all rattle off the same problem, I can see that there's an issue. It would be very foolish for FIRST or LM to ignore the anecdotal evidence that is cropping up, as while the numbers now are small, it very well could be an early indicator of a larger issue.
{edit} I should say I don't have any reason to believe that FIRST and/or LM is ignoring this issue... but I do get concerned when I hear statements to the effect of "The failure rate is acceptable" since that suggests that it might not be taken seriously. {/edit}
|
Quote:
Originally Posted by utlinebacker
My purpose of actively posting is to attempt to retrieve useful information from those "anecdotally" (to use your parlance  ) reporting failures on this forum... as the small percentage of returns AND the fact that we cannot reproduce any failures even after repeated vicious testing is not getting us any closer to understanding a root cause.
Our engineers are literally begging me for more information from you guys (again, see my previous post on questions we are interested in), and I hope that anyone who has worked with us feels that we are quite responsive and easy to work with. Please help us to help the community. support@luminarymicro.com
Regards,
Scott |
Guys, Please ... A vendor (Luminary Micro) has asked us (The
FIRST teams) to assist in collecting data so that they may resolve a potential issue with their product (Jaguars).
We, in
FIRST, have a massive engineering group that should be able to help out, so that those within our group that do not have access to this information will be protected.
Please, if you have a failure, let LM know. Give them all the information they've asked for, and more. Maybe your bit of 'anecdotal' information gives them the key to fixing whatever may be wrong with the Jaguars. We need to do this now so that week 1 will not have a bunch of rookies with robots that don't run ... and they will not know why.
This is what GP is all about. Helping without being asked ... because it is right to do so.