Bizarre Droid Auto-Focus Bug Revealed 275
Posted
by
timothy
from the each-droid-has-a-moth-enclosed dept.
from the each-droid-has-a-moth-enclosed dept.
itwbennett writes "Pity the poor engineer who had to find this one. One of the more interesting of the handful of bugs that have appeared since the launch of Verizon's Droid smartphone has to do with the on-board camera's auto-focus. Apparently it just didn't work. And then suddenly it did. Naturally, this off-again, on-again made the theories fly. But the real reason for the bug was revealed in a comment on an Engadget post by someone claiming to be Google engineer Dan Morrill: 'There's a rounding-error bug in the camera driver's autofocus routine (which uses a timestamp) that causes autofocus to behave poorly on a 24.5-day cycle,' said Morrill. 'That is, it'll work for 24.5 days, then have poor performance for 24.5 days, then work again. The 17th is the start of a new 'works correctly' cycle, so the devices will be fine for a while. A permanent fix is in the works.'"
A timestamping overflow error (Score:1, Insightful)
hoses up the camera autofocus?
Is this Slashdot, or The Daily WTF [thedailywtf.com]?
IIRC, this is the same sort of bug (Score:5, Insightful)
that caused Windows 95 to require a reboot in about the same timeframe.
Kudos to Google for being so open about the bug (Score:5, Insightful)
Re:iPhone 3G/3GS GPS bug (Score:3, Insightful)
GPS works by trilaturation [wikipedia.org], not triangulation... just sayin'
Re:So they tested it for less than 24.5 hours ... (Score:2, Insightful)
Re:When Signed/Unsigned Strikes (Score:5, Insightful)
Whenever I write something I think it particularly clever, I comment it out and write something simpler. The clever stuff I find is nearly impossible to figure out next year when you have to go back and add a feature or change something. It doesn't help that I usually think, "Oh that's so clever, there's no way I would forget how that works. It's so elegant." and don't bother to comment the hell out of it.
Simple == good
Re:Bigger bugs afoot... (Score:5, Insightful)
They should've just made it to lie about its policy enforcement to Exchange server like the iPhone did. That way it'd be banned from my corporate network like my iPhone was. Thanks Steve, you're such a smart guy.
Re:When Signed/Unsigned Strikes (Score:4, Insightful)
"causes autofocus to behave poorly on a 24.5-day cycle [...] That is, it'll work for 24.5 days, then have poor performance for 24.5 days, then work again"
Except if it works for 24.5 days, then doesn't work for 24.5 days, then works again, it's a 49 day cycle. Sorry to be pedantic, but someone abbreviating that statement might mislead someone else to believe they have a different problem when they see they have a 49 day, and not a 24.5 day, cycle of inoperability.
Re:When Signed/Unsigned Strikes (Score:3, Insightful)
This is what I hate about "smart" programmers at Google. They come up with the cleverest algo's, and they demand from their tests the ability to develop these algo's. Then they release something and along comes this sort of bug that simply does not need to happen!
It is a smart algorithm -- it's just a slightly poor implementation. As someone else mentions, it's probably a miscast to a signed 32-bit integer. The algorithm, on the other hand, is sound.
Your rant says more about you than it does about Google. You're intimidated by their ability to think of something elegant, and then do it right (unfortunately in this case only half the time).
4.3BSD had a bug like that (Score:5, Insightful)
The initial release of 4.3BSD had a bug like that. It wouldn't interoperate with implementations that chose TCP sequence numbers in the upper half of the 32-bit address space. BSD itself didn't do this until it had been up for 2^31 seconds, so it got through testing. Other implementations cycled faster. We were losing network connections for two hours out of every four.
It took a 1-line fix, after three days of looking at the generated machine code to figure out exactly how the sequence number arithmetic worked. Too many casts in the source.
Re:When Signed/Unsigned Strikes (Score:5, Insightful)
Obligatory quote:
"Debugging is twice as hard as writing the program, so if you write the program as cleverly as you can, by definition, you won't be clever enough to debug it. "
Re:New Moon? (Score:3, Insightful)
Of course. That's when the world was created, fully-formed. How do you know the world existed before Jan 1 1970? :-)
Re:When Signed/Unsigned Strikes (Score:1, Insightful)
The huge failure that both the interface designers of VI and of Clippy make, is that efficiency and easy usage would be mutually exclusive opposites.
This is actually a very interesting point, on the other hand you do realize that no one here is talking about UI design right?