Installation / Removal > Apple - Installation/Removal
iPod first generation install?
AlexP:
I assume the reason there was a check there in the first place is that there is a downside to applying it indiscriminately, and therefore we want a reliable way of checking between first and second gen iPods.
Llorean:
Just as we previously hadn't known there were 1Gs that didn't report their version the way we expected, what is the effect this will have if their are 2Gs with a similar problem or quirk?
So, "why not" is "we might be fixing the bug by simply shoving it off on other users we don't know about yet."
As for the "unsupported" nature of this build, we've truthfully had bug reports filed where the patch "should not" have affected the area the report was on, but did. It would be best if you tried to give it a week or two for an official fix for this problem, then verify any bugs you find in it, because the build environment can really cause some interesting quirks in builds, and patches (especially on portalplayer) can have some very interesting side effects at times, but if you come across something you feel absolutely confident is something we wrote, rather than something misbehaving unexpectedly, speaking up now may be best.
bidmead:
@BigBambi But this check (as described) isn't indiscriminate. It's not a kluge. AIUI it specifically doesn't apply the scrollwheel handling to 2G. If there are 2Gs reporting as 1G I can see there would be a problem. Similarly if there are 1Gs that need the scrollwheel handling but report as 2G. Or even if there's a logic bug in the RB generation reporting. This patch will expose those anomalies, and so be A Good Thing(tm).
@Llorean I take your point about knock-on effects, and certainly won't be jumping the gun with a shower of bug reports, for that and other reasons.
--
Chris
AlexP:
--- Quote from: bidmead on March 21, 2008, 04:41:29 PM ---@BigBambi But this check (as described) isn't indiscriminate. It's not a kluge. AIUI it specifically doesn't apply the scrollwheel handling to 2G. If there are 2Gs reporting as 1G I can see there would be a problem. Similarly if there are 1Gs that need the scrollwheel handling but report as 2G. Or even if there's a logic bug in the RB generation reporting. This patch will expose those anomalies, and so be A Good Thing(tm).
--
Chris
--- End quote ---
Right, but the point is that it is now applied to everything that doesn't report as a 2G (within a 1G2G ifdef) rather than everything that reports as a 1G, so if there are 2Gs that don't report as such (not just reporting as a 1G), or if we want to extend the code in the future to other similar players it could cause a problem. I agree that it isn't a completely horrendous hack (i.e. just enabling for everything), but a way of properly detecting 1Gs would be preferred, and that is what I believe the devs are searching for.
yapper:
--- Quote from: Llorean on March 21, 2008, 04:28:21 PM ---Just as we previously hadn't known there were 1Gs that didn't report their version the way we expected, what is the effect this will have if their are 2Gs with a similar problem or quirk?
--- End quote ---
I actually did some testing on a 2G, with a version of the patch that was set to trigger scroll wheel enable on anything that didn't return a version of '7', and there are no noticeable side effects. I believe one of amiconn's early versions of powersaving for the 1G2G also sent the powersave regardless of 1G or 2G (and he noted it was 'safe' as the GPIO output port used as the enable on the 1G is used as an input on the 2G for headphone detection).
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version