New MacBook Pros Fail to Earn Consumer Reports Recommendation | Hacker News Reader

New MacBook Pros Fail to Earn Consumer Reports Recommendation

consumerreports.org

50 pointsQUFB7 comments

the touch bar is simply garbage. In 2 weeks of usage including heavy xcode use I have yet to find a single instance where having the touch bar was better than having the function keys.

Not only that I'm pretty sure the battery issues are due to not being able to turn off a constantly updating touch bar.

It may not be the hardware; it may be the the OS. Since upgrading to Sierra, I've been plagued by various runaway kernel processes randomly eating 100% of the CPU. Have finally managed to kill most of them through deleting .plists and the like -- all for native MacOS apps like Contacts by the way, not in any 3rd-party apps -- but this kind of bugginess could easily produce the kind of results that Consumer Reports saw.

"we experimented by conducting the same battery tests using a Chrome browser, rather than Safari. For this exercise, we ran two trials on each of the laptops, and found battery life to be consistently high on all six runs"

Specifically it looks like it might be a Safari problem.

Mind to share the list of .plists that you have deleted and disabled?

Oh gods, I didn't keep a list. Most had no effect, but I eventually traced it to Contacts, and with the exception of a couple of even more mysterious rogue issues with the windowserver, that seems to have fixed it.

I cannot consolidate their results with my own experience and those of many reports I've read: their results seems way too good! Real world annecdata: a google hangout seems drained the battery in about two hours on my top-spec, one-month old, MBP. I never had experiences like this on the 2013 MBP.

Google Hangouts consistently drains at minimum of 20% of my battery life per 30min meeting. It doesn't matter if I'm using my top spec previous generation 15", my top spec previous gen 13", or my 2016 12" MacBook. As such, using it to gauge battery life is about as fair as running a Hadoop cluster in VMs to gauge battery life. Actually, I think the active Hadoop cluster would use less battery.

The reason why their test results are ridiculously long is how little they stress the machine. Loading ten basic web pages (in the stock browser, so no Flash, and something tells me no WebGL usage either), so outside an errant background process, what's really using any watts there besides the screen?

Reply on news.ycombinator.com