Slashdot Mirror
Consumer Reports Stands By Its Verdict, Won't Recommend Apple's MacBook Pro (mashable.com)
Consumer Reports took many by surprise last week -- certainly Apple -- when it said it doesn't recommend the company's new MacBook Pro models. The American magazine, which has garnered credibility over 80 years of its existence, said battery life on Apple's new laptops was all over the place -- hitting 19 hours in a test, but less than four hours in another. Last week, Apple's VP of Marketing, Phil Schiller insisted that Consumer Reports' findings didn't match the company's field data, and that Apple was working with Consumer Reports to understand its review. Now Consumer Reports has responded: The nonprofit organization is standing by its initial verdict in which it did not give the MacBook Pro (2016) its "recommended" rating. The organization has now said it doesn't think re-running the tests will change anything. "In this case, we don't believe re-running the tests are warranted for several reasons. First, as we point out in our original article, experiencing very high battery life on MacBooks is not unusual for us -- in fact we had a model in our comparative tests that got 19 hours," it said. "Second, we confirmed our brightness with three different meters, so we feel confident in our findings using this equipment. Finally, we monitor our tests very closely. There is an entry logged every minute, so we know from these entries that the app worked correctly," it added.
The problem is not "over inflated battery life" - and actually, Apple has (in the past) gotten kudos for being one of the few companies that consistently provided reasonably accurate battery numbers for their products.
No, the issue is there's something as-yet-unexplained which, under some circumstances, causes the battery life of the newest MacBook Pros to plummet to ridiculously low levels. Consumer Reports saw it in their testing; but, even before that, some customers were experiencing it (and justifiably complaining).
#DeleteChrome
Sounds like they have stopped being "objective" and have moved on to "defensive".
HOWEVER, an interesting anecdote comes from reading another online forum (MacRumors.com), last evening, where a poster with a tbMBP 15" noted that, ONE TIME, when he unplugged an external Thunderbolt display (TB displays FORCE the MBP to use the dGPU), "Activity Monitor" said in the "Energy" tab that, instead of the 10 or 11 hours he was getting on average, it was showing that he was expected to get 3 hours.
However, no Processes were showing as being Energy-Hogs, and, he also stated that the "CPU" Tab showed that nothing was using over 1.5% CPU (which was reasonable for what he had running). And what he did have running SHOULD (and probably was) running on the iGPU. (???)
But, what was really "telling", was that he reported that the area under the "E" and "R" keys on the Keyboard was getting REALLY HOT. Hot enough that he panicked, and Rebooted the laptop.
Everything returned to normal, battery life report back to normal, no heating, hasn't happened since...
So, looking at the iFixit teardown of the 15" MBP, you can see in Step 6, that the components that would be under that area of the Keyboard would plainly be the AMD GPU (outlined in Yellow) (and not the CPU, which is over nearer to the "I" and "O" keys, basically).
So, something is (maybe) occasionally causing the AMD GPU, not the CPU, to run amok (or even be in some sort of power-guzzling "SCR-Lockup" state (hopefully not!)), sucking down the juice. Obviously, CR and others haven't triggered this behavior in the same way as the MacRumors poster; but there may be more software paths to this bug, likely involving switching between dGPU and iGPU modes, and/or power-savings involving same.
More than likely this is still a software issue; but it is not one that Users can see in Activity Monitor (other than it does seem to "know" that the battery is being drained by something, hence the low "Time Remaining" number). Apparently, Activity Monitor doesn't report separately on GPU Energy usage (they need to change that!)
Just an interesting little tidbit, that belies the assertion that a "retest" wouldn't make a difference (after Apple has a chance to address this issue, of course).
CR does not accept vendor payments, nor does it accept advertising.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
Is that Apple had no interest in actually sending the logs and test data to their engineers to figure out what went wrong and develop a solution. Instead, they wanted to solve the issue with PR: insist that CR somehow ran incorrect or non-stringent testing, have them re-run the tests according to how Apple wants them to be run, and have them revise their recommendation. Obviously I'm extrapolating a bit here, but it feels consistent with Apple's action up till now. Not to mention they put their head of marketing on the case, not any actual engineers. Good on Consumer Reports for sticking to their standards instead of caving to pressure.
I've been waiting since 2011 to upgrade but every model they put out has been more and more retarded. Soldered memory. Proprietary storage. Removing ports even when it destroys compatibility between the few products in their own meagre lineup.
Apple has basically abandoned the professional market, and are now exclusively catering to rich people who sit all day in Starbucks looking at Facebook. I can think of no other reason for their direction in the last few years.
It's not CR's job to find out why the product isn't working as expected/advertised. Their only job is to test the system in controlled and repeatable ways that can be demonstrated and are consistent with current quality assurance methods, and then to report on those tests to their paying subscribers. CR does not take money from anyone but their subscribers and buys off-the-shelf/lot products in order to ensure that there is no appearance of impropriety.
In this case, they were comfortable enough with their results, even after Apple contacted them, to keep them. If they felt that the consistency was in issue with the tests (the same tests/test-systems that are run/used on other computer systems) then they would have stated that and reworked the tests. They have done this in the past when their tests were not working as expected.
Arstechnica's test shows similar results to Consumer Reports. Stop parroting Apple's head PR guy.
Under a very basic web browsing test, which the author admits that "this test is probably actually too light", the Mac book pro got good battery life.
https://cdn.arstechnica.net/wp-content/uploads/2016/11/charts.010.png
But under a heavier test using webgl, the Mac book pro got 2hrs 13m
https://cdn.arstechnica.net/wp-content/uploads/2016/11/charts.011.png
Look in a copy of CR magazine. Notice what's missing? Advertisements. Same with the website.
Also Consumer's Union is a non-profit which publishes its financial statements. The income statement is particularly simple in that operating revenue comes from the following sources: subscriptions, newsstand sales, tax-deductable grants, and interest.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
You're behind the times. The newest Macbook Pros have their NAND storage soldered to the mainboard.
The previous iteration of their proprietary SSD had encrypted communications. It took OWC over a year to reverse-engineer it and offer compatible SSD upgrades. I guess Apple took that as a sign that they needed to eliminate any possibility of a third party upgrade. After all, you can't have customers modifying their hardware to their liking.
What exactly does Consumer Reports have to lose by a re-test?
Time. If they start giving some companies special attention, then everyone else will start demanding that, too. If, on the other hand, they stick to the "We give you one chance and that's that" they can actually get a lot more work done.
Oh FFS...
Since this thread is full of fanboys rationalizing Apple's failures, I think I'll eat their mod points by recounting my personal experiences with their failures.
I bought a 2007 MBP. It's battery swelled and had to be replaced. Eventually, it's 3d graphics card died and the only way to use it was to boot into safe mode.
I bought a 2012 MBP. It's trackpad quit working and had to be replaced. The replacement trackpad also failed within a month, but by then it was out of warranty. I quit trying to get it fixed because I use a mouse anyway, and I'm sure those cunts would try to charge me because I didn't buy "Apple Care".
I was given a 2015 MBP. So far it hasn't failed, but it has behavior that is intolerable. With the lid closed, it goes to sleep unless there is a keyboard plugged in. Apple says "Fuck you, software KVM users". And even with a keyboard plugged in, it immediately goes to sleep if the power cord is yanked out. Apple says, "Fuck you, cat owners".
I have no interest in their new crippled laptop and its gimmicky function key overlay. That shit was lame when it was called the Optimus Maximus in 2008 and it is just as lame now. Apple says, "But muh innovation! Muh courage!"
My first laptop, a ThinkPad from 1998, still works and boots to a 2.4 kernel. (Many nostalgia, such rugged, wow.) My other Toshiba, Dell, and HP laptops also worked up until I got rid of them, and they all took way more abuse than my precious, delicate MPBs.
So this year, I bought a cheap laptop from Dell. I'm using Linux again for the first time in a decade, and it is liberating. Buh-bye Apple, you prissy, shark jumping freaks. I can't wait until I retire and never have to touch your shit again.
First of all, no, they are under no obligation. They would still likely do it since it would be in the best interest of their subscribers and of their reputation.
Furthermore, there is no official fix from Apple yet. As far as I know there are rumors of some beta version faring better, but nothing more. CR didn't say they won't re-test the devices if/when Apple releases a fix, they won't re-test the devices *as they are* since they are confident of their previous test's methodology and findings.
Note that CR shared the diagnostic files from their tests with Apple and will definitely re-test the devices as soon as Apple claims they figured out the battery problems and fixed them. Until then they trust their results and find a re-test pointless.
So no, there is nothing suspicious on the part of CR.