947 words on Apple defects
Just as I stopped thinking about failure possibilities for the MacBook, yet another failure – if not EPIC failure – in the design of both the hardware and the software ‘crafted’ to run on it revealed itself. This time it’s about temperature.
Let me just recap what I consider good design for a consumer machine like the MacBook: it’s reasonably powerful, it’s quiet and – being a laptop – it doesn’t burn my laps. If the technology Apple use is too inferior to achieve that, I’d still expect the machine to be noisy and hot only in extreme usage situations. Ripping and re-encoding a DVD would be one – where loads of mechanical and CPU resources are used. Playing modern games which stress both the CPU and graphics hardware might be another such situation.
Unfortunately the MacBook’h history is one of poor thermal design. In loads of the first generation machines temperature failure was built-in and they just shut down when heating up too much (while I had a load of defects my first MacBook, I amusingly didn’t have that one – probably because it was too common). With some firmware update Apple reprogrammed the fan of the machines to run at a higher level – I suspect to reduce the number of times that failure showed up, i.e. to reduce their financial cost at the expense of their users’ nerves. When I got my second and current MacBook it was clear that Apple had completely given up on a quiet design. It seems that the fan is always running, and it only takes a short burst of computation to make it clearly audible. I sighed.
For the past months, though, I had the impression that the fan’s noise was becoming stronger and stronger. I was literally hoping for the construction site around the corner to be noise, so I didn’t need to hear the fan while using the machine. I just blamed that on ever cheaper and lower quality components being used for the machine that can’t take a year of usage.
Then I had a shutdown of the machine. I didn’t think much of it at the time, blaming my commong bad luck or immature software instead. But as I saw such shutdowns more frequently and each time in situations where the machine was running hot – when both the CPU and GPU were used, I started being concerned. In some of these situations I was using the machine on a mattress. The vents weren’t blocked but it certainly became a bit hotter that way. In others I wasn’t, the problem just happens more easily on a mattress – and probably even more easily on a cushion.
But to be honest I don’t give a damn. I bought a portable machine, one that can presumably be used pretty much anywhere. In particular it wasn’t advertised as ‘usable on straight table-tops only’ and I’m sure it’s not meant to be usable in that way only as that’d be missing the point. And besides me not giving a damn, the machine can easily be led to exhibit the same problem on the straight and solid surface commonly known as a desk. I’m sure that – however they spin it – that’ll be considered an ‘adequate’ way of using the machine.
So we have an overheating situation: Using the numbers given by iStat Pro (which don’t necessarily represent the correct temperatures but are as good as things get), my machine runs just below 70°C CPU temperature when idle, a bit above 70°C when used a bit, and heads for 80°C when there is a high load. Adding some more stress drives the temperature up to 83°C, 85°C or even 87°C. Somewhere between those numbers the machine usually just shuts down. I can fairly easily reproduce that by running
yes along with a Quartz Composition, the machine just needs to heat up itself and the desk it is on for a while.
Yeah, that’s right. Instead of mimicking the behaviour when the battery runs down where you are first warned about a critical situation coming close and the machine is then sent to sleep carefully, Apple carefully ‘crafted’ their hard- and software to simply shut down when things get too hot (thus making all your unsaved data go AWOL, and triggering all the additional nuisances of making sure that your file systems, indices and backups are in a good state). Of course completely shutting down can be a reasonable course of action when the machine heats up by surprise. Say, because you put it on a stove or because the battery just went up in flames. But if the temperature is rising step by step while there is a high load on the system and the FAIL is completely predictable to a human onlooker, that’s not exactly what I’d call a surprise, it’s simply not caring about the users’ data, i.e. not caring for the only reason of the machine’s existence.
I’ll probably learn what is causing the problem tomorrow. To me it sounds a bit like the ventilation isn’t working correctly (to be honest there has never been a direct stream of air coming out of the machine’s back no matter how loud the fan was. Perhaps it’s just blocked. In which case – I was told – it’s not a warranty repair to ‘fix’ it. Because Apple didn’t ship the machine with the blocking stuff in place. They just shipped a faulty design that lets stuff get in and not out again while making it not user cleanable, I suppose… or it’ll be a new mainboard once more. It’s still the first one in this machine.
My stats. Those numbers you’re seeing are highly irregular. I am interesting in seeing what the problem is in your case.
Another case of Apple shitty hardware flaws, or perhaps you’ve pushed that machine to its limit. I wonder if they even check to see if there is a limit, or do they just push out the products without any “real world” testing.
Thanks for the comparison numbers. I took the Book to the shop and they managed to reproduce the problem. It took them 6 hours, though (I manage in 20 minutes if I try…). Tomorrow they should be able to tell me what’s going on there and how it’ll be fixed.
[I keep hoping for a ridiculous aswer à la ‘that t-shirt you just couldn’t find got stuck in there’…]
Every once in a while my MacBook Pro will get “stuck” after a time of high activity, and remain in “fans at maximum mode” for all eternity (even if I shut down every app running and do nothing). It’s only after I reboot that things go back to normal. It’s rare, but it’s very frustrating when it happens.
Sounds like your fans are cluttered with dust. A rather typical problem with many laptops, used as desktop replacement, after a while. Had it first on a Vaio, then on a Dell… and plan on changing the MacBook before it happens (usually some time between 18 and 24 months in my case).
That sounds quite odd as well but at least it doesn’t turn the machine off and destroy your work. (Even sleeping the machine keeps the fans on?)
If dust in the fans is a problem (it wasn’t with my Titanium Powerbook and none of the machines I had afterwards was recommended for user-opening, so I didn’t try), I’d say the machine should be designed for easy cleaning of the fans. Just like it’s done in a vaccum cleaner or tumble dryer.
I’m having the exact same problem with my Macbook - with almost identical numbers. PLEASE PLEASE PLEASE let me know what the outcome is - I would love to know if this is a warranty repair, or something I need to tear apart and fix myself (and the cause of it!).
Sorry I can’t help you. I never found out myself. My dealer said they didn’t find anything, so they didn’t fix it. I haven’t had the problem again since, so I don’t know.
The idea of disassembling the machine was too scary for me. I’m still wondering whether it could just be dirt (dust, dead mosquitoes – they just love the machine’s heat) inside the machine, which apparently wouldn’t be covered by warranty.
My problem was solved by installing the Quicktime update and rebooting. I’ve heard similar reports all over the place.
Received data seems to be invalid. The wanted file does probably not exist or the guys at last.fm changed something.