Slashdot Mirror
Inside Apple's Anechoic Testing Chambers
As part of Apple's press conference on Friday, they mentioned their state-of-the-art testing facilities and released a brief video showing some of their anechoic chambers. They later invited journalists on a tour of the rooms and explained some of the experimentation process. Quoting:
"There are four stages. The first is a passive test to study the form factor of the device they want to create. The second stage is what Caballero calls the 'junk in the trunk' stage. Apple puts the wireless components inside of the form factor and puts them in these chambers. The third part involves studying the device in one of these chambers but with human or dummy subjects. And the fourth part is a field test, done in vans that drive around various cities monitoring the device's signal the entire time (both with real people and with dummies). ... The most interesting of these rooms was one that Caballero called 'Stargate.' Why? Because, well, it looks like it belongs in the movie/TV series Stargate. Inside this room, there's a giant ring that a human sits on a raised chair in the center of. This chair slowly rotates around as signals are passed around the entire outer circle. This creates a 360 degree test area. I was told this room is completely safe for humans. And people typically spend 40 minutes in there at a time for testing. By comparison, devices can stay in the other anechoic chambers for up to 24 hours at a time. ... We then went into a room that contained fake heads."
And nowhere do I read a description of the simulated conductive hands covering the antenna gap. Might they have failed to consider one key variable to test for?
I don't think it's a matter of saying "We're better than they are" as it is a matter of saying "before you accuse us of not testing, take a good look at our investment in testing facilities". Sure, the testing procedures may have been (probably were) flawed, but that's a separate issue from the rampant accusations of them not giving a shit.
A lot of oscilloscopes run Windows.
The linked techcrunch article sure does have some pretty pictures, but it just makes it that much more sad that Apple missed something with their million-dollar test chambers that any left-handed person will notice in a day or two.
the iPhone 5
Also for HATS testing (in fact Apple had advertised positions that required HATS-testing experience).
I'd say a lot of instrumentation runs on Windows only.
Sad but true.
A little known feature of iphone 4 is that if you sit in the 'stargate' and have a finger bridging the 'gap' of the antennae, you are able to cross into an alternate dimension that can only be described as 'insanely great'. Take my word for it.
Nearly everything about how Apple has handled this has been wrong. From their disingenuous attempt to rebrand the problem "Antennagate" to stop the media from calling it the "Death Grip", to their feigned surprise that the iPhone signal bar calculation was heavily weighted to make the iPhone look like a strong performer.
Now they're showing off how much testing the phone went through, which seems indicate they knew it was glitchy from the start. Or did they? I mean after all, in one of the first reviews of the iPhone 4 before it was even released, Walt Mossberg said:
So the very first review picked up on it, but they didn't have an explanation? They said they waited to have a press conference because they wanted to do testing to determine the problem, but doesn't that undermine the point that you've done adequate testing? Why after their press conference, is it still so unclear if they knew whether skin connecting the antennas was a problem or not?
The really bizarre thing is I've had an iPhone 4 since day 1, I've seen the glitch and until I got a case it had been affecting my data connections, but I still really like this phone! Is Apple turning us all into battered wives?
*sigh* The way the press has scented blood on this - there's nothing they like better than to tear down what they've built up. iPhones. Good machines, over-priced, sometimes innovative with a somewhat irritating and closed development model attached to them. Apple wasn't producing immaculate products from Heaven before, and they're not producing bricks from Satan's arse all of a sudden. Something got fucked up along the way this time, it'll get fixed. The hype and the derision in both directions is irritating to me. Maybe we can stop the rollercoaster and start treating Apple like any other company soon, please? I don't see constant stories about Nokia's phones (which are pretty nice, imho), for example.
Aide-toi, le Ciel t'aidera - Jeanne D'Arc.
many people do this!
I use my right hand for dialing, etc
but honestly, even right-handers hold the damn phone in their left hand most of the time...
In the anechoic chamber there is going to be one source of RF and there will be no reflections or other paths, only line of sight from antenna to antenna.
In the real world you are exposed to far more RF. From your cell phone, from the cell phone of everyone else in the neighborhood, from the microwave oven, from every monitor, cpu and everything else.
The real danger in an anechoic chamber is sanity. The non-reflective cones also absorb acoustics, which make the space a very strange aural experience, which can do funny things to your brain. For one you feel really, really alone, you can't even hear the echo of your own voice.
I'd be ironic if it did. But it doesn't.
> I don't think it's a matter of saying "We're better than they are" as it is a matter of saying "before you accuse us of not
> testing, take a good look at our investment in testing facilities".
I agree. They're trying to show what goes into this kind of testing. Engineers and technology people aren't going to be surprised by Apple's facilities (though it's cool to see the photos of the anechoic chambers), since other major mobile phone manufacturers will have similar facilities.
Apple's trying to show some of the ways that they control conditions while they're testing. Sitting in a Starbucks holding the phone in weird ways and watching the bars change isn't a good way to measure a problem since there is zero control of the fading conditions. The fact that they had a bug in their signal strength algorithm is bad, but one can't complain the problem happened because they weren't testing.
I think there's been a huge overreaction to the issue. However, what did Apple expect? One could argue there was a huge overreaction when the iPhone/iPad was announced (albeit, positive in those cases). This antenna thing just reminds Apple that the knife cuts both ways.
Insert simplistic political, ideological, or personal proselytization here.
Now, as much as Apple annoys me, and they do enough that I stopped using my iPhone and got an HTC desire, I do feel compelled to point something out to you folks, most of whom are not in the wireless industry.
Apple, and to that extent, all wireless manufactures must perform TRP and TIS testing as laid out in the CTIA Test Plan for Mobile Station Over the Air Performance, which I think are currently at 2.2.2.
The thing is, OTA testing takes a long time and is actually a lot of money.
Please note, that for certification, a company can NOT perform this testing on their own. They must use a PTCRB test house, which is independent for what should be obvious reasons.
As I mentioned, the CTIA test plan looks at both TRP (Total radiated power) and TIS (total isotropic sensitivity) under a few conditions, which are head adjacent(left and right cheek) and free space. This is done in all bands and all modes. That's to say you test the 850 band in GSM. GPRS, EGPRS and UMTS(3g). Each band is tested in full on three channels, the low, mid and highest of the band. Then a single point offset method is applied to all intermediate channels relative to the 3 primary channels in both position and power level to save time.
This still takes a LONG time.
A GSM 850 L/M/H TRP in free space takes about 1 hour in a non stargate system (note almost no labs use this system since it uses power meters which have trouble to properly trigger a EGPRS pulse)
about the same for the same conditions in TIS.
UMTS though takes about 4 hours for the TIS.
Now, you take a phone like the iPhone and account for charge times and the like and you are looking at about 3 - 4 weeks of lab time since you can only use 1 phone!
I also assume that would be lots of cash in lab time. Granted, that's crackers to Apple.
The point is, all phones on a PTCRB network, to witch ATT is, MUST pass these requirements. This means that Apple had to have passes ALL requirements.
They did was they were required to do. It just goes to show what you can't catch everything with this testing, but given that it's a rare problem..you can catch most.
How the fuck could it be for show? Were they not having a PR problem regarding antennas right now, the inside of this facility would have remained secret. They haven't managed to build this testing facility in Cupertino in the last two weeks.
It wouldn't be much of a certification process if companies did the certification testing for their own phones themselves.
I can't watch the video because it's in quicktime and I'm on a linux machine, but it is true that a lot of instrumentation runs windows only. I had the opportunity a few years back to visit the manufacturer of some scientific microscopes. I asked them why in the world they were using such an unstable and complicated platform as windows XP to run their software when what you really wanted was something that was dedicated to running a microscope. E.g., we used to comment on how back in the days of dos, the software for these microscopes was actually better, because dos had few (no?) abstraction layers to the hardware and the software had direct control over the vesa bus cards that controlled the microscope and running that software was the ONLY thing that computer was doing. In the days of NT and XP, software glitches and lag time (e.g. screen updates, etc.) have gotten worse and I think some of that is due to the fact that a modern operating system has a lot of things going on in the background that interrupt the microscope software.
Anyhow, I brought up this problem with the manufacturer and told him that something like linux might be better since it's easier to have a more fined-grained control over which processes are running under what conditions. Their response was sort of typical, the engineers knew about this already and even had an alpha quality version of the software that ran on linux. The managers, on the other hand, couldn't even pronounce linux correctly and didn't even understand the problem. They said that if enough users ask for it, they'll do it. I guess the users don't ask.
I have noticed that on some of the non-production machines, such as the software controlling instrumentation at synchotrons, the software is running on some form of unix. So there's hope, but I think we're stuck with windows until the general user actually sees the benefit of a dedicated instrumentation OS over a perhaps ill-fitting, but familiar, OS. For those of us forced to use mission critical windows software, we still have a lot of computers that are forbidden to be plugged into the internet since obviously if just the OS is getting in the way, AV software would get in the way as well. It makes the validating the MS Genuine Advantage a fun experience when you don't have cell reception in a basement lab (nor land line) and no internet connection.
Gentlemen! You can't fight in here, this is the war room!
As has been shown endless times while testing software, testing in controlled facilities often belies real life experiences.
You can lead a man with reason but you can't make him think.
I don't think it's a matter of saying "We're better than they are" as it is a matter of saying "before you accuse us of not testing, take a good look at our investment in testing facilities". Sure, the testing procedures may have been (probably were) flawed, but that's a separate issue from the rampant accusations of them not giving a shit.
Is it separate? If they have all these testing facilities and the testing procedure were in fact not flawed, then this problem is not caused by negligence but rather deliberate prioritization (i.e. time to market and/or development costs were more important). It other words, it would mean they really did not "give a shit".
I'm not certain boasting about their testing abilities is the rhetorically smartest thing to do at this moment.
“This lab used to be secret. Most people don’t know it exists,” Caballero told us. Dubbed the “Black Labs,” when I asked about the black cloaks, Caballero said that “we have a lot of other projects going on.”
Other secret projects? Alien research!!! That's how they stay ahead of the curve. I knew it!
It sounds like a newer version of the testing facilities we were using at HP 15 years ago.
good job on the photography but these are pretty standard anechoic RF testing chambers. The only news worthy thing is that Apple is main-steam enough that people actually looked at these photos.
Any company doing serious RF development will either have their own and rent time in a dedicated testing facility.
Search google for "anechoic chamber" and you'll find hundreds of photos of such facilities.
The US Air Force has one big enough to park a C-130 in :)
The fact that they had a bug in their signal strength algorithm is bad, but one can't complain the problem happened because they weren't testing.
For the last time.. it was not a bug. A bug has unintentional consequences. What they were doing was intentional.
"His name was James Damore."
iPhone 4 is out. Some people have signal issues due to a design decision. Many people think it's the best phone they've had. Many people think it's the spawn of Satan. Apple held a press conference to give away a fix to the problem. Some people think the fix is ugly and doesn't do anything about the Satan problem. The End.
This flamewar has been pounding Slashdot for a long time, but since the lost/recovered prototype iPhone 4, it's been ridiculous. Every . Single . Day on Slashdot there has to be an Apple flamewar, and the Anti-Apple jokes now begin to bleed into other stories. Too much coverage, Slashdot. More physics, less phones. Leave the intensive, by-the-minute coverage of mobile phones to Gizmodo and Engadget.
STOP . AMERICA . NOW
The tour was for show because it sidestepped the key points. That is,
I believe they knew about the issue early on. I further believe it's quite possible the engineers had intended to coat the antenna but Jobs didn't like the look of a coated antenna. When it came down to "what are we going to do about this?" the logic that prevailed was "It only affects a minority (left-handed customers) so we'll put the bumpers out there and charge extra. That'll address the problem and bump our ROI on the phone. Problem solved." They failed to anticipate how the decision would blow up in their face and since it's probably Jobs who made the call, it's taken this long for the rest of Apple to convince him he had to acknowledge the mistake.
iPhone 4 is out. Some people have signal issues due to a design decision. Many people think it's the best phone they've had. Many people think it's the spawn of Satan. Apple held a press conference to give away a fix to the problem. Some people think the fix is ugly and doesn't do anything about the Satan problem. The End.
This flamewar has been pounding Slashdot for a long time, but since the lost/recovered prototype iPhone 4, it's been ridiculous. Every . Single . Day on Slashdot there has to be an Apple flamewar, and the Anti-Apple jokes now begin to bleed into other stories. Too much coverage, Slashdot. More physics, less phones. Leave the intensive, by-the-minute coverage of mobile phones to Gizmodo and Engadget.
Sorry, but your post really doesn't make it clear whether you are for or against the iPhone... How the hell are the Slashdot crowd supposed to mod that?
Just pick a side and start whining - you'll get the hang of it soon enough. They'll be another iPhone 4 submission tomorrow, so you can try again then.
Steve Jobs said explicitly in his press conference that Apple's decision to use an external antenna was part of a design tradeoff to house the phone in a slim case while offering extended battery life. He even acknowledged that there are designs that would provide substantially improved reception, such as an antenna protruding from the case. So Apple tested the reception of the design, and found that no matter how you hold it, performance was similar or better to that of their previous phone, as independent testers have since found, and concluded that Apple's customers would be happy with the design. So far, the phone is selling quite well, and returns are lower than previous models in the line, suggesting that Apple's estimation of its customers' priorities is pretty accurate.
I'm not sure how that translates into not giving a shit.
... industrial designers that did the iPhone case design and overruled the antenna/RF engineers, put them in the test chamber and turn the microwaves up to 'bake'.
It really doesn't matter how many fancy anechoic chambers you've got. If the art majors who spec the kewl stainless steel antenna have the last word, its a culture problem, not a technology problem.
And still most people don't buy technology, they buy products. I mean, they even buy bicycles with no fenders on them. How crazy is that? And last I heard HD even sells vehicles with no roof! You get soaked if it rains! Must also be one of those culture problems. This is clear flaw of their products. Someone should sue them.
People buy these bicycles and cars because they understand the implications of no fenders or roof. But most customers can't be expected to understand the performance issues associated with poor antenna design. If iPhone customers were informed of the tradeoffs of the cool stainless steel antenna (crappy reception) do you think they'd still make the same choice they did? Its the engineering department's job to ensure a minimum level of performance in the design, particularly when the tradeoffs aren't apparent to potential customers.
Have gnu, will travel.