sudaltsov

Mike Wuerthele said:

sudaltsov said:

Mike Wuerthele said:

sudaltsov said:

Mike Wuerthele said:

sudaltsov said:

Mike Wuerthele said:

sudaltsov said:

I cannot believe you did not mention the cameragate of the new one. There are plenty of images around there showing how dramatically worse the new camera became.
https://www.macrumors.com/guide/macbook-air-camera-quality-issue/

Why? It is hardly a "gate," and service calls for the "issue" aren't statistically significant. As with any other issue, if you have a problem with a piece of Apple hardware, go to a service center and get a case number so it can be tracked.

So far, the 2018 MacBook Air has about the same failure rate over the corresponding period of time as every other MacBook Air ever released past 2010, and a notably lower one than the original Air. If there was a widespread problem, it would manifest in the service data.

Also? Watergate was the name of the hotel. The scandal in question wasn't about water. Adding -gate to things is... silly.

If you look at the number of complains and returns on Apple support site (need links?) - you'll see it is serious statistics. And you'll see that all Apple can offer is 'yes we designed it that way, just return/replace and shutup if unhappy'. People do return those devices, in bulk. It is design issue, not an issue with individual items. Go to any Apple store, compare the camera of Air 2017 and Air 2018. You'll be surprised. I did that with at least 5 different stores in different countries. Also, I returned my Air 2018 for replacement - just to get the new one with the same horrible camera. Other people did that too. What other kind of proof do you need?

"It would manifest in the service data" - do you have access to it? I doubt Apple publishes it.

BTW "watergate" created a naming pattern, it is not a hotel name anymore. There was antennagate with iPhone

I have access to return data as well. There is no increase over any other model. And yes, I have access to a pretty large subset of the entire set of Apple's service data going back to 2010. It is where the keyboard failure numbers came from.

I'm aware of the naming pattern. It was stupid during every other -gate too.

What kind of proof I need is actual data supporting the fact that there is a large issue and the actual data says that there isn't one. I'm not saying that you didn't experience the problem, and that there aren't complaints about it. What I am saying is that it isn't a giant issue affecting the entire line, beyond more than a small percentage of purchasers because the data does not say that it is one. Should the data show a shift, then the conclusion will change.

I respectfully disagree (even knowing that you have access to the stats).
That is the design issue affecting the entire line, which was confirmed many times (including Apple support people - 'functions as designed' looking at those photos). Look at the poll https://9to5mac.com/2018/12/02/some-2018-macbook-air-users-experiencing-poor-facetime-hd-camera-quality-are-you-poll/ near 80% (small percentage?) out of 4K people experience that. So far noone could present good (at least not seriously worse than Air 2017) quality photos made in less than 'Apple-grade perfect conditions'. Could you please by any chance do that, if you happen to have access to two devices? There was no single report of the scenario "I had a bad one, sent back for replacement, got the good one" - but there are numerous "replaced bad one with the same bad". I sincerely would like to see some hope here, some proof that there exists good Air 2018 devices with decent (not worse than Air 2017) camera. Then perhaps I would shut up on that issue forever. Please prove me wrong, not with statistics, just with the photos made on 2017 and 2018 devices. Proving that mythical "good-camera-enabled" Air 2018 exist at all.
PS. BTW, there was a huge thread on Apple support website about it. What did Apple do? Deleted it. Now, there are smaller threads but they still grow.

Now I'm starting to wonder if you read the article, since the only comments you've made here, ever, have been about this camera. There is a side-by-side in this very article you're commenting on, and in the video.

You're welcome to disagree based on your own problem and a disproportionately loud forum-going population, but data is data. Also, regarding the support forums, casual users who get their problem solved have a tendency to not return and report success, which is why we can use it as a general proof of life (or death, as the case may be), but can't use it for actual data.

I definitely read the article, including those 2 photos. They are made exactly in the conditions I described as 'Apple-grade perfect conditions'. In particular, the bluish shade of white, and very bright environment. For home users, the light is slightly less bright (and the white is of yellowish shade) - in those conditions Air 2017 works fairly well, 2018 is a disaster.
I would really love to see the Apple service data about it, but I suspect you are under DNA (sorry, that slightly undermines my trust for obvious reason...). I can only operate publicly available data - polls, forums etc (accepting your comments about the necessary bias of those sources...). Oddly enough, with this story going on since November, you are the first person I am talking to trying to claim there is no design issue here...
I am absolutely sure at this stage that if we would have at least one case 'Air exchange solved my problem' - it would be reported. I m sure of that just because the number of affected users is fairly large. Also, it would be very difficult to explain the large number of users saying 'replaced did not fix my problem'. What are the chances of the shells hitting the ground in the very same location?

BTW, another fact established on forums is that MacBook Pro 2018 with touchbar has the same issue (the camera is just slightly better than Air 2018). Good cameras ended with Air 2017 and touchbar-less Pros. Kudos to Apple.

For what it's worth, the picture was taken at Vadim's desk. and not in a studio environment. To be clear about what I'm saying, I am not saying that there is absolutely no design issue -- what I am saying is that there is no affirmative proof of one in the service data. There is plenty of anecdotal data from pissed off people, but there isn't as of yet a corresponding number of returns or service claims on the hardware.

"What are the chances of the shells hitting the ground in the very same location?" -- I'm not sure I understand what you mean with this?