Any insight into why Apple seems to have a much better bluetooth stack? I do a lot of BLE development that has to work cross platform, and we see constant GATT connection issues on android compared to almost none on ios.
Apple has a unique relationship to its hardware, and the money to bend vendor's firmware to their will. Half of the problem with Bluetooth is the host stack. The other half is the firmware running on the controller on the other side of HCI. If the controller ever gets screwed up, the host stack can only disconnect from HCI and totally restart the controller.
The third half is the bluetooth firmware on the other device. The fourth half is the other firmwares on the other device. The fifth half is the specification(s). The sixth half is the RF environment.
Haha -- yes, indeed. Bluetooth the spec and implementation are an absolute dumpster fire.
"Bluetooth is a layer cake of sadness" is the turn of phrase we used for a while on the Glass connectivity team. One of our project founders, Thad Starner actually apologized to me for the mess it became; apparently it was supposed to be simpler, but when Nokia took ownership back in the 90s, it started to go downhill.
Our lead on the connectivity team at the time had a crazy idea to rewrite the spec using only ACL and L2CAP, but never really went anywhere with it because of the the Glass org implosion.
A huge factor is that iPhone has a large, long-standing market share, with relatively few different OS and hardware versions. This means that everyone else has been able to test their Bluetooth implementations for interoperability with iPhones.
Anecdotal, but my new 16 inch MBP drops BT connections all the time. I had to give up the keyboard and mouse I've used for years across at least 5 computers, mostly Mac's because they disconnected so much. Even the Apple keyboard and mouse I switched to drop occasionally.
FWIW, I also have a 16" MBP and I've also had a ton of problems with Bluetooth on this thing.
I've noticed that Bluetooth connectivity is significantly worse when the laptop is closed. You might see if keeping it open helps.
Resetting the Bluetooth module also helped resolve some persistent connectivity problems I was having (shift+option+click on the Bluetooth menubar item; choose "reset" from the menu).
Eventually this thing started having kernel panics every time I plugged something into a USB-C port and I had to send it back for replacement. Not a great experience.
I think there might be some issue on that hardware where 2.4ghz wifi and bluetooth share an antenna, and can interfere with each other? If you're using 2.4ghz and can use 5, give that a try.
There's something odd about the 2019 MBP Bluetooth hardware. The most interesting part is that just plugging in a Cambridge Silicon -based USB dongle can kill it permanently: https://discussions.apple.com/thread/250944058
Doesn't kill it permanently; there's a fix involving connecting a CSR 2.0 dongle. There's also a workaround that sets an nvram var. Overall it's an absolute clusterfuck, though, because Apple still haven't fixed the problem.
The workaround with nvram doesn't work if you already made the mistake of plugging in the CSR dongle. I don't know about the fix with CSR 2.0 dongle, it seems that nobody's selling them any more. Could work, could be an urban legend like all the reported fixes with nvram and pram resets and restoring various files.
I can confirm the 2.0 BT dongle fix works because I had that exact issue and that's how I fixed it. Of course, I still made Apple replace the laptop (was less than 2 weeks old), but doesn't really seem like they've noticed seeing how it's still not patched almost a year later.
My guess would be that they can apply much more pressure to the Bluetooth chip vendor. The buying power that Apple has for designing in a particular chip is much bigger than any individual Android manufacturer (even Samsung). That gets them the leverage to get the chip vendor to do what they want.
> The buying power that Apple has for designing in a particular chip is much bigger than any individual Android manufacturer (even Samsung)
Why does Samsung have smaller buying power than Apple? Doesn't Samsung sell more phones than them? Or is it because while Samsung sells more phones in total, Apple still has the most successful single models?
A typical manufacturer approach is to beat the hell out of your suppliers on price to make your margin which is likely most of what Samsung does with its buying power. Apple does that but is also willing to throw money at issues and go as far as financing facilities for vendors. Where most manufacturers don't care about the device they sold the second it's out of warranty[1], Apple takes a longer term view of the relationship. So Apple is far more likely to say 'we need this fixed and are willing to provide X million dollars and a team of our own engineers to solve the problem' while Samsung probably goes more like 'make cheaper and make it better... and make it cheaper!'
So the reason Samsung typically has less influence is that when all you do is crush your suppliers margins to make your own, said suppliers don't tend to make much of an investment in making things better since they are incentivized to just make them cheaper.
[1] in fairness, they can't afford to: while Apple has an ongoing revenue stream from its devices, most other manufacturers don't. It's Google/Facebook/etc who monetize the devices post-sale while for the original device manufacturer it's merely a liability at that point. This is a factor in why Android has a rather dismal track record re: updates on older devices.
It's because Samsung can get away with selling total garbage. I work on an audio-related app, and a huge majority of the hacks in the app to work around bugs in phones are for different Samsung models. Still more than half of our users keep buying them.
This, but slightly revised: it's because Samsung is in the market of selling cheap hardware. Apple sells luxury products and part of their value-add is leverage with their vendors.
Apple can come to a vendor and say "these are our constraints on what we are willing to buy. Here's testing benchmarks. You are required to meet them. Failure to do so voids the purchase contract."
The vendor will then, of course, say "We'll have to charge you more if we're spinning up a test infrastructure we don't have."
And then Apple will negotiate a price point and pass the anti-savings onto the consumer.
They do this at multiple levels at multiple points in their hardware story. I met someone once who worked on the USB integration specs back in the first-couple-generation iBooks. Apple built a rig for physically testing the connectors (including multiple cycles of intentional mis-plug, i.e. flipping a USB-A connector over wrong-side-up and pushing hard against the socket). They told the vendors selling them USB sockets that the sockets had to survive the rig, or the sale was void. Vendors priced accordingly. But the resulting product is more robust than others on the market.
My social network orbits around several people who experience apple’s testing rigor- and the stories I’ve heard aligns with the parent comment.
Apple sounds like it really is that rigorous with the quality of hardware and drivers/firmware they use.
Some of the stories of nitpicking that I’ve heard are truly awe inspiring. On the other hand, I’d hate to be the engineer on the vendor side trying to please apple. (Mostly because some crap PM or sales person made promises without consulting engineering on what’s possible with the available time and resources)
> Samsung is in the market of selling cheap hardware
I'm not sure about this. There are people who pay >1000€ for their flagship phones and believe them to be premium products, but they are just as buggy as the cheap ones. Huge amount of CPU power and impressive camera specs, though.
IIRC that model doesn't give apps audio route change notifications when Bluetooth is disconnected, which screws up audio timing. Or something similar, each model has different bugs. But yeah, it doesn't spontaneously catch fire, so on Samsung scale it's good. Reviewers generally don't know about this stuff, because app developers have already worked around it.
My impression as an Android developer (not backed by any systemic research, just personal experience) is that Samsung feels less bound by the platform standards because - due to their large market share - they can simply get away with it.
We've been together 4 years, its USB port is a little loose now. Too much charging everyday, but the screen is still pristine. The camera is fine so we can go out and take photos at the beach. I'm happy man. It does what it said it would do and did it. And still doing it.
You don't get to see the hacks in the apps you're using, starting with comments like "On some Samsung phones X happens even though API documentation says Y. We work around it by Z" or don't directly pay for the work that went into discovering the workaround.
> Apple still has the most successful single models?
Not just that, every apple device has wireless access, and Apple has thrice the operating income and more than twice the net income of Samsung Electronics.
But yes the "value" of individual devices is also part of the equation, in the sense that Samsung has a lot of cheap-ish devices with fairly short lifetimes, they're not going to fight for device quality. Apple has a very limited number of devices they support for a long time. And they're probably bringing in a lot of baggage from having been screwed over by the plans and fuckups of their suppliers in the past.
And even then, the more time passes the more they just go "fuck'em all" and move chip design in-house, not just the "main" chips (AX, MX, SX) but the ancillary as well: the WX and HX series integrate bluetooth on the SoC. There's no doubt they'll eventually go their own way on larger devices as well, the U1 is probably the first steps towards that.
I didn't realise that Samsung actually sold more phones than Apple (I've just looked at the numbers and you're right). Apple's smaller set of devices does help them a little but there isn't a lot to choose there.
I think my overall point is still valid because I have had Samsung phones for a while and have found their Bluetooth to be pretty good. This is not surprising as Samsung actually bought one of the biggest Bluetooth chip vendors (CSR) at one point, so they do have control over the full stack.
Samsung bought what was advertised as mobile handset business after CSR failed to spot that combo BT/Wifi chips were the future and threw away their lead. Qualcomm got the rest a few years later.
("what was advertised as" because there wasn't really a differentiation in the R&D bits, so there was a somewhat arbitrary split and hasty redacting of repos given to Samsung to avoid names of other customers in comments).
Samsung's phone division and electronic parts division aren't the same thing, so there was no guarantee that the phones would buy the Bluetooth/Wifi from their new acquisition, although I hear they did eventually.
I'd imagine lots of reasons. Apple has the hardware guys to design whatever chip they need. They also have a mountain of reserve capital (Something like $1 trillion I believe?). Further, Apple's closed ecosystem means that if you want them to sell them your hardware you have to conform to their standards.
With Samsung and android, it's a different story. There are many android vendors and the people producing the Bluetooth chips are selling to many of them (broadcom). Samsung has the ability to make their own chips, but to get everything working flawlessly they not only have to make their chips awesome but also improve the android driver stack to work with their new awesome chips. That stack has to also be compatible with the other bluetooth manufactures on the market making it a harder change to make.
In other words, with apple and a vendor, there are pretty much just the 2 parties involved which control everything. With Samsung and vendor it's not just them but also the likes of google and other bluetooth vendors that can get in the way of really fixing things.
I get what you're saying, but Samsung is also huge and Bluetooth is used in lots of electronics that Samsung sells -- including laptops, tablets, smart watches, TVs... or conceivably could sell, like future IoT products.
If someone senior at Samsung said to their vendor "good Bluetooth or you lose the Samsung account", that would provoke some, um, intense conversations at the vendor between sales and engineering.
Incidentally Apple sometimes has more than one vendor too, so it's not just two parties. I know cases where they've had two suppliers. Displays and modems come to mind, although I've not Googled to verify.
> If someone senior at Samsung said to their vendor "good Bluetooth or you lose the Samsung account", that would provoke some, um, intense conversations at the vendor between sales and engineering.
The problem: There aren't that many suppliers left in the field, and Broadcom knows that their customers are pretty much locked in. The notable exception is once again Apple, they have proven that they can and will go and implement the technology on their own if their suppliers fail to meet their expectations.
Besides that they also lack the amount of control over the software side. Google is the entity that controls the stack, not Samsung - their responsibility ends at the kernel / HAL interface.
So why should Samsung invest more than the bare minimum when they can't get anything measurable in return?
I get what you are saying but they are simply in different positions. Broadcom can say "Hey, there's a bunch of work here to make this not suck" and Samsung, even if they wanted to do their own thing, realizes they too would have to go through that effort to make everything work. It wouldn't be a simple matter of just making non-sucky bluetooth, they'd have to also work with the android OS to improve the bluetooth stack and no guarantee that those changes can be merged.
Apple controls their whole stack. They've already written the Bluetooth stack for their OS. They only have to service their devices.
> Incidentally Apple sometimes has more than one vendor too, so it's not just two parties.
Not the point I was making. It's not an issue with multiple vendors, its and issue of who controls what. Those other vendors also have to conform to apples standards if they want to sell apple their products. What I'm talking about is the fact that a bluetooth device manufacture has to conform to google's android standards if they want to sell their chips to android manufactures, not to samsung standards. That's where the leverage goes away.
If samsung ever pulls the trigger and uses Tizen everywhere, then they'll be more in Apples position. Until that happens, they need to work with google to get stuff done.
Except that's not how it works? Chip manufacturers just care about selling chips, not about being in the iPhone or being in an Android phone. It doesn't matter that some chip works in OnePlus phones, if it doesn't work in Samsung phones Samsung's not going to buy it.
If broadcom says it's going to be expensive to fix, either you pay or you don't. But it has nothing whatsoever to do with "controlling the whole stack".
As an aside, and it's totally irrelevant to the above, but there's nothing other than the amount of work involved preventing Samsung writing their own bluetooth stack. They write plenty of custom drivers and they created a folding device prior to OS support. If they wanted to they could; they just apparently don't think it's worth the cost.
Edit: according to a comment down thread they have done exactly that.
> If someone senior at Samsung said to their vendor "good Bluetooth or you lose the Samsung account", that would provoke some, um, intense conversations at the vendor between sales and engineering.
The same thing could be said for Qualcomm's failure to support it's SOCs for more than a couple of years. Device makers could force their hand.
> support it's SOCs for more than a couple of years
That works, even works well, as long as device makers feel that supporting SOCs for a limited period helps them sell more phones.
Apple has changed the rules of the game by supporting its devices for longer -- and as phone upgrades become more infrequent due to plateauing technology, I think device makers will realize this.
Samsung has already committed to supporting recent Galaxy devices for at least 4 years -- at least with security updates[1]. I suspect this was also because they found that their extremely short-sighted prior policy re security updates encouraged corporate phone procurers to ditch Samsung and go with Apple.
I would be surprised if they didn't. They advertise their own Bluetooth chips H1 and W1 in AirPods and some Beats products. These chips definitely have their own firmware and it would be rather ridiculous not to also have their own firmware on the host side, maybe even re-using code.
There's quite a bit of evidence that they do. Apple's strength has always been at the point at which hardware is built and interfaces to the software, including firmware.
I still find it buggy on my iPhone 12 Pro, as a user at least. Often says 'connected' to my headphones (Sony WH-100XM3's) when it's not, connecting to Alexa devices to stream audio often fails and requires a reboot to connect properly.
I can't believe Bluetooth is still such a pain in the bum in 2021.
I had the idea in my head (so take with a grain of salt) that parts of the BT stack are underspecified such that different implementations tend to have slightly different interpretations of the standard, so the problem isn't even which vendor you use so much as going all-in on any single vendor so the devices all agree on which reading to use.
Although of course Apple might well be better anyways; one would hope that billions of dollars in R&D plus caring about quality makes a difference.
I've actually worked in this area. You are basically correct.
Sometimes you have to make a choice on which brands/chipsets you support. Devices on different ends of the compatibility spectrum can basically be mutually exclusive. IIRC if you advertise A2DP some devices supporting only HSP won't work, so you can make some hacky workaround but then your nicer A2DP equipment is harder to use. If you only need to guarantee support for X subset of devices you control, it's easy to tweak the settings so they work well together.
Actually the spec is overspecified and repeats itself multiple times in different layers. L2CAP specifies a TTL, so does RFCOMM, so does SOC, so does...
They don't smooth out, for me, at least. Airpods Max, for example, can just decide to transmit nothing but deafening static to the other side in the conversation.
As a long time Apple user, I really don't know where that's coming from. Endless issues with bluetooth over the years, including full system crashes. Curiously iOS bluetooth stack seems to be more stable than on macOS. Or maybe it's just hardware differences.
Apple's stack does work great with Apples hardware, though.
They have less hardware variability and also their bluetooth stack is still quite buggy but everyone works around their bugs on device side due to popularity.
Could it be because they only have a handful of hardware variants to support, vs hundreds on Android? Presumably, the stack itself is somewhat sane on both sides, but the hardware's bugs, poorly specified or undefined behavior is constantly throwing wrenches in the machinery - Apple managed to fix most of them for the hardware they support, but Android has no chance as they have to support hundreds of different chips.
This is not specific to bluetooth or drivers, but I don't fully buy the thesis/deflection that Apple just has a much smaller set of hardware to support. I run a hackintosh system myself and it's more stable than Windows. I'm starting to think Apple just has stronger general QC. Windows is just as glitchy and crashy on Surface devices as anywhere else.
The counter point does apply to drivers, but it's really not just that.
