Firmware release 0x39 will go into M72 (maybe M71) and should help address these issues. CL: https://chromium-review.googlesource.com/c/chromiumos/third_party/linux-firmware/+/1307358

 Issue 781882  has been merged into this issue.

These issues affect Caroline too.  See https://bugs.chromium.org/p/chromium/issues/detail?id=781882#c67

Correct - Caroline also uses Stone Peak 2.

Now that https://chromium-review.googlesource.com/c/chromiumos/third_party/linux-firmware/+/1307358 has been merged. Is there any more information about this being fixed in M71 or M72?

Is this fix available in the Beta or Dev Channels?

Thank you.

BT should be better in M71 b/c of the new firmware that was checked in. In M72 (or maybe more likely M73), we may be able to enable the rfkill on newer hardware (Pixel Slate, Nami, and later, but not Pixelbook or Caroline). Those CLs are in review.

#9 I'm on Pixelbook 71.0.3578.49 now and BT is still flaky - has the new firmware been pushed in that version?  Will the "rfkill" ever be enabled on PB or Caroline?

wpwoodjr, you have the latest BT controller FW. You are right that it is still flakey. It may be better, but it's still not best. Realistically it will be a challenge to make it flawless, but we are hoping right now to eliminate the need to reboot.

rfkill is a hardware solution to power-cycle the BT controller. It requires new hardware, so it's not available on PB or Caroline.

#11 so are Pixelbook owners SOL then? There's really no other way to power cycle the Bluetooth controller? It's been a standard feature on laptops for as long as I can remember...

#11, the other day I was listening to Youtube on a Bose Soundwave and BT kept cutting in and out. At least once I had to manually reconnect.  Will that behavior ever be fixed?

This is honestly an outrage. A thousand dollar computer shipped with an essential feature broken, and there's no fix after a full year, and now we're being told it might be unfixable? 

jshee is right, this is totally unacceptable. Just like the screen tearing color saturation bug. https://bugs.chromium.org/p/chromium/issues/detail?id=781483

Unfortunately I have to agree with the negative sentiment here. For such a serious issue to go on for so long it's really frustrating and disappointing. I am not sure how this problem hasn't received more airtime as I can't imagine anyone using bluetooth not being frustrated by it (I haven't seen any reason to believe that the issue is restricted to only a subset of Pixelbooks).

The Pixel Slate is currently being marketed and reviewed in the press.  A quick search has revealed that some of the more thorough reviewers have brought up bluetooth issues presumably resulting from these 7265 firmware problems.  

I think it's worth frustrated parties contacting reviewers and asking them to read these threads and inform their readers of the chances of the issues being resolved (given that Pixelbook users have waited a long time and still haven't had any joy).  

If it transpires that rfkill is the only solution, then effectively we have faulty hardware on our hands, given that it wasn't designed to drive the rfkill bodge that Intel provide, without which (so far) it seems that the bluetooth system is all but useless.  "Have you tried switching it off and on again?" Bad enough to have to do that to the BT controller, let alone the whole flipping system.

If certain other leading consumer electronics vendors had lumbered users with this, there would be a massive deal about it from day 1.

Have people tried this?
https://chromeunboxed.com/news/google-fixing-bluetooth-issues-chromebooks/
There's a new experimental bluetooth stack you can enable after v69.
Anecdotally my bluetooth is working right now with it.

chrome://flags/#newblue

I have switched as well, although I have heard that the issue still
persists from others. If I can go two days without the BT adapter
"disappearing" I'll post again.

On Tue, Nov 27, 2018, 2:35 PM col… via monorail <
monorail+v2.2623127217@chromium.org wrote:

Honestly not sure why it took so long to realise the Bluetooth adaptor was able issue. The solution presented is pretty much forcing us to upgrade to the Pixel Slate (was this intended by Google) having said that I have already seen reviews of the slate talking about Bluetooth connectivity issues. Regardless Google needs to be held accountable and realise that they have provided faulty hardware and need to provide phyiscal replacements for the pixelbook with a new chip or a significant price discount on an upgrade to a Pixel Slate. I will do my best to spread this issue and make it aware to the tech media and suggest others do the same because if Google are not ready to take responsibility they will be in line for a class action. Very disappointed

To be fair to Google here, this is an Intel issue that affects other laptops too.

This post led me to try a test: https://communities.intel.com/thread/61208

Here's some highlights from the post:

"Furthermore, if there no Bluetooth devices to be connected to your computer, turn it off because the wireless and Bluetooth* combo share the same wireless connection and antenna so turning off the Bluetooth can let the wireless connection has the entire usage of bandwidth and the antenna. Also, by turning off the Bluetooth*, it will help to decrease the signal interference since Bluetooth devices uses the same band."

and

"If you're using a bluetooth device, I think it's the fact that the adapter is trying to handle both bluetooth and the wifi connection - and as far as I can tell, it can't."

So I went to speedtest.net and did a speed test. With BT on but nothing connected, I got 106 Mbs down and 12 Mbs up. Then I connected my BT Bose Soundwave and got 10 Mbs down, 12 Mbs up. That's without playing any music.

That's a 10X difference in download speed!!

With my BT game pad connected, I got 33 Mbs down, 12 up. Better than BT audio, but still 1/3 the speed vs having no BT device connected.

I've since done some more tests, sometimes I get as high as 50 Mbs with Bluetooth audio connected, but not 106 Mbs.

On the 5 GHz band, there seems to be no interference and I get 135 Mbs down with and without a BT audio device connected. So maybe that's the answer, use 5 GHz.

However I'm inclined to purchase a separate BT audio device such as this: https://www.kickstarter.com/projects/humanthings/genki-bluetooth-audio-for-the-nintendo-switch

#21: Bluetooth and 2.4 GHz wifi have always interfered, and that is "normal". They occupy the exact same frequency and try to talk over each other. Your phone will have the exact same issue, as will anything else using both Bluetooth and 2.4 GHz wifi. Honestly, 2.4 GHz wifi should always be avoided when possible, it just has way too much interference. 

That being said, it could be the cause for the disconnects that people are experiencing. But it's not the reason for the controller dying and requiring a reboot.

#22 I don't have any issues with my Pixel 3, connected to 2.4 GHz wifi and BT audio.  I think the issue here is the shared antenna.

#22 although its true, as you say, that the interference is still there, as Speedtest is slower on the Pixel 3 too at 2.4 GHz when BT is in use.  However  BT audio doesn't disconnect on the Pixel 3, like it does on the Pixelbook.

Since moving to the beta channel and installing Version 71.0.3578.49 (Official Build) beta (64-bit), bluetooth has always been available after sleep (so far), so that meets my requirement of smart unlock working.  Hopefully when this goes to stable I can get off the beta channel.  I have no idea whether other bluetooth devices work better with it now, and I almost don't want to try as it'll just anger me again if they don't.

Presumably being a bit more liberal with the settling timer has paid dividends when waking from sleep, anyway.  I really didn't want to have to return the Pixelbook as I find the  industrial design of it absolutely fantastic - the keyboard is almost perfect for such a thin device, and the general build quality matches it.

I had the problem before, but also since switching to the beta channel for Linux support I didn't had the problem anymore. Now for a few weeks in Stable with Linux enabled I also don't have the problem anymore. So smart unlocking is working since then. Not sure if it have to do with the Linux support enabled.

On 5 GHz wifi, I'm still seeing several BT disconnects per hour.

#25 How long have you been on build Version 71.0.3578.49 (Official Build) beta (64-bit)?  I am also on that build and bluetooth is no better at all.  Still grays out after sleep.  Constantly disconnects from devices during ues etc.
Update: Now on Version 71.0.3578.71 (Official Build) beta (64-bit) and bluetooth still not working properly.  All the same issues.

#30 I've been on 71.0.3578.49 for 2 days now.  I have no greying after sleep, whereas before (on latest stable) it was quite easy to reproduce the issue.

Have you fully done the "Better Together" thing with your phone (as that's also in the beta)?  I did (with my P20 Pro), and was wondering if that had made any difference too.  I'd say the Pixelbook takes slightly longer to unlock via phone than it did before, but it has succeeded 100% of the time so far.

As I mentioned before, I don't use bluetooth with it for anything other than "Smart Lock", which is beginning to resemble its name now.  Really sorry to hear it's still failing for you.

 Issue 892985  has been merged into this issue.

After a couple of days, bluetooth is MIA. New bluetooth stack has no effect.

After all these years of technological advancement, we still have such a simple problem that cannot be fixed?  I didn't even have such problem back in all time using Bluetooth 1.1 mouse, keyboard, and headset with WiFi on Dell, HP, Lenovo, IBM, Mac, windows, Ubuntu, MacOS.  And now only Pixelbook has this issue and there is no fix and there should be a recall just like cars.

It's amazing how a standard such as bluetooth is as far from a standard as possible. You would think being on version 5 with multiple versions within each number, it would be pretty fine tuned by now. 

And I can't buy into the "in fairness to google" just because other companies used the same chip and have the same issue. All companies should be held accountable and have a generous (read beyond fair) solution for its customers who have been living with the issues after shelling out a fair amount of money for a premium product. While it's not quite up to Samsung's combustible issue, they gave 50% off the Note 8 to all the Note 7 owners. Sounds like a start to me for the PB. And not the Pixel Slate. If I wanted a tablet and separate keyboard, I'd buy an x2 and save the extra coin. 

#36 Here is the real rub.  I reached out twice to file a warranty claim with google about the bluetooth issue during my warranty period. They pushed me off each time with, "it is software issue being addressed and a fix is coming soon" and would not go any further.  My response was this is almost going on a year now and no fix, I think that is fair on my part I want to file a warranty claim.  Which they did not oblige.  Now I am out of warranty and there still no fix.  And now it comes to light it is an Intel hardware piece.   

I'm still experiencing this on 11151.61.0 (Official Build) stable-channel eve with newblue enabled.

Issue 907361 has been merged into this issue.

 Issue 919155  has been merged into this issue.

Issue 890906 has been merged into this issue.

 Issue 888252  has been merged into this issue.

Issue 879233 has been merged into this issue.