Should we upgrade the reference Galaxy S4s to Android 5.0?

Vladan Djeric vdjeric at mozilla.com
Fri Jun 12 22:20:40 UTC 2015


As it turns out, FHR didn't report the Android device model. Even though it
has strong opt-in bias, Telemetry numbers are all we have, so let's upgrade
the phones to Android 5.0

On Thu, Jun 11, 2015 at 2:30 AM, Avi Halachmi <ahalachmi at mozilla.com> wrote:

> The Lolipop updates for the various variants of the S4 started rolling out
> around 2 months ago, and are still rolling out (E.g. in Bulgaria it rolled
> out today, in Italy - last week, etc).
>
> I think it's likely that most users who care enough would get it updated
> sooner or later.
>
> So even if right now the majority is not on Lolipop yet, we should
> probably use Lolipop as reference. Currently the latest version for the S4
> is 5.0.1 in most places.
>
> On Wed, Jun 10, 2015 at 6:22 PM, Benjamin Smedberg <bsmedberg at mozilla.com>
> wrote:
>
>> We should get as close as possible to what most of our users will have.
>>
>> --BDS
>>
>> On Wed, Jun 10, 2015 at 11:20 AM, Vladan Djeric <vdjeric at mozilla.com>
>> wrote:
>>
>>> Hey all,
>>>
>>> The S4's are prompting us to upgrade to Lollipop, but we may want to
>>> continue using Android 4.4 if that's what most of our users have.
>>>
>>> I asked Roberto to look at Fennec Telemetry to get the Android version
>>> distribution for S4 devices.
>>>
>>> He examined 4389 Fennec Telemetry pings from the North American version
>>> of the S4 (GT-I9500) submitted on June 9th:
>>>
>>> ~50% of pings from KitKat (4.4)
>>> ~35% of pings from Lollipop (5.0)
>>> ~15 % Jelly Bean (4.2, 4.2.2, 4.3)
>>>
>>> There's opt-in bias with Telemetry as always, but it seems like we
>>> should upgrade. What do you think?
>>>
>>> P.S This is our new contentperf mailing list
>>>
>>> ---------- Forwarded message ----------
>>> From: Roberto Agostino Vitillo <rvitillo at mozilla.com>
>>> Date: Wed, Jun 10, 2015 at 9:07 AM
>>> Subject: Re: Fwd: About:telemetry
>>> To: Vladan Djeric <vdjeric at mozilla.com>
>>>
>>>
>>> Hi Vladan,
>>>
>>> Looks like KitKat is the winner:
>>> http://nbviewer.ipython.org/gist/vitillo/59c7f14eb484d8a38e3a
>>>
>>> Roberto
>>>
>>> On 9 June 2015 at 22:22:14, Vladan Djeric (vdjeric at mozilla.com) wrote:
>>> > Hi Roberto,
>>> >
>>> > Can you do a Telemetry analysis to find the most popular Android OS
>>> > versions among Fennec users whose device.manufacturer field =
>>> "samsung" AND
>>> > device.model = "GT-I9500"? This is Fennec Telemetry, so it might be a
>>> bit
>>> > different from what you're used to. We need this info to decide which
>>> > Andorid version to put on the reference Samsung Galaxy 4 hardware we're
>>> > using for the Content Performance
>>> > program.
>>> >
>>> > Thank you,
>>> >
>>> > Vladan
>>> >
>>> > ---------- Forwarded message ----------
>>> > From: Aaron Klotz
>>> > Date: Tue, Jun 9, 2015 at 5:12 PM
>>> > Subject: About:telemetry
>>> > To: Vladan Djeric
>>> >
>>> >
>>> > General Data Click to toggle section
>>> > General Data Name Value
>>> > type main
>>> > id 8c42322c-b4c9-416d-be70-c7365cc15757
>>> > creationDate 2015-06-09T20:57:13.506Z
>>> > version 4
>>> > application.architecture arm
>>> > application.buildId 20150609081916
>>> > application.name Fennec
>>> > application.version 41.0a1
>>> > application.vendor Mozilla
>>> > application.platformVersion 41.0a1
>>> > application.xpcomAbi arm-eabi-gcc3
>>> > application.channel nightly
>>> > clientId 533d1ce4-1e39-471d-bc0c-355ffc84a04a
>>> > Environment Data Click to toggle section
>>> > build Name Value
>>> > applicationId {aa3c5121-dab2-40e2-81ca-7ea25febc110}
>>> > applicationName Fennec
>>> > architecture arm
>>> > buildId 20150609081916
>>> > version 41.0a1
>>> > vendor Mozilla
>>> > platformVersion 41.0a1
>>> > xpcomAbi arm-eabi-gcc3
>>> > hotfixVersion null
>>> > partner Name Value
>>> > distributionId null
>>> > distributionVersion null
>>> > partnerId null
>>> > distributor null
>>> > distributorChannel null
>>> > partnerNames []
>>> > system Name Value
>>> > memoryMB 1861
>>> > cpu.count 4
>>> > cpu.vendor null
>>> > cpu.family null
>>> > cpu.model null
>>> > cpu.stepping null
>>> > cpu.extensions [hasMMX, hasSSE, hasSSE2, hasSSE3, hasSSSE3, hasSSE4A,
>>> > hasSSE4_1, hasSSE4_2, hasEDSP, hasARMv6, hasARMv7, hasNEON]
>>> > device.model GT-I9500
>>> > device.manufacturer samsung
>>> > device.hardware universal5410
>>> > device.isTablet false
>>> > os.name Linux
>>> > os.version 19
>>> > os.kernelVersion 3.4.5-3270240
>>> > os.locale en-US
>>> > hdd.profile.model null
>>> > hdd.profile.revision null
>>> > hdd.binary.model null
>>> > hdd.binary.revision null
>>> > hdd.system.model null
>>> > hdd.system.revision null
>>> > gfx.D2DEnabled null
>>> > gfx.DWriteEnabled null
>>> > gfx.adapters.[0].description Model: GT-I9500, Product: ja3gzs,
>>> > Manufacturer: samsung, Hardware: universal5410, OpenGL: Imagination
>>> > Technologies -- PowerVR SGX 544MP -- OpenGL ES 2.0 build 1.10 at 2359475
>>> > gfx.adapters.[0].vendorID Imagination Technologies
>>> > gfx.adapters.[0].deviceID PowerVR SGX 544MP
>>> > gfx.adapters.[0].subsysID null
>>> > gfx.adapters.[0].RAM null
>>> > gfx.adapters.[0].driver null
>>> > gfx.adapters.[0].driverVersion OpenGL ES 2.0 build 1.10 at 2359475
>>> > gfx.adapters.[0].driverDate null
>>> > gfx.adapters.[0].GPUActive true
>>> > settings Name Value
>>> > blocklistEnabled true
>>> > e10sEnabled false
>>> > telemetryEnabled true
>>> > locale en-US
>>> > update.channel nightly
>>> > update.enabled false
>>> > update.autoDownload true
>>> > userPrefs.browser.cache.disk.capacity 204800
>>> > defaultSearchEngine google
>>> > addons Name Value
>>> > activePlugins []
>>> > activeGMPlugins.gmp-gmpopenh264.version 1.3
>>> > activeGMPlugins.gmp-gmpopenh264.userDisabled false
>>> > activeGMPlugins.gmp-gmpopenh264.applyBackgroundUpdates 1
>>> > persona null
>>> >
>>>
>>>
>>>
>>> _______________________________________________
>>> Contentperf mailing list
>>> Contentperf at mozilla.org
>>> https://mail.mozilla.org/listinfo/contentperf
>>>
>>>
>>
>>
>> --
>> Benjamin Smedberg
>> Engineering Manager, Firefox
>>
>>
>> _______________________________________________
>> Contentperf mailing list
>> Contentperf at mozilla.org
>> https://mail.mozilla.org/listinfo/contentperf
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.mozilla.org/pipermail/contentperf/attachments/20150612/05caec9a/attachment.html>


More information about the Contentperf mailing list