slow slower slowest scroll on Android

Ralph DiMola rdimola at evergreeninfo.net
Wed May 29 17:50:11 EDT 2019


Jerry,

First of all... The armv7 will work on all architectures. The Intels have an
arm emulator that in my tests are about 7-10 times slower than a native
Intel build. Arm 64 will only work on 64 bit arm architectures. Intel will
work on all Intel architectures and Intel 64 will work on only 64 bit
Intels. (I think a have this straight)

Testing: If you push all 4 types to a test device the test device will use
the best architecture. This will make the apk huge and some devices have a
apk size limit. The build/install also takes longer. When testing use only
one architecture and pick the best for your test device.

PlayStore distribution: If you build 1 apk for all 4 architectures the apk
will be huge and you may exceed the max apk limit for some devices. The best
way is to build 4 apks one for each architecture. Up the version code by at
least 1 for each build. Upload and release all 4 apks to production on the
PlayStore. When a user installs your app from the PlayStore the best apk for
their device will be installed. This makes for a faster/less data install
for users.

The restrictions for multiple apks on the PlayStore is 1) the version codes
must be different 2) the apks must not target duplicate architectures. This
is why you can have 4 apks in production at the same time.

Ralph DiMola
IT Director
Evergreen Information Services
rdimola at evergreeninfo.net

-----Original Message-----
From: use-livecode [mailto:use-livecode-bounces at lists.runrev.com] On Behalf
Of JJS via use-livecode
Sent: Wednesday, May 29, 2019 5:10 PM
To: JJS via use-livecode
Cc: JJS
Subject: Re: slow slower slowest scroll on Android

Wow, what a difference does it make.

I saw that the setting was container again so i let it that way.

Turned off the nice border glow around the grid, turned border off and set
it to zero (the latter won't make difference i guess.)

This is how it should be. Thanks for pointing me to it Monte/Brian.

So in standalone i only now had the ARM64 selected and then build it in
dev/test mode.

(What will it push to the phone if all 4 options are selected? does it check
what is connected?)


This looks very good. Now i want to test further with the grid 
containing images and text fields, see how that performs.


Regards,

Jerry


Op 29-5-2019 om 21:24 schreef JJS via use-livecode:
> Thanks Brian,
>
> i will do some tests and report back.
>
> Regards,
>
> Jerry
>
> Op 29-5-2019 om 18:08 schreef Brian Milby via use-livecode:
>> DP2 will fix the missing option in the PI to set container mode, but 
>> it can be done from the message box now.
>>
>> If you need a static border or shadow behind the DG, then use a 
>> separate object for the effect behind the DG so you can get the 
>> performance boost from accelerated rendering.
>>
>> Thanks,
>> Brian
>> On May 29, 2019, 2:28 AM -0700, Sphere via use-livecode 
>> <use-livecode at lists.runrev.com>, wrote:
>>> Strange is, when i opened the stack in 950dp1 and looked at the setting
>>> to set static, dynamic, scrolling. Container was selected, but when you
>>> click on it you can only choose from the 3 previous mentioned and not
>>> container anymore.
>>> Unadorned (had to translate it first...) so no nice edge around it.
>>> Ok will test tonight.
>>>
>>> Thanks!
>>>
>>> Monte Goulding via use-livecode schreef op 2019-05-28 23:23:
>>>>> On 29 May 2019, at 5:49 am, JJS via use-livecode
>>>>> <use-livecode at lists.runrev.com> wrote:
>>>>>
>>>>> The DG2's where in container mode, now i've set them in scroller.
>>>>>
>>>> You should not fiddle with the layerMode of the datagrid. The scripts
>>>> already set the correct groups to container and scrolling. The actual
>>>> scrolling layerMode group in the datagroup is about 3 levels in with
>>>> its parents being container. As mentioned on other posts you need to
>>>> ensure the datagrid is unadorned (no border etc). Also if you have it
>>>> placed inside a group then that group must also be set to container
>>>> layerMode and be unadorned.
>>>>
>>>> Cheers
>>>>
>>>> Monte
>>>> _______________________________________________
>>>> use-livecode mailing list
>>>> use-livecode at lists.runrev.com
>>>> Please visit this url to subscribe, unsubscribe and manage your
>>>> subscription preferences:
>>>> http://lists.runrev.com/mailman/listinfo/use-livecode
>>> _______________________________________________
>>> use-livecode mailing list
>>> use-livecode at lists.runrev.com
>>> Please visit this url to subscribe, unsubscribe and manage your 
>>> subscription preferences:
>>> http://lists.runrev.com/mailman/listinfo/use-livecode
>> _______________________________________________
>> use-livecode mailing list
>> use-livecode at lists.runrev.com
>> Please visit this url to subscribe, unsubscribe and manage your 
>> subscription preferences:
>> http://lists.runrev.com/mailman/listinfo/use-livecode
>
> _______________________________________________
> use-livecode mailing list
> use-livecode at lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your 
> subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode

_______________________________________________
use-livecode mailing list
use-livecode at lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode





More information about the Use-livecode mailing list