cordova-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Andrew Grieve <agri...@chromium.org>
Subject Re: Measuring Perf of Cordova apps (Android/iOS)
Date Mon, 20 Oct 2014 18:27:00 GMT
I was thinking you could just write a post that links to your
already-excellent post here:
http://blog.nparashuram.com/2014/10/measuring-rendering-performance-metrics.html

You'd be the first post to be re-posted on the Cordova blog, so you'd need
to do some work to figure out what format's best, but I think pretty much
anything would do :P

On Wed, Oct 15, 2014 at 1:54 PM, Parashuram Narasimhan (MS OPEN TECH) <
panarasi@microsoft.com> wrote:

> @Michal - I tested the WkWebView using a simple iOS App and loaded a
> webpage on it - not Cordova yet and that was the reason I did not publish
> the results. Can't wait for WkWebView to be in Cordova and then see how
> much of an improvement we could get :)
>
> For the metrics, the idea was not to compare iOS vs Android, but rather
> compare 2 deploys of the same app. This idea is, a developer runs the tests
> per commit (in a continuous integration system) and can see which commit
> slowed down the app. This way, they could also figure out the CSS or JS
> event handlers they added during that commit. The graphs would look
> something like http://nparashuram.com/perfslides/perfjankie - you can see
> how the 3rd commit on that graph is bad since I added an onScroll event
> handler.
>
> @Andrew - I would love to do a blog post. Should I write it on my blog and
> refer to it in the official Cordova blog, or can this be in the official
> cordova blog? Note that I developed this tool independently and is not an
> official Cordova release, but if the folks in the community feel that this
> could be interesting to any Cordova developer and makes sense to put in the
> official Cordova blog, I would love to write a quick tutorial.
>
> -----Original Message-----
> From: agrieve@google.com [mailto:agrieve@google.com] On Behalf Of Andrew
> Grieve
> Sent: Tuesday, October 14, 2014 11:48 AM
> To: dev
> Subject: Re: Measuring Perf of Cordova apps (Android/iOS)
>
> Super Cool. This would definitely be useful to add to a continuous build
> system. Certainly would make a good referral blog post if you want to have
> a stab at it :)
>
> On Tue, Oct 14, 2014 at 2:28 PM, Michal Mocny <mmocny@chromium.org> wrote:
>
> > Very cool Parashuram.
> >
> > How did you test WkWebView?  non-cordova?
> >
> > Also, the results have different labels on iOS/Android, which I assume
> > means you are just reporting what the tools report.  Would you
> > consider mapping the values which have direct parallels to the same
> > labels so we can do more direct comparisons?
> >
> > -Michal
> >
> > On Mon, Oct 13, 2014 at 2:47 PM, Parashuram <n.parashuram@gmail.com>
> > wrote:
> >
> > > Over the weekend, I created a tool to measure the rendering and
> > > network performance for Cordova apps, and I thought of sharing it on
> > > the list. I
> > am
> > > able to pick up numbers from iOS and Android (4.4).
> > >
> > > Here is a demo video - https://www.youtube.com/watch?v=TG_eTe_H-s4
> > > More details on how to run it on your cordova app : [link <
> > >
> > http://blog.nparashuram.com/2014/10/measuring-rendering-performance-me
> > trics.html
> > > >
> > > ]
> > > Browser-perf : http://npmjs.org/package/browser-perf
> > >
> > > Would love to hear feedback from the community on this. Initial
> > > tests
> > with
> > > WkWebview and UIWebView show some good numbers.
> > >
> > > Since a lot of people worry about perf of hybrid apps, I am hoping
> > > that this tool would give them a way to look at hard numbers and see
> > > that
> > hybrid
> > > apps can be as smooth as native apps.
> > >
> > > Parashuram
> > > http://nparashuram.com
> > >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@cordova.apache.org
> For additional commands, e-mail: dev-help@cordova.apache.org
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message