Performance testing a Flex client - performance

While I have used performance test tools for a number of years now, they all operate at the network level, recording traffic between the client and server.
Most of the time, managers are interested in the performance that the end user sees.
With RIA applications such as Flex, part of the performance that the user sees is comprised of calculations etc on the client desktop itself. But most tools do not record that time.
Some RIAs include a profiler, but the profiler is only simulating one client.
Has anyone found a good solution to this? The best idea I can come up with is to have a network level tool (e.g. Neoload) creating load on the server, then using either a profiler or some GUI-automation to record full client times. This seems a bit clunky though.
Thanks,
Gordon

There's a larger number of tools available for testing the overall performance of a Flex application, and the majority focus on network communication performance (mostly AMF for larger apps) and the load tests for the backend.
Here's a short list of some tools or services:
http://www.spoon.as/ecosystem/testing-load-and-performance/
http://www.genilogix.com/products-hp-performance.aspx / Article: Performance Testing Flex Applications using LoadRunner
When you are testing a client application, there are a number of parameters to test:
Application load and initialization time
Framerate
Memory consumption
Impact of garbage collection / System.pauseForGCIfCollectionImminent() API
Check if application is using hardware acceleration (GPU) (can improve graphics performance drastically, but has some drawbacks as well)
Here are some links which should be useful:
Flash Builder Profiler (commercial)
Adobe TV: Measuring Framerate Performance
Adobe article: Garbage collection internals for Flash Player and Adobe AIR
Providing Hints to the Garbage Collector in AIR 3
DemonsterDebugger Great for debugging and CPU profiling
The Miner (formerly Flash Preload Profiler) / The Miner documentation
Adobe Max 2011 demo of Monocle profiler Adobe is working on Monocle, a much improved profiler, which was demoed at Max 2011. Release is planned for the end of 2012 (Flash Player "Dolores")
Adobe TV: Concurrency in Flash Runtimes (starting with Flash Player 11.4 beta)
Adobe TV: Performance-tuning Flex Mobile Applications
Stage3D hardware acceleration
Much depends on what kind of devices you are targeting, or which features you are using, e.g.
Mobile vs desktop deployment
Do you load large amounts of data
Real-time communication requirements
Number of visual items or components
Number of animations / items animated
Screen resolution (e.g. building applications for Smart TVs with very high resolution)
Do you use HD video playback
Video streaming - which quality and resolution are you using

There is an extension for Selenium to test Flex UI:
http://code.google.com/p/flex-ui-selenium/

try to use monocle....is really powerfull

Related

Show performance/system logs on the Microsoft Mesh application

I've been playing around with Microsoft Mesh (a remote AR/VR/MR collaboration solution for Teams) for a few days. The thing I want to know is if it’s possible to log some performance/system indicators when using the application. For example, frames per second, network I/O, CPU/Memory usage.
Since it runs on only HoloLens 2 (for now), whose OS is a modified version of Windows and CPU is ARM-based, it seems difficult to grab detailed information as in Linux where almost everything is open-sourced (I'm a newbie in Windows-related development).
I'm wondering if there are some APIs for these logs (whether system level or application level). Thank you all!
You can use the Performance Tracking in the device portal, refer to Using the Windows Device Portal - Mixed Reality | Microsoft Docs, you can refer to Performance tracing and analysis with WPA - Mixed Reality | Microsoft Docs for further analysis of this data, and you can also get this data via Device portal API reference - Mixed Reality | Microsoft Docs.

How can i maintain Agora.io video call memory usage

Platform Unity 2019.4.8.f1
Apk tested on Android devices
Api lvl 21+
Test failure on some devices that have 2gb memory.
Our team has implemented a unity project and we use agora for video call. Our streaming configs are very low, 15fps, 65bit. But when we test the app on phones that have 2gb memory, some times later(about 5-10min) app crashes. We try to find out the problem and according to diagnostic tools agora uses lots of memory, We cannot find any limitation from agora sdk and cannot reduce memory usage, is there any idea about this. Thanks.
First, did you make sure if this is an Application level memory management issue? e.g., you created instance of renderers but did not release in time?
If it is SDK level issue, you probably won't find a solution from the public. I suggest you to create a ticket to ask the question directly to the Agora support team. Please provide your quantified analysis and the crash log when you submit that.

My website pictures loading are slow in the server and failing google mobile testing

My website pictures loading are slow in the server and failing google mobile testing- In google mobile testing pass my website but some of the (18 pictures) are loading slowly. This is basic hosting I am running my website. Can anyone helps on this(suggestion and recommendation). Thanks.
Generally you can use Photoshop or an online service to optimize your pictures, there are also free optimization services up to certain limits.
This question is a little bit old, but can quide you in the right direction. You can also have look here.
Personally I do not have Photoshop and I use Kraken free service.
The optimization can be lossy or lossless, I prefer lossless (without a visual decrease in quality).

Google PageSpeed: Big gap between Desktop and Mobile

I'm trying to improve/optimize my website performance and cannot make it work.. And I running out of ideas.
This is my Site in Desktop with 99/100 score.
And this is my website performance in Mobile, with 40/100 score.
There are some big differences in server stuff, no idea why or how to improve this..
Thanks.
Edit: Site URL: creatufrase.net
The main big difference in the new Google PageSpeed is on the way how the Desktop and Mobile is tested, where on Mobile is applied internet connection limits and slowed down CPU.
Currently, Lighthouse (Google PageSpeed) simulates a page load on a mid-tier device (Moto G4) on a mobile network. With these following network limits:
Latency: 150ms
Throughput: 1.6Mbps down / 750 Kbps up.
Packet loss: none.
As well, the mobile device is with slowed down CPU power.
In this case, if your website is JavaScript heavy then lower CPU powered mobile devices parse and process JavaScript much slower and the network limits slow down overall speed if you have a large website with large images or other elements.
As I can't find official confirmation, I'm guessing that desktop tests are tested without these limits and that for some website could make a huge difference between mobile and desktop data in the new Google PageSpeed tests. Especially, if your website is a bit with more JS or contains heavy images.
For confirmation about these limits I did a quick test using Lighthouse Audit tool in the Google Chrome Dev Tools and used for mobile network and slowed down CPU limits, but for desktop didn't apply any limits and I got similar timing results as in Google PageSpeed mobile vs desktop.
Based on all this information, have a closer look at your script usage under Reduce JavaScript execution time. Start with reviewing ad/tracking scripts, Add This widget and other 3rd party scripts.

How will web programming change as tablet PC's become more prevalent?

In case you've been living under a rock, you may have noticed that Apple introduced a tablet PC. HP, Microsoft, Google, and others aren't too far behind. A lot of discussion and thought has been put into how a user would interact with traditional applications in a large touch screen environment, but how do you think this will affect the user interface of traditional web programming? What do you think will have to change with our current HTML controls? What new kinds of controls will be possible if we have touch-and-drag technology on our web pages?
EDIT: In response to some of the comments and answers, I would argue that this is fundamentally different than either mobile phone technology or previous iterations of tablet PCs. Whether you love or hate Apple, the sales of the iPad have demonstrated a significant consumer interest in tablets that hasn't been present before. When the number of users of tablets reaches critical mass - which it eventually will - designers will have to take notice and adapt our web sites accordingly.
Having a large-size tablet computer is also a significant departure from touch screen smart phones. Development of web pages for mobile devices has generally been optimized for restrictions of the device such as low bandwidth, absence of functionality (Javascript for many phones), and reduced screen resolution. Multi-finger touch based tablets will provide a host of opportunities to improve user interaction.
The same effect is visible on mobiles that have suported html for a while.
Its basic case of controls degrading to provide click/touch features instead of mouse hovers.

Resources