ART vs. Dalvik cache: Are you seeing a performance difference?

[patruns]

Interestingly, I do not find it on the Kit Kat update for the Nexus 10. Anyone else?

 

[Shilohcane]

Interestingly, I do not find it on the Kit Kat update for the Nexus 10. Anyone else?

I went looking for ART to test out on my 2012 Nexus 7 that is Quad-core Tegra 3 processor and it isn't there either. It is a shame since the 2013 Nexus 7 could use better battery life.

This thread thinks;

Apparently ART is only available for Snapdragon-based devices. I can switch the runtime on N4, N5 and new N7 which are all based on Snapdragon but not on the first gen N7 (Tegra 3) and N10 (Exynos). It is plausible to make such restriction (because it can break the system) if the devs aren't tested it enough on devices other than Snapdragon.

 

[patruns]

Interesting. I am not sure why that is. I would not think that a run time would be processor dependent.

 

[stackberry369]

Interesting. I am not sure why that is. I would not think that a run time would be processor dependent.

Where do I find this on my note 3?

Posted via Android Central App

 

[patruns]

Where do I find this on my note 3?

Posted via Android Central App

In devices with Kit Kat that have it, it is in settings/developer options/about the 4th down "select runtime"

 

[devilsfl]

I think this is only true for photos taken directly from Google backup. I can upload just fine if the pic is local and and I'm not using the Photos app.

Sent from my Nexus 5 using Tapatalk

Ah okay, I'll check that out, thanks. I switched back to Dalvik because it was too annoying. But I don't think it worked whether I was starting out from the app or from Photos. I'll check from the gallery.

 

[oplix]

edit:

 

[downhillrider]

i switched to art today and i noticed my phone getting warm. which it really never did in the couple of weeks ive had it. will give it another day or 2

 

[wsrobert]

Can someone give me a brief rundown of the difference between art and dalvik? I just switched to art on my moto x and I can definitely feel a difference switching between apps that are open. But if the app is opening from a cold state it seems to definitely take longer than with dalvik. Facebook is painfully slow for example. I tend to close apps after I'm done using them...is that dumb? Is android efficient enough to not need to do that? If so, than I'll definitely stay on art!

Posted via Android Central App

 

[smoledman]

The problem for me is that ART killed Whatsapp, so I had to go back to Davlik.

 

[downhillrider]

what kind of problems? ive been using whatsapp all day and havent noticed anything.

 

[sethdaniel2011]

I tend to close apps after I'm done using them...is that dumb? Is android efficient enough to not need to do that? If so, than I'll definitely stay on art!

In short, yes. Longer explanation follows. Note: I know you didn't say you used a task killer, but much of the logic here remains unchanged whether killing an app through a task killer or through the native android system or any other way.

There was a proliferation of task killers on Android in the early days because people carried over thinking from Windows that they need to kill apps. I think around version 4.0 it actually became impossible for one app to close another, but task killers still exist because people still think they need them (they're purely placebo, or worse - malicious). A simple google search for "task killer android" will surface some stupid apps, but also many articles about why they are bad. Here's a good summary:

When you leave an Android app, going back to your home screen or switching to another app, the app stays “running” in the background. In most cases, the app will be paused in the background, taking up no CPU or network resources. Some apps will continue using CPU and network resources in the background, of course – for example, music players, file-downloading programs, or apps that sync in the background.

When you go back to an app you were recently using, Android “unpauses” that app and you resume where you left off. This is fast because the app is still stored in your RAM and ready to be used again. Killing a task will regain the used RAM for you, but that's not necessarily a good thing. Empty RAM is useless. Full RAM is RAM that is being put to good use for caching apps. If Android needs more memory, it will force-quit an app that you haven’t used in a while – this all happens automatically, without installing any task killers.

 

[19h_rnn]

It increased my battery life by 20% now my phone can last a day and a half
Besides being power friendly, it also increased the speed of my phone. The transition between apps and the long lags have improved and disappeared
However it does eat up large amounts of my storage space.
But its a very nice virtual machine and I do recommend it if you want a more productive phone and a bit of curiosity.

 

[npark2012]

What exactly is ART and Dalvik? If someone could explain in simple terms.

 

[HNNNNNGHHH]

What exactly is ART and Dalvik? If someone could explain in simple terms.

ART and Dalvik are two different runtimes currently for android, the latter being the one used for android for years but will be replaced with ART in the coming years.

 

[HNNNNNGHHH]

So far, ART has improved my battery life tremendously but be warned that you cannot use ART on a custom ROM without flashing ART-compatible gapps.

 

[Ted Leung]

Just switched to art on my nexus 4. Not seeing any significant speed improvements. Lag of the switcher button is still there.

Sent from my Nexus 4 using AC Forums mobile app

 

[physioprof]

Opening and switching apps seems faster on my Nexus 7 2013.

Sent from my Nexus 7 using AC Forums mobile app

 

[johnathin15noob]

So far, ART has improved my battery life tremendously but be warned that you cannot use ART on a custom ROM without flashing ART-compatible gapps.

what do you mean "cannot"... with what consequence?. i'm running a custom rom and am using ART. i didn't check if my gapps were ART-compatible first.

 

[HNNNNNGHHH]

what do you mean "cannot"... with what consequence?. i'm running a custom rom and am using ART. i didn't check if my gapps were ART-compatible first.

Actually depends on the device but ART has been known to cause gapps to FC if the gapps package you're using isn't optimized for ART.