Froyo 2.2 to bring 450% performance increase?

[baller827]

danggg, I just rooted my droid yesterday, overclocked it today and am completely amazed at how it performs. to think now that it's gonna be going 450% better along with a flash player?

 

[tomato88]

I'm completely sold to this info. Now where is 2.2?

 

[Doubledee90]

lol.he.broke.his.spacebar.

I.had.the.exact.same.problem.many.years.back.and.I.typed.like.that.too.

A 450% increase would be amazing, but I highly doubt it is accurate. None the less, I shall be waiting on pins and needles for the release.

No, it just made what I wanted to say more intense

 

[inssane]

A 450% performance increase is possible because the majority of Android apps are built to run on a davlik java virtual machine, which has advantages, but also limits the capabilities of the apps. 3D games such as ExZeus are an exemption.

The new virtual machine is much more efficient, makes use of just in time compilation, etc, and allows for certain apps (such as the Linpack benchmark) see these major performance increases.

Sent from my Droid using Tapatalk

+1
I have read amazing things about Dalvik. 450% seems optimistic, but in Linpack, the top ten include Nexus Ones that are in the high 30 and one at 40 mflops. There is a device at the top with 80mflops but it looks like a virtual machine off of a PC.
If a few N1s are officially there and there was no tom foolery, we should be in for a nice ride with 2.2.

 

[Optimus]

When 2.2 comes out looks like I'm finally going too root. I'm counting on cyanogen .

 

[Fadelight]

When 2.2 comes out looks like I'm finally going too root. I'm counting on cyanogen .

So you are willing to handle dog poop until 2.2 comes out? Why not root now and be 500% more happy with 2.1 while you wait?

 

[OMJ]

So you are willing to handle dog poop until 2.2 comes out? Why not root now and be 500% more happy with 2.1 while you wait?

I love the analogy Fade.

 

[ssantan2]

I wondering how much of that 450% speed increase we will actually notice day to day?

 

[gonzoguy24]

I just hope everything gets a bit snappier at stock and apps open and close a bit smoother. For instance opening a large 3D game like Backbreaker Football takes foreverrrrr

 

[nader2013]

I wondering how much of that 450% speed increase we will actually notice day to day?

That is 450%! I'm guessing this will be a tremendously noticeable difference.

 

[riffdex]

i dont know what mflops are, but is froyo expected to increase battery life for android devices?



Sent from my Droid using Tapatalk

 

[johnnybirdman]

i dont know what mflops are, but is froyo expected to increase battery life for android devices?
Sent from my Droid using Tapatalk

I hope so.
And ever since 2.1 update my custom ringtones don't 'fire' like they did before. It's like the phone isn't fast enough to match the number to the tone and all I get is the default ringtone even if a custom one is set.

More..faster..better.

J.

 

[sorceror]

i read that too, but find a 450% performance increase hard to believe. very skeptical.

That's 450% for a specific, cpu-bound task. LINPACK is a test of number-crunching - it's doing a lot of the same calculations, over and over, in tight loops. JIT (Just In Time) compiling is great for this - it identifies places in an application where a lot of work is going on, and converts those spots to native code.

Java normally is interpreted. A developer writes a program, and this is compiled into "byte-code" - an intermediate form that the Java virtual machine executes. (Dalvik on Android.) So, when executing Java, there's the overhead of examining that bytecode, and then telling the underlying hardware (ARM CPU on Android) what to do. JIT bypasses that - it "pre-converts" the bytecode directly into 'native' code (in this case, direct ARM CPU instructions). Thus, no need to 're-interpret' the code each time it's run.

For most typical applications, they won't see that kind of speed boost. But most applications do spend most of their time in small parts of their code. (The old 80-20 rule, 80% of the time in 20% of the code on average.) Speeding up just those heavily-used parts can have a big impact on speed. Not 450% - that's probably the best possible case - but I'd believe a typical app would see at least a 50% speed boost, and most would probably see 100% or better.

Two caveats: First, this won't make the network any faster. If the browser's waiting on a website to give it data, JIT won't make any difference. Second, a lot of games are already compiled to native code for speed - JIT won't make any difference for them.

 

[Fabolous]

Very well put, informative post!

 

[virtus]

So is this the same JIT that is already available on rooted phones?

Also, I seem to recall reading that Turbo Davlik sacrificed memory for speed. If this is the case, I can see it working great on the N1 but not so great on the Droid.

 

[sorceror]

So is this the same JIT that is already available on rooted phones?

Also, I seem to recall reading that Turbo Davlik sacrificed memory for speed. If this is the case, I can see it working great on the N1 but not so great on the Droid.

It's probably close, but not exactly the same. Dalvik has had some (disabled) JIT support, but I assume Google's done some testing, polishing, and tuning in preparation for turning it on in Froyo.

And yes, there are indeed tradeoffs. Reading my above explanation, people might ask, "why not just compile everything to native code"? But native code often uses more memory than Java bytecode, which can end up slowing things down when stuff can't fit into RAM.

So there's a balancing act, and JIT tries to find the best tradeoff between memory space and execution speed.