really need answer not excuses

[pool_shark]

So from that picture you posted, what am I supposed to be seeing, that handcent also takes 0.0, so it also never uses any CPU?

OMG OMG, I just checked task manager on my computer, and guess what, google updates, chrome, avast, spybot, are all taking 0 CPU right now, that means they never use any CPU or power!

Since I'm not currently using it, it isn't using any cpu. If I send/recieve a text or mms it will change.
Until then it's just waiting.

At no point did I say the processes NEVER use any CPU, they will use CPU if they are being used, if they aren't, they won't.

 

[PSkeptic]

Show me where I told one person that it shouldn't bother them.
What I said was that I don't understand why it does.

As far as things running, if you ever have the opportunity, look on an AIX server. You'll see AIO running when you do a pstat -a, however AIO doesn't do a thing until there is a system call for it.
It's just waiting in ready mode at the bottom of the process list.

If any of you want proof of what's using cpu and what isn't, go to the command line and run the top command.
You will see countless apps running, most with 0.0 cpu.

Yes, and on AIX, AIO does run when it's called. However, Android is NOT AIX, and the system is designed differently. Apps sit and wait for a call, or map into memory their public functions that other apps can use.

And, they take up slots in the process table. And, they take up a chunk of RAM.

 

[pool_shark]

So, you do have components of the VCast Suite, right, that are preinstalled? CityID?

Come on, what are you arguing here? That VZW doesn't load about 50MB of bloat on all phones?

Sure, there is different bloat on all phones. My Ally didn't come with NFS:Shift... My Fascinate didn't have SongID. Both phones had about 50MB of bloat, taking up space.

I didn't say they don't install apps, I said there are different apps on different devices. When you started listing all of the apps I don't have I told you I didn't have them.
Mine came with blockbuster, NFS, cityID, amazon mp3, hotspot, quickoffice, swype, vvm, backup assistant, kindle, and youtube.

 

[pool_shark]

Yes, and on AIX, AIO does run when it's called. However, Android is NOT AIX, and the system is designed differently. Apps sit and wait for a call, or map into memory their public functions that other apps can use.

And, they take up slots in the process table. And, they take up a chunk of RAM.

Android isn't AIX? Really?
Whew, good thing I ran into you.

The point was that just because they are in the list of running processes, doesn't mean they are actually using any CPU.

 

[PSkeptic]

Android isn't AIX? Really?
Whew, good thing I ran into you.

Well, you're trying to equate daemons to running applications in different system architectures.

 

[pool_shark]

Well, you're trying to equate daemons to running applications in different system architectures.

Dammit! You're responding too fast, give me time to edit.

 

[Outlaw71]

For me it is a big deal as well. Sponsorship is generally done to bring a product to the consumer for free. In order to watch HBO I have to pay because there are no advertisers sponsoring it. But I can waych CBS for free because their programing is sponsored.

So I say if you're going to load my phone down with these sponsors that I can't get rid of, then refund the money I paid for the thing back immediately. If it's sponsored, then why am I paying for it?

When the iPhone comes out it won't have any sponsors on it, so why should my Android? I pay Verizon for minutes and data. All of that extra crap should be removable. Oh sure it is their device, so I'm not arguing that they don't have the right to put it on in the first place, indeed they do. But as soon as I paid for the thing I took ownership of it, therefore I should be able to remove any of it that don't want on it.

I don't care if they use battery, don't use battery or even save battery. I should have the ability to remove it regardless.

 

[Mayhem]

So basically it's completely irrational, got it.

It' a matter of preference.

 

[PSkeptic]

Android isn't AIX? Really?
Whew, good thing I ran into you.

The point was that just because they are in the list of running processes, doesn't mean they are actually using any CPU.

Ok, I slowed down a bit to let you add more, respond, etc.

You should know (Having an AIX background) that any process, using memory actively or not, used memory and CPU (Albeit, sometimes only sipping CPU cycles).

The Kernel needs to ping the process to see if it needs to wake up. The kernel pings the process to see if it can be swapped/remapped/etc. The kernel pings all of the processes to check if the other processes are still needed. The kernel needs to manage it's memory map.

So, even and unused, sleeping process uses resources. If it didn't, a fork bomb would be useless.

Why do you think people don't like having zombie processes hanging around? Enough of them, and your machine starts to crawl.

Android is based on the linux kernel, and it's memory management is well documented. Yes, it handles memory a little better than stock Linux. But, zombie processes still use system resources.

 

[pool_shark]

Ok, I slowed down a bit to let you add more, respond, etc.

You should know (Having an AIX background) that any process, using memory actively or not, used memory and CPU (Albeit, sometimes only sipping CPU cycles).

The Kernel needs to ping the process to see if it needs to wake up. The kernel pings the process to see if it can be swapped/remapped/etc. The kernel pings all of the processes to check if the other processes are still needed. The kernel needs to manage it's memory map.

So, even and unused, sleeping process uses resources. If it didn't, a fork bomb would be useless.

Why do you think people don't like having zombie processes hanging around? Enough of them, and your machine starts to crawl.

Android is based on the linux kernel, and it's memory management is well documented. Yes, it handles memory a little better than stock Linux. But, zombie processes still use system resources.

Zombie processes are different than sleeping processes. Zombies can actually spend time trying reconnect to their parent and go nuts.
But you can have a ton of defunct processes that were incorrectly terminated from their parent and sit there using no cycles whatsoever.

Sleeping processes do get cycles, but if you look at the average amount of cpu time used since those processes were started you will see 0.0.
That means that even though they do get cpu cycles, the time spent on those processes are at a minimum which isn't hurting performance in any way.

 

[Outlaw71]

Ok, well at one point I felt like an equal participant in this discussion, but now you two are talking so far over my head that I've decided to sit in the corner and hum the theme songs to some of the more popular 80's sitcoms.

Let me know if/when one of those zombie processors ever bites the head off of one of the sleeping ones, okay.







........ da dant da... da da da. Dunna da da dant da... da dunna da.....

 

[PSkeptic]

Zombie processes are different than sleeping processes. Zombies can actually spend time trying reconnect to their parent and go nuts.
But you can have a ton of defunct processes that were incorrectly terminated from their parent and sit there using no cycles whatsoever.

Sleeping processes do get cycles, but if you look at the average amount of cpu time used since those processes were started you will see 0.0.
That means that even though they do get cpu cycles, the time spent on those processes are at a minimum which isn't hurting performance in any way.

If a process takes up CPU cycles, guess what: It impacts performance. That's the bottom line. It reduces battery life, and slows down the phone.

Something many people do not tolerate. Myself included. I want my device to be operating at it's peak, not slowed down by things that don't have to be there.

And, sleeping processes DO use CPU cycles. They have to, in order to be in the process list!

Ok, well at one point I felt like an equal participant in this discussion, but now you two are talking so far over my head that I've decided to sit in the corner and hum the theme songs to some of the more popular 80's sitcoms.

Let me know if/when one of those zombie processors ever bites the head off of one of the sleeping ones, okay.







........ da dant da... da da da. Dunna da da dant da... da dunna da.....

It's ok. Zombie processes only want to eat your RAM

 

[Outlaw71]

It's ok. Zombie processes only want to eat your RAM

Ah ha... well lets talk about more pressing matters, shall we? Did you recognize the theme song I sang there at the bottom of my post? It's the 'A-Team'! Now go back and read it with that jingle in your head.... you can see it now, can't ya!

 

[pool_shark]

If a process takes up CPU cycles, guess what: It impacts performance. That's the bottom line. It reduces battery life, and slows down the phone.

Something many people do not tolerate. Myself included. I want my device to be operating at it's peak, not slowed down by things that don't have to be there.

And, sleeping processes DO use CPU cycles. They have to, in order to be in the process list!

You are incorrect. Processes are supposed to use CPU, but just because they do does not mean performance is IMPACTED.
Here is the output of the sar command on one of my servers.

13:02:18 %usr %sys %wio %idle physc %entc
13:02:21 0 6 0 93 0.02 18.1
13:02:24 0 5 0 95 0.01 13.8
13:02:27 0 4 0 95 0.01 13.2
13:02:30 2 6 0 92 0.02 17.9
13:02:33 0 4 0 95 0.01 13.2


Average 1 5 0 94 0.02 15.3


The formatting didn't work, but there is 94% idle time, is my server IMPACTED because the cpu is being used, of course not.


If the average amount of time a process uses the CPU is 0.0 it means that process for the entire time it was been running hasn't done very much at all, that's why if you run the top or sar command on your device you will see how much idle time there is on the system.
That will let you know just how much work your device is actually doing.

...and if you were paying attention I did say that yes, sleeping processes use CPU cycles, but it is very minor which is why the average amount of time reads 0.0.

 

[PSkeptic]

Ah ha... well lets talk about more pressing matters, shall we? Did you recognize the theme song I sang there at the bottom of my post? It's the 'A-Team'! Now go back and read it with that jingle in your head.... you can see it now, can't ya!

No that you said it, ayep

 

[PSkeptic]

You are incorrect. Processes are supposed to use CPU, but just because they do does not mean performance is IMPACTED.
Here is the output of the sar command on one of my servers.

13:02:18 %usr %sys %wio %idle physc %entc
13:02:21 0 6 0 93 0.02 18.1
13:02:24 0 5 0 95 0.01 13.8
13:02:27 0 4 0 95 0.01 13.2
13:02:30 2 6 0 92 0.02 17.9
13:02:33 0 4 0 95 0.01 13.2


Average 1 5 0 94 0.02 15.3


The formatting didn't work, but there is 94% idle time, is my server IMPACTED because the cpu is being used, of course not.


If the average amount of time a process uses the CPU is 0.0 it means that process for the entire time it was been running hasn't done very much at all, that's why if you run the top or sar command on your device you will see how much idle time there is on the system.
That will let you know just how much work your device is actually doing.

...and if you were paying attention I did say that yes, sleeping processes use CPU cycles, but it is very minor which is why the average amount of time reads 0.0.

Does you server run on battery? Especially one that is as constrained as a phone's battery?

Impact on a server vs a phone is huge difference. As long as a user is serviced in a reasonable time, then a server is considered "Not impacted". A phone, however, has to juggle many RTOS operations. Every process in the process list impacts how responsive each of those RTOS operations respond, not to mention it adversely impacts battery life.

Especially the bloat that send out data requests (Which most do). Every packet out of a phone drains the battery.

 

[pool_shark]

Does you server run on battery? Especially one that is as constrained as a phone's battery?

That isn't the point, you said just having it there is impacting performance which is an untrue statement.

Impact on a server vs a phone is huge difference. As long as a user is serviced in a reasonable time, then a server is considered "Not impacted". A phone, however, has to juggle many RTOS operations. Every process in the process list impacts how responsive each of those RTOS operations respond, not to mention it adversely impacts battery life.

Essentially it is the same. If the processes aren't getting very much CPU they are not the cause of your performance issues. Your time would be better spent on the processes that are getting more time.

Especially the bloat that send out data requests (Which most do). Every packet out of a phone drains the battery.

Look at my post with the pic.
It shows NIC 0.0, which means nothing was going out.
That was from the output of top -n3 -d5.
It wasn't just a snapshot for a 1 second interval.

I'm done with this argument because I've offered proof to back up my statements, you just keep offering statements.
Have a nice day.

 

[PSkeptic]

Look at my post with the pic.
It shows NIC 0.0, which means nothing was going out.
That was from the output of top -n3 -d5.
It wasn't just a snapshot for a 1 second interval.

I'm done with this argument because I've offered proof to back up my statements, you just keep offering statements.
Have a nice day.

You're posting "proof" from a server, not from an Android device... Which really isn't proof at all.

If all the bloat on a phone didn't impact performance and battery life; why pray tell does a phone with a de-bloated ROM perform better, and have better battery life?

As for your screen-shot there, are you telling me that kjournald has no impact on performance? Because it shows 0.0% usage. And, everyone knows is kicks into gear whenever there is a write action.

 

[pool_shark]

You're posting "proof" from a server, not from an Android device... Which really isn't proof at all.

If all the bloat on a phone didn't impact performance and battery life; why pray tell does a phone with a de-bloated ROM perform better, and have better battery life?

The picture in my above post was from my Droid 2.

How can you prove that you get better performance or battery life than I do since we most likely aren't doing the same things on our device?

 

[PSkeptic]

The picture in my above post was from my Droid 2.

How can you prove that you get better performance or battery life than I do since we most likely aren't doing the same things on our device?

Oh that's easy. Controlled testing.

I can bet that if I were to take a Droid 2, and run the same battery of tests on it stock, and then run the same battery against a debloated unit; the debloated unit would be snappier, and have longer battery life.

I did that with my Samsung Fascinate. I got better battery life, and a snappier system once it was debloated.

 

[pool_shark]

I started blockbuster, skype, visual voicemail, and cityID.
Ran top again, and as you can see only skype is even in the process list even though in manage applications all of the above were listed as running.

 

[Outlaw71]

I can do 50 push-ups.

 

[pool_shark]

I can do 50 push-ups.

Your phone have video?
Prove it.

 

[Outlaw71]

Your phone have video?
Prove it.

* * * G*U*L*P* * *



Hey I know, hows about I put up a video of someone demonstrating what I look like when I do 50 push-up?

 

[pool_shark]

* * * G*U*L*P* * *



Hey I know, hows about I put up a video of someone demonstrating what I look like when I do 50 push-up?

That might work.
If you post a video of someone else doing 50, I'll give you credit for 25.