Listen, I'm not professing that I know how to design an OS. I'm just noticing that certain OSes do a better job than others of keeping the windows manager fluid and smooth. I don't know if you are a designer for Gnome and taking this personally or what, but it just seems to me that there are better ways of getting it done then what linux window managers are doing.
How much CPU should it take to draw a window? Next to nothing, so the OS should be able to get it out of the way first. If it takes that much CPU to draw windows then there is a huge flaw in the window manager and that's not anyone else's fault.
These are preemptively scheduled like any other processes, and are not typically CPU bound.
Then that is the problem. A window trying to draw should be the highest possible priority on desktop workstation and should not be treated like other processes. There is good reason to keep other processes waiting while I/O completes but window drawing operations should never be kept waiting. I suspect you have hit the nail on the head.
I think most people here understand that. The thing is, ultimately any given distro makes the decision to bundle with a certain window manager and often people are stuck with that window manager. Using myself as an example, I am very knowledgeable but this is a system I have to use every day, day in and day out. I use software day in and day out that is tested to work with this window manager. If I attempt to migrate to 'window manager x', not only do I need to spend my own time doing it but I will be almost assured of running into issues here and there that may not be insurmountable but will take even more of my valuable time.
So that is why for all intents and purposes. here in the real world, an OS distro might as well be coupled with the kernel and the window manager.
The linux laptop I use comes pre-built and unfortunately I don't have the flexibility to modify it in that way. I'm stuck with Gnome and whatever scheduler the OS comes with. If there are schedulers that are much better, I would question why the particular distro I am using wouldn't just pick the one that works better.
I've noticed on my machine that applications take longer to open, and windows stay in the 'partially drawn' state for a longer time. Also, it takes windows 7 longer to navigate through contents of folders in explorer. That said, Windows is still my preferred OS to use personally and I don't find it all that inconvenient, but that is just what I have noticed.
In my experience, Linux desktop response suffers way more heavily under high disk load then Windows desktop response. Something with the way Gnome and KDE are prioritized in the kernel loop I would expect. Run something in the background that is chewing up the disk and expect windows to draw very slowly.
Doesn't work. Now you are dealing with two devices and a cable. Too cumbersome on a trail. Personally I spend a lot of time sitting at the side of a swimming pool these days for swim meets and I don't want to deal with two devices there either.
Let me add, a USB power pack doesn't work on a hike either. I wouldn't even use a USB power pack and I am considerably more tolerant than my friend. Say you are walking along a trail and you want to check your phone, or take a picture. Now you have a cord dangling off your phone so you are not just grabbing it out of your pocket, it is in your bag with a cord attached. So you have to use both hands now and either pull out phone with power pack or go into your bag and disconnect the pack, hoping you have had the phone on there long enough for it to charge enough while the GPS is going. It's just not convenient at all.
I find battery stats to be quite wrong in most cases. For one thing it is a known fact that smartphones drain the battery faster just when you are in an area with a weak or slow connection. So many variables involved it's a bit rediculous for any phone maker to make a claim on their battery life. Therefore I cannot rely on that number when dumping close to $1K on a handheld device.
The specific conversation was that my friend wanted to track a hike he was going on with his son. I asked him why he doesn't just use a GPS tracking app on his phone and he said he worried that the phone battery wouldn't last all day. Being an Apple fan, he likes his phone to be thin and lean so he would never consider attaching a battery pack.
Question: Does Apple specifically say that a user is allowed to open the device? Because I think if not I wouldn't open it. Definitely not if it was still on warranty.
I'm curious how people are finding that a battery pack is the solution, when:
1) Usually the battery dies when you are using the phone, so you can't just put it aside at that point and charge it. I can use my phone while it is on a battery back but then it hardly charges and i'm stuck with the pack.
2) A battery back is physically bigger than a phone battery and most require an extra cable. How is that ever going to be more convienent then just carrying another charged battery?
Sounds like a general phone design issue as opposed to a removable battery issue. On my phone there is no individual door, the whole back comes off and there is no risk of it just coming off. Furthermore, the back is usually under the case that I have that clips tightly onto the phone.
Slashdot Mirror
So let's just leave it as you are one person who had bad luck.
It's really not that difficult to change a battery in a phone that is made to do it, no matter where you are.
Listen, I'm not professing that I know how to design an OS. I'm just noticing that certain OSes do a better job than others of keeping the windows manager fluid and smooth. I don't know if you are a designer for Gnome and taking this personally or what, but it just seems to me that there are better ways of getting it done then what linux window managers are doing.
That's not because the window drawing has higher priority. That's because the system isn't ready when it asks the window to draw.
How much CPU should it take to draw a window? Next to nothing, so the OS should be able to get it out of the way first. If it takes that much CPU to draw windows then there is a huge flaw in the window manager and that's not anyone else's fault.
Issues I am having are with Gnome.
Let me put it in simple terms. If I do anything to fuck up my laptop, even for a couple hours, I'm totally screwed.
These are preemptively scheduled like any other processes, and are not typically CPU bound.
Then that is the problem. A window trying to draw should be the highest possible priority on desktop workstation and should not be treated like other processes. There is good reason to keep other processes waiting while I/O completes but window drawing operations should never be kept waiting. I suspect you have hit the nail on the head.
Actually you might have a point. If I hadn't already commented I would mod you up.
If you can get your company to allow you to use a VM engine that is any good, and give you a machine with enough resources to run it.
I think most people here understand that. The thing is, ultimately any given distro makes the decision to bundle with a certain window manager and often people are stuck with that window manager. Using myself as an example, I am very knowledgeable but this is a system I have to use every day, day in and day out. I use software day in and day out that is tested to work with this window manager. If I attempt to migrate to 'window manager x', not only do I need to spend my own time doing it but I will be almost assured of running into issues here and there that may not be insurmountable but will take even more of my valuable time.
So that is why for all intents and purposes. here in the real world, an OS distro might as well be coupled with the kernel and the window manager.
The linux laptop I use comes pre-built and unfortunately I don't have the flexibility to modify it in that way. I'm stuck with Gnome and whatever scheduler the OS comes with. If there are schedulers that are much better, I would question why the particular distro I am using wouldn't just pick the one that works better.
GUIs are also for old tired unix admins that grow weary of typing on the command line all day and just want to play a movie.
I've noticed on my machine that applications take longer to open, and windows stay in the 'partially drawn' state for a longer time. Also, it takes windows 7 longer to navigate through contents of folders in explorer. That said, Windows is still my preferred OS to use personally and I don't find it all that inconvenient, but that is just what I have noticed.
In my experience, Linux desktop response suffers way more heavily under high disk load then Windows desktop response. Something with the way Gnome and KDE are prioritized in the kernel loop I would expect. Run something in the background that is chewing up the disk and expect windows to draw very slowly.
Clearly something you should have researched better before buying a second phone with the same issue.
Doesn't work. Now you are dealing with two devices and a cable. Too cumbersome on a trail. Personally I spend a lot of time sitting at the side of a swimming pool these days for swim meets and I don't want to deal with two devices there either.
I'll reboot my phone. No worries there.
Let me add, a USB power pack doesn't work on a hike either. I wouldn't even use a USB power pack and I am considerably more tolerant than my friend. Say you are walking along a trail and you want to check your phone, or take a picture. Now you have a cord dangling off your phone so you are not just grabbing it out of your pocket, it is in your bag with a cord attached. So you have to use both hands now and either pull out phone with power pack or go into your bag and disconnect the pack, hoping you have had the phone on there long enough for it to charge enough while the GPS is going. It's just not convenient at all.
This is why I'm saying.. removable battery, just bring as many charged ones as you need, problem solved. For my use I only need one extra.
I find battery stats to be quite wrong in most cases. For one thing it is a known fact that smartphones drain the battery faster just when you are in an area with a weak or slow connection. So many variables involved it's a bit rediculous for any phone maker to make a claim on their battery life. Therefore I cannot rely on that number when dumping close to $1K on a handheld device.
The specific conversation was that my friend wanted to track a hike he was going on with his son. I asked him why he doesn't just use a GPS tracking app on his phone and he said he worried that the phone battery wouldn't last all day. Being an Apple fan, he likes his phone to be thin and lean so he would never consider attaching a battery pack.
Question: Does Apple specifically say that a user is allowed to open the device? Because I think if not I wouldn't open it. Definitely not if it was still on warranty.
I'm curious how people are finding that a battery pack is the solution, when:
1) Usually the battery dies when you are using the phone, so you can't just put it aside at that point and charge it. I can use my phone while it is on a battery back but then it hardly charges and i'm stuck with the pack.
2) A battery back is physically bigger than a phone battery and most require an extra cable. How is that ever going to be more convienent then just carrying another charged battery?
Sounds like a general phone design issue as opposed to a removable battery issue. On my phone there is no individual door, the whole back comes off and there is no risk of it just coming off. Furthermore, the back is usually under the case that I have that clips tightly onto the phone.
You think when they put the battery in at the factory they are just going to inject pure battery 'juice' into the phone??