If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
|
Thread Tools | Display Modes |
#16
|
|||
|
|||
Printer prints blank page, not an ink problem
On Mon, 28 Mar 2016 01:50:42 -0400, Micky
wrote: Early Sunday morning I printed it, and though the paper came out, it was blank. I did this 3 more times, because the last time iirc what was once blank printed a 400K file when I printed it to a file, but this time all 4 times it was blank. .... Well this time it asked what name I should give to the file. This was Sunday's puzzle. It's after midnight so I tried to print Monday's. Again it printed blank, so, rather than keep trying that, I printed to CutePDF, again the file was about 70K (instead of 400K), and I printed that to the same old Samsung laser printer that printed blank a few minutes earlier. It does seem, doesn't it, that there is something wrong with the driver? Despite what I said. Because the webpage seems able to provide the data, and the printer seems able to print. Maybe the problem starts when the computer goes to sleep, or hybrid sleep. Does that sound like it could be the reason? Sometimes it has a hard time waking up from hybrid sleep, takes much longer than other times, and sometimes it's very slow for 20 minutes or more afterwards. Also, sometimes there are script errors when I leave the computer alone for a couple hours, even though there almost never are when I'm sitting here doing things. Does that make sense? How would I stop that? This is vista but it's a lot like 7, I hear. Conceivably it was still the first print job asking after 20 minutes, but if so, the second print job never asked. One time it did not ask. Instead of 400K it's only 70K but I displayed it with Adobe Reader and printed it from that, to the same printer, and it printed fine. It might be a little lighter, the lines not as dark, than printing directly, not sure. So, so far, this appears to be a way around the blank page issue, which was the problem this thread is named after. |
Ads |
#17
|
|||
|
|||
Printer prints blank page, not an ink problem
Micky wrote:
On Mon, 28 Mar 2016 01:50:42 -0400, Micky wrote: Early Sunday morning I printed it, and though the paper came out, it was blank. I did this 3 more times, because the last time iirc what was once blank printed a 400K file when I printed it to a file, but this time all 4 times it was blank. .... Well this time it asked what name I should give to the file. This was Sunday's puzzle. It's after midnight so I tried to print Monday's. Again it printed blank, so, rather than keep trying that, I printed to CutePDF, again the file was about 70K (instead of 400K), and I printed that to the same old Samsung laser printer that printed blank a few minutes earlier. It does seem, doesn't it, that there is something wrong with the driver? Despite what I said. Because the webpage seems able to provide the data, and the printer seems able to print. Maybe the problem starts when the computer goes to sleep, or hybrid sleep. Does that sound like it could be the reason? Sometimes it has a hard time waking up from hybrid sleep, takes much longer than other times, and sometimes it's very slow for 20 minutes or more afterwards. Also, sometimes there are script errors when I leave the computer alone for a couple hours, even though there almost never are when I'm sitting here doing things. Does that make sense? How would I stop that? This is vista but it's a lot like 7, I hear. Conceivably it was still the first print job asking after 20 minutes, but if so, the second print job never asked. One time it did not ask. Instead of 400K it's only 70K but I displayed it with Adobe Reader and printed it from that, to the same printer, and it printed fine. It might be a little lighter, the lines not as dark, than printing directly, not sure. So, so far, this appears to be a way around the blank page issue, which was the problem this thread is named after. What does Task Manager show you ? Have you been sleeping the computer, over and over again for months, without a reboot. Or is the computer rebooted regularly, and the "sleep" is just the odd time during the day ? When returning from sleep, the drivers have to be "warm started". The registers on the chips have to be re-loaded with correct values. None of your symptoms match an issue there. You can have activity like Windows Update, Windows Defender, Search Indexer, file system health, as stuff that runs when you don't want it to. The Search Indexer can even be caught working, when the index is up-to-date. It never sleeps. You can have slowness, from certain memory usage patterns. But I lack a good paragraph to insert here, as for how that works. If a program uses all the RAM, things like the system file read cache get purged, which may make some things a bit slow later. I've also noticed how some programs that use memory mapped I/O, if the mapped space is large enough, it seems to affect performance later. (As seen when VPC2007 reads an entire DVD. Not a problem with a CD sized image. In fact, VPC2007 can't read a whole DVD, because the mapping method used fails.) I find a reboot usually works around such side effects. Paul |
#18
|
|||
|
|||
Printer prints blank page, not an ink problem
[Default] On Mon, 28 Mar 2016 04:13:59 -0400, in
microsoft.public.windowsxp.general Paul wrote: Micky wrote: On Mon, 28 Mar 2016 01:50:42 -0400, Micky wrote: Early Sunday morning I printed it, and though the paper came out, it was blank. I did this 3 more times, because the last time iirc what was once blank printed a 400K file when I printed it to a file, but this time all 4 times it was blank. .... Well this time it asked what name I should give to the file. This was Sunday's puzzle. It's after midnight so I tried to print Monday's. Again it printed blank, so, rather than keep trying that, I printed to CutePDF, again the file was about 70K (instead of 400K), and I printed that to the same old Samsung laser printer that printed blank a few minutes earlier. It does seem, doesn't it, that there is something wrong with the driver? Despite what I said. Because the webpage seems able to provide the data, and the printer seems able to print. Maybe the problem starts when the computer goes to sleep, or hybrid sleep. Does that sound like it could be the reason? Sometimes it has a hard time waking up from hybrid sleep, takes much longer than other times, and sometimes it's very slow for 20 minutes or more afterwards. Also, sometimes there are script errors when I leave the computer alone for a couple hours, even though there almost never are when I'm sitting here doing things. Does that make sense? How would I stop that? This is vista but it's a lot like 7, I hear. Conceivably it was still the first print job asking after 20 minutes, but if so, the second print job never asked. One time it did not ask. Instead of 400K it's only 70K but I displayed it with Adobe Reader and printed it from that, to the same printer, and it printed fine. It might be a little lighter, the lines not as dark, than printing directly, not sure. So, so far, this appears to be a way around the blank page issue, which was the problem this thread is named after. What does Task Manager show you ? Not sure. What should I be looking for. One strange thing, unrelated to sleeping, that I've been meaning to ask about, is that often Firefox, and less often Eudora, other programs, and even once in a while Solitaire, will say Not Responding (I think that's it) but the amount of RAM used can be 60 to 80%, and the CPU can be 5%. What's holding things up if there is unused RAM and CPU? Have you been sleeping the computer, over and over again for months, without a reboot. Or is the computer No more than 3 weeks and the problem can happen after only 3 days after rebooting. rebooted regularly, and the "sleep" is just the odd time during the day ? More like the latter. When returning from sleep, the drivers have to be "warm started". The registers on the chips have to be re-loaded with correct values. None of your symptoms match an issue there. You can have activity like Windows Update, Windows Defender, Search Indexer, file system health, as stuff that runs when you don't want it to. The Search Indexer can even be caught working, when the index is up-to-date. It never sleeps. You can have slowness, from certain memory usage patterns. But I lack a good paragraph to insert here, as for how that works. If a program uses all the RAM, things like the system file read cache get purged, which may make some things a bit slow later. I've also noticed how some programs that use memory mapped I/O, if the mapped space is large enough, it seems to affect performance later. (As seen when VPC2007 reads an entire DVD. Not a problem with a CD sized image. In fact, VPC2007 can't read a whole DVD, because the mapping method used fails.) I find a reboot usually works around such side effects. Thanks. Paul |
#19
|
|||
|
|||
Printer prints blank page, not an ink problem
Micky wrote:
One strange thing, unrelated to sleeping, that I've been meaning to ask about, is that often Firefox, and less often Eudora, other programs, and even once in a while Solitaire, will say Not Responding (I think that's it) but the amount of RAM used can be 60 to 80%, and the CPU can be 5%. What's holding things up if there is unused RAM and CPU? "Not Responding" should be measuring some aspect of the Windows program. Like whether the program Message Loop responded in five seconds or not. https://en.wikipedia.org/wiki/Messag...rosoft_Windows If you ever wrote your own program, where a window is rendered on the screen, you have to put in the fake event handler to keep everything happy. Even if your intention is not to actually use the window created. If Windows sends you an event, you reply "yah, got it", then just ignore it. And the fast response back like that, counts in your favor ("Is Responding"). The programs I write, are normally command line type, but one program (which I cannot find now), I had to put in that fake message loop since there was a rectangular window on the screen. If an event sent to a program does not receive a response in a fixed period of time, then the more modern OSes go nuts and put up the Not Responding thing. I've even seen problems, where a piggish program sucks up all the CPU cycles, and a "victim" program sitting next to it, oscillates in and out of the "Not Responding" state. The victim simply didn't have a chance to respond. So that's an extreme example of bad system design (not enough cycles to keep all programs properly responsive). When you're not running piggish programs, you shouldn't be seeing Not Responding, unless the thing is actually in an infinite loop and no longer cares about your input. ******* Modern Firefox is a very busy program - it now runs a compositing loop at 60FPS. It no longer appears to be event driven (i.e. only updates the screen based on some state information inside the program). The compositing loop appears to me, to be the equivalent of a wasteful polling loop. In the sense that it constantly redraws the screen (even if it's only juggling some pointers, and handing off the task to your GPU). On my other machine, it can use 40% of a CPU core, doing nothing. Now, if you had a messaging loop running there, you'd hope it would be a separate thread, and the thing would remain responsive. If your CPU is weak, Firefox should simply be dropping FPS and modifying it's behavior. I haven't attempted to analyze that aspect of it, to see how it deals with really weak computers. Or deal with computers that don't have the minimum hardware requirements to composite like that. Firefox has a built-in Profiler, which should work in the various OSes. You can see in this diagram, I've taken a trace, and had it display the result. The green fuzz, is the compositing loop. It's too bad I didn't capture a CPU graph to go with it. I consider it bad design to chew up a CPU like that. Across North America, think of the electricity that design is wasting right now :-) The developers should be proud. http://s12.postimg.org/bisuvijwd/profiler.gif Paul |
#20
|
|||
|
|||
Printer prints blank page, not an ink problem
Paul wrote:
Micky wrote: One strange thing, unrelated to sleeping, that I've been meaning to ask about, is that often Firefox, and less often Eudora, other programs, and even once in a while Solitaire, will say Not Responding (I think that's it) but the amount of RAM used can be 60 to 80%, and the CPU can be 5%. What's holding things up if there is unused RAM and CPU? Modern Firefox is a very busy program - it now runs a compositing loop at 60FPS. It no longer appears to be event driven (i.e. only updates the screen based on some state information inside the program). The compositing loop appears to me, to be the equivalent of a wasteful polling loop. In the sense that it constantly redraws the screen (even if it's only juggling some pointers, and handing off the task to your GPU). On my other machine, it can use 40% of a CPU core, doing nothing. Now, if you had a messaging loop running there, you'd hope it would be a separate thread, and the thing would remain responsive. If your CPU is weak, Firefox should simply be dropping FPS and modifying it's behavior. I haven't attempted to analyze that aspect of it, to see how it deals with really weak computers. Or deal with computers that don't have the minimum hardware requirements to composite like that. Firefox has a built-in Profiler, which should work in the various OSes. You can see in this diagram, I've taken a trace, and had it display the result. The green fuzz, is the compositing loop. It's too bad I didn't capture a CPU graph to go with it. I consider it bad design to chew up a CPU like that. Across North America, think of the electricity that design is wasting right now :-) The developers should be proud. http://s12.postimg.org/bisuvijwd/profiler.gif Paul And it appears this compositing thing has been around for a while. This blog is from 2012. https://benoitgirard.wordpress.com/2...hy-it-matters/ Paul |
|
Thread Tools | |
Display Modes | |
|
|