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 |
#11
|
|||
|
|||
why won't write-behind stay disabled?
System.ini shows no obvious mention of write-behind.
Vcache settings: maxfilecache=3072 chunksize=512 minfilecache=3072 Reasons are that I'm ways to inprove disk performance because I'm tired of getting file system problems because my system too often hangs up before data gets written to my drive (lost clusters and orphaned pieces of files, etc.) I've tried everything else believe me. Second I'm tweaking my MTU/port settings for faster internet data transfer. I was reading microsoft website about setting port speed http://technet2.microsoft.com/Window...4f1a41033.mspx when I read what this same article said about write-behind "You may want to disable the write behind cache function, especially if you own system critical applications, and ALWAYS shut down Windows AFTER closing ALL running programs! This means all data will be immediately written to disk, bypassing the cache." I take it from my reading that write-behind is also implictated in problems with the data layer of internet connections. I've been trying ever since to disable write-behind. Please if you wish to comment on my reasoning OK but help me disable write-behind. |
#12
|
|||
|
|||
why won't write-behind stay disabled?
I didn't say system, I said disk system. I don't know why you want to test
your disk system, but I want to test it to compare hardware, and to do that you have to turn off write behind caching in the operating system. -- Jeff Richards MS MVP (Windows - Shell/User) "Noel Paton" wrote in message ... "Jeff Richards" wrote in message ... It must be turned off if you want to do any serious disk system performance testing. That is total BS! - the whole point of system benchmarking is to test the system "as it will be used" - not in some airy-fairy-land of the dealer's choice! |
#13
|
|||
|
|||
why won't write-behind stay disabled?
In that case, you don't use Windows to do it, but a direct-access OS
-- Noel Paton (MS-MVP 2002-2006, Windows) Nil Carborundum Illegitemi http://www.crashfixpc.com/millsrpch.htm "Jeff Richards" wrote in message ... I didn't say system, I said disk system. I don't know why you want to test your disk system, but I want to test it to compare hardware, and to do that you have to turn off write behind caching in the operating system. -- Jeff Richards MS MVP (Windows - Shell/User) "Noel Paton" wrote in message ... "Jeff Richards" wrote in message ... It must be turned off if you want to do any serious disk system performance testing. That is total BS! - the whole point of system benchmarking is to test the system "as it will be used" - not in some airy-fairy-land of the dealer's choice! |
#14
|
|||
|
|||
why won't write-behind stay disabled?
Did you see the date on that article? - 1995!!
it is totally irrelevant to anything resembling a modern machine, having been written (presumably) during the Beta for Win95. In fact, its zip file is dated at 2002, so cannot have been updated since then - and the page itself (http://www.cerberus-sys.com/~belleis...in.html#BOTTOM) admits it to being 1996 In the real world, the machine is faster than the output through any sort of COM modem port! If you've been reading articles about memory configuration, and you're still using Cacheman, then you've obviously been reading the wrong ones - or you haven't taken any notice of the right ones! vache does NOT use up RAM - contrary to what is stated in many article - and has even less to do with any HD optimisation/tweaking you may want to do with 48 MB of RAM, I have to wonder what the rest of your setup is like? - in particular the CPU and HD, since if they are from that era then we may be able to better advise on settings details on your hardware would be useful at this point! -- Noel Paton (MS-MVP 2002-2006, Windows) Nil Carborundum Illegitemi http://www.crashfixpc.com/millsrpch.htm http://tinyurl.com/6oztj Please read http://dts-l.org/goodpost.htm on how to post messages to NG's "Olive" wrote in message ... Right now I'm playing with drive write-behind. As of now, I still can not turn it off. I'm trying to report results about the effects of disabling write-behind on disk performance and on internet speed. I've been reading this article. See the paragraph that starts with "Another cause is poorly written 32-bit disk drivers" http://www.cerberus-sys.com/~belleis...aq/overrun.htm The article says a tempopary solution to slow internet speeds until you can update your hard drive, bios and other drivers is to temporarily disable drive-write behind. I installed a new drive in 2003, can't update my bios [don't ask], and have the latest win98 drivers available. So I thought I would try to temporarily disable write-behind to test the effects. But I can not report results to this NG until I can disable write-behind. Sir, my current vcache settings are just the latest temporary settings from months of playing with Cacheman and reading articles about memory management. My current vcache settings use Cacheman to set min and max vcache to 3072 and set chucksize to 512. Then I let windows manage files and buffers by remming out lines files= and buffers= from my autoexec.bat These current settings give me some stability and (most importantly) free up for other uses about 13 meg of ram of my miniscule 48 meg ram. Otherwise, sir, my machine starts swapping early and often like a sinful couple in an open marriage. Soooooooo, right now it's about disabling wite-behind. I have faith in this NG. I know posters are searching for options. This newsgroup has never failed me. |
#15
|
|||
|
|||
why won't write-behind stay disabled?
Right now I'm playing with drive write-behind.
As of now, I still can not turn it off. I'm trying to report results about the effects of disabling write-behind on disk performance and on internet speed. I've been reading this article. See the paragraph that starts with "Another cause is poorly written 32-bit disk drivers" http://www.cerberus-sys.com/~belleis...aq/overrun.htm The article says a tempopary solution to slow internet speeds until you can update your hard drive, bios and other drivers is to temporarily disable drive-write behind. I installed a new drive in 2003, can't update my bios [don't ask], and have the latest win98 drivers available. So I thought I would try to temporarily disable write-behind to test the effects. But I can not report results to this NG until I can disable write-behind. Sir, my current vcache settings are just the latest temporary settings from months of playing with Cacheman and reading articles about memory management. My current vcache settings use Cacheman to set min and max vcache to 3072 and set chucksize to 512. Then I let windows manage files and buffers by remming out lines files= and buffers= from my autoexec.bat These current settings give me some stability and (most importantly) free up for other uses about 13 meg of ram of my miniscule 48 meg ram. Otherwise, sir, my machine starts swapping early and often like a sinful couple in an open marriage. Soooooooo, right now it's about disabling wite-behind. I have faith in this NG. I know posters are searching for options. This newsgroup has never failed me. |
#16
|
|||
|
|||
why won't write-behind stay disabled?
|
#17
|
|||
|
|||
why won't write-behind stay disabled?
Why not use Windows (provided you disable the write-behind feature)? It's
available and it's already running. Or if you object to that one so much, perhaps I have a logging application that buffers its disk writes within the application so that I can implement very aggressive disk drive power management. Write-behind has to be disabled or the disk drive stays active for much longer causing a reduction in battery time. Or perhaps I use one of those data base systems where the manufacturer has warned me that data corruption can occur in a multi-user environment when write behind is enabled. Or perhaps I am using a high performance raid system where the manufacturer advises disabling write behind so that the intelligent algorithms in the controller can operate most efficiently. Or any number of other possible reasons. -- Jeff Richards MS MVP (Windows - Shell/User) "Noel Paton" wrote in message ... In that case, you don't use Windows to do it, but a direct-access OS -- Noel Paton (MS-MVP 2002-2006, Windows) Nil Carborundum Illegitemi http://www.crashfixpc.com/millsrpch.htm "Jeff Richards" wrote in message ... I didn't say system, I said disk system. I don't know why you want to test your disk system, but I want to test it to compare hardware, and to do that you have to turn off write behind caching in the operating system. -- Jeff Richards MS MVP (Windows - Shell/User) "Noel Paton" wrote in message ... "Jeff Richards" wrote in message ... It must be turned off if you want to do any serious disk system performance testing. That is total BS! - the whole point of system benchmarking is to test the system "as it will be used" - not in some airy-fairy-land of the dealer's choice! |
#18
|
|||
|
|||
why won't write-behind stay disabled?
If your recovery procedure involves restoring an earlier version of the
registry then this could cause the setting to revert. Then you wouldn't know whether or not it had changed before the problem occurred. Of course, it's always possible that the registry is getting corrupted, for instance as a result of a memory error. However in this case I would expect that the symptoms were more variable and much more widespread. Consider what utilities you are running that might be trying to 'optimise' your system behind your back, or utilities that need to manipulate Windows' settings for their own purposes, such as diagnostics, and may not be restoring things to how they were. -- Jeff Richards MS MVP (Windows - Shell/User) "Olive" wrote in message ... I still have not received a workable answer. I want to disable drive write-behind and make the settings stick. How do I do this? Thanks. |
#19
|
|||
|
|||
why won't write-behind stay disabled?
Will report back. This NG gave me four things to work on. Thanks.
To Mr. Paton your arguments proceed from at least two false assumptions: 1) The lion's share of computers use Microsoft's latest and greatest technologies and 2) Computing principles change as fast as computing technologies. All I know is using the principles in this article sped up internet speed of my Win98 system dramatically. The principles, paraphrased from the article, must be done in order--1, 2 then 3: 1) Start with your data link layer and make it solid (eliminate com port overrunns). Enable FIFO. Reduce rate setting your com port advertises to modem. For example, set modem speed to 56K but set comm port to 38K. You have to change comm port speed within each software program that uses internet. Otherwise Microsoft article says you cannot make comm port change change using Control Panel. 2) Next work on the PPP or SLIP layer. Turn off the "speed-up" option on your S3-chip-based video cards. If current video software does not allow you to change speed-up mode then manually add a line to the [display] section of system.ini saying bus-throttle=on. Turn off write-back cache for all drives. Write-behind caching for VCACHE is turned-off with a line in the [386enh] section of system.ini that says ForceLazyOff=C 3) Next use EasyMTU to tune the MTU, MSS and RWIN for maximum speed. |
#20
|
|||
|
|||
why won't write-behind stay disabled?
Olive
The COM port is severely limited in its bandwidth - while any Pentium-II-class machine is quite capable of filling that bandwidth with no impact on system speed, unless perhaps it also has to cope with compression on-the-fly (which is better done within the modem, anyhow) You would appear to be using Windows 98 - and applying Win3.1 tweaks to it. Many of these tweaks don't work, simply because Windows changed significantly between Win3.x and Win98, and the hooks for the tweaks to hang on just don't exist. Many of those that still apparently work are not supported, or can lead to system instability or data loss. In response to 1) This would appear to expose a weakness in your modem, rather than a problem with the OS - in that the modem is relying on the OS to do compression. Have you updated the drivers for the modem? in response to 2) - again, the s3 video tweak would appear to expose a hardware problem rather than an OS problem - again, a driver update would seem to be a possible solution. This is the first time I have ever heard of the ForceLazyOff - I'm trying to get some information on its working in Win98 as to 3) - the author of EasyMTU says that it's designed for DUN 1.2 (on Win95) - and may not work with higher versions. Since an updated Win98 has version 1.4 as a minimum, I suspect that the program may have problems. ....and you've still given no details as to your hardware! In view of my comments above - perhaps you had also better include the driver versions in use for the relevant items. -- Noel Paton (MS-MVP 2002-2006, Windows) Nil Carborundum Illegitemi http://www.crashfixpc.com/millsrpch.htm http://tinyurl.com/6oztj Please read http://dts-l.org/goodpost.htm on how to post messages to NG's "Olive" wrote in message ... Will report back. This NG gave me four things to work on. Thanks. To Mr. Paton your arguments proceed from at least two false assumptions: 1) The lion's share of computers use Microsoft's latest and greatest technologies and 2) Computing principles change as fast as computing technologies. All I know is using the principles in this article sped up internet speed of my Win98 system dramatically. The principles, paraphrased from the article, must be done in order--1, 2 then 3: 1) Start with your data link layer and make it solid (eliminate com port overrunns). Enable FIFO. Reduce rate setting your com port advertises to modem. For example, set modem speed to 56K but set comm port to 38K. You have to change comm port speed within each software program that uses internet. Otherwise Microsoft article says you cannot make comm port change change using Control Panel. 2) Next work on the PPP or SLIP layer. Turn off the "speed-up" option on your S3-chip-based video cards. If current video software does not allow you to change speed-up mode then manually add a line to the [display] section of system.ini saying bus-throttle=on. Turn off write-back cache for all drives. Write-behind caching for VCACHE is turned-off with a line in the [386enh] section of system.ini that says ForceLazyOff=C 3) Next use EasyMTU to tune the MTU, MSS and RWIN for maximum speed. |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
write protection error | kel via WindowsKB.com | Disk Drives | 3 | January 17th 06 10:21 PM |
Whoa. What was that? 98 load failure and.... | keith | General | 20 | March 3rd 05 06:46 AM |
Restart 3-4 times before it can be use | frustrated 98se user | General | 18 | February 12th 05 04:14 PM |
Please help! Display settings !! | Mitzi | Monitors & Displays | 12 | July 11th 04 05:19 AM |
Disk write errors | Bob Ninow | Disk Drives | 4 | June 6th 04 07:00 PM |