alphaa10

For those running an external USB hard drive for imaging / backup purposes, there is every reason to suppose fragmentation builds up as fast as on an internal partition. The obvious concern-- although we are already severely limited in I/O speed by USB2.0 limits (to a lesser extent by USB3.0) the major problem is defragmenting an external USB HD partition. Under USB2.0, this becomes almost a disincentive to regular maintenance-- defragmentation must be done overnight and into the next morning, or not at all. More often than not, defragmentation is not done at all. Has anyone an opinion whether the quickest way to resolve fragmentation is simply to reformat the external drive before a backup session, and be done with it in a few minutes?

Willy2-- Cleanmem was run manually during its test, and did reduce (from its own indicator) the RAM in use by a few percentage points. However, I rarely have exceeded 40 percent of installed and available RAM-- installed, in part, to reduce time-consuming resourse to the PageFile-- so this utility may be a marginal benefit to the original concern about PageFile fragmentation. In addition, I found running task scheduler a waste of system resources and disabled it, so having an immediate (manual) response from CleanMem rather than scheduled assistance is my only option, and the more desirable mode, anyway. BTW, I have installed a total of 6gb DDR2 in this XP Pro 32bit system in a plan to move to a 64bit version soon. Obviously, I cannot use more than 3gb of the 6gb installed under a 32bit system.

WILLY2-- Thanks, I'll try it. Presumably this utility is a minor load on the system. kroozer-- Thanks, I'll look into what MS means by "clearing" the file. Does it mean content / data is removed, but the file structure is left intact? Or does it mean the file itself is removed? I'll determine that when I check with MS on the issue tomorrow on this KB article. * As it turned out, in doing the manual hack to clear the PageFile on shutdown, I found my key value was already set properly..This means my Defraggler message about PageFile fragmentation was issued with the proper key value to "clear" the PageFile already in place.

MTA-- My OS is XP Pro, SP3 and all the patches since. Effectively, my Page File has the partition all to itself, because the other partition is seldom used, and I may wipe it from the HD altogether. No traffic / i/o goes to the other partition. I like the idea of wiping the Page File at session shutdown, but am unsure how to manage that easily. WILLY2-- My Page File partition is large enough to place a 4gb single file, with extra "handling room" for the OS. The Page File size is the maximum permitted by the OS, so there is no expansion. And since I set the Page File minimum size to the same value as maximum size, there is no contraction. Thanks for the Cleanmem reference-- but I'll need to read more, since I like to keep many apps going at the same time.

Unlike most Windows users, I have placed my system Page File on a separate partition (on a different HD), on the idea this will reduce or eliminate fragmentation. As a test of whether that configuration actually reduces or prevents fragmentation, I attempted to run Defraggler on a heavily-used Page File partition. I was surprised to find Defraggler reported some fragmentaton. 1. Can you explain why fragmentation occurs on a separate Page File partition? 2. If my purpose is simply to keep the Page File reasonably integrated, is it better to skip time-consuming defragmentation, and simply format the Page File partition periodically, before starting a Windows session?

Speccy displays the partitions / volumes on each hard drive, numbering them 0,1, etc. However, it seems of value for Speccy to display the volume name, also. For most user operations with Speccy, the volume name is at least as important as the partition number Since Speccy already displays each partition heading ("Partition 0") in blue type, why not display the volume label here, as well?

Alan_B asked, "Were you defragging System C:?" Thank you for your response. On my system, the boot volume is L:, and the data volume I attempted to defrag is D:. The data volume resides on an older Seagate Barracuda 7200 RPM 400 GB IDE. At least 50gb of free space exists on that volume.

DeFraggler version 2.13.670 OS Windows XP Professional-- 32bit Mainboard is fully populated at 2gb Hard drive under defragmentation-- Seagate 7200 rpm Barracuda 250gb IDE The drive being defragmented was partitioned into three differently-sized volumes-- two in NTFS and one in FAT32. On two successive attempts to defrag the same NTFS 186gb partition (51gb free), DeFraggler aborted at 55 percent. No other apps were running during the session. When I let DeFraggler continue to run, despite the error message (attached), the status indicator stayed on 55 percent. The abort was a system error message, asking for permisson to send the DeFraggler crash report to Microsoft. Of course, I did not bother to bother Microsoft. For what it is worth, I was able to transcribe this much of the Error Signature-- AppName-- defraggler.exe, AppVer: 2.13.0.670 ModVer: 5.1.2600.5755 Offset: 000033bc

That would be a nice feature for AMD systems-- any plans for that?

Like most users, I have more than one physical drive running at all times. On occasion, as many as five separate, physical devices. In addition, many drives have more than one data volume / partition. As a result, I make it a rule to name each data volume uniquely, since this rapidly identifies the data volume for drive maintenance. SUGGESTION FOR SPECCY-- Display the data volume name-- for some reason, the current version does not. PS-- I really like the "fit and finish" of your product line-- I just discovered Speccy, after Googling the web for "system information utilities". Your website is gorgeous-- open, accessible, and logical. And best of all, unlike many sites, it does not drop the user into deep black space, with hard to read bright characters against a dark background. HINT (see next) ANOTHER SUGGESTION FOR SPECCY-- Display all information in black or colored characters against a white background-- or allow the user to configure background/foreground color.