UltraDefrag 6.0.0 beta1 has been released

2012-11-03
2012-12-17
  • Stefan Pendl
    Stefan Pendl
    2012-11-03

    Hello friends,

    This release is improving the usability, stability and reliability of UltraDefrag.


    Here is a brief list of the changes:

    • Improved performance of all processing algorithms.
    • Introduced FAT optimization.
    • Introduced sorting by path, size, creation time, modification time and last access time.
    • Introduced filtering by file fragment size.
    • Introduced tuning of disk optimization.
    • Introduced pausing of the processing in the GUI.
    • Introduced minimizing to the system tray.

    We would like to receive as much feedback as possible, so we can further improve this release.

    You are welcome! Your help will be much appreciated!


    Stefan Pendl
    UltraDefrag Team - Maintainer

     
    • Tomasz W.
      Tomasz W.
      2012-11-04

      Thank you very much.

      I'd just like to ask one question. I saw some time ago somewhere (don't remember where exactly) that Windows NT 4.0 and 2000 weren't going to be supported in UltraDefarg 6.0.

      I've just installed beta1 in Windows 2000 and it does work. Does it mean that Win2k will still be supported?

       
      • Stefan Pendl
        Stefan Pendl
        2012-11-04

        Hi Tomasz,

        we dropped support for Win2k and below initially while developing v6, since we wanted more streamlined and clean code.

        Win2k and below put quite some restrictions onto us in respect to Windows defrag API support.

        About a week before releasing v6 beta1 Dmitri included support for Win2k and below once more due to user requests.

        For now Windows NT 4.0 up to Win7/Server2008R2 is supported.

        --
        Stefan

         
  • Tomasz W.
    Tomasz W.
    2012-11-04

    Thank you for clearing the situation up.

    I myself would probably be able to make the program work even if the support had been dropped (by modifying the files) but it's always nice to have this support provided natively, especially because there doesn't seem to exist any other defragmentation software supporting such a wide range of Windows versions.

     
    • Stefan Pendl
      Stefan Pendl
      2012-11-05

      Modifying the PE header won't work, since you will end up with API functions being called that don't exist on Win2k and below which results in crashing.
      In this situation enabling boot time processing would result in a system that can't be accessed any more.

      Windows NT 4.0 and 2000 are no longer listed at the Windows life-cycle page and there are no longer updates and security fixes supplied.

      From my point of view it is hard to support something that has been dropped by the vendor many years ago.
      I know that Dmitri has a different opinion, but that doesn't make development easier.

      I know there are sayings like "never change a running system" or "if it is not broke don't fix it", but I don't see any good in those with respect to computers.

      Software is under heavy changes due to the environment it is being run in is changing fast, so spreading your legs till you tear apart is no good.
      At some point in time one needs to let go old things to be able to introduce new ones.

      Finally systems were nothing changes are likely to not suffer from fragmentation, so a tool to remove fragmentation doesn't make sense for those.

      --
      Stefan

       
  • Tomasz W.
    Tomasz W.
    2012-11-07

    We've already got kernel extensions for Win2k which allow to run even applications compiled in VS2012. They cover not only XP specific APIs but also several APIs from Windows 2003 Server and Vista. I don't know which APIs exactly you're talking about when it comes to UltraDefrag but I'm almost sure that they are already included in the new updated kernel files. You've probably heard about KernelEx for Windows 98. It's a similar idea although much more extensive. Most of the new XP updates have been also ported so I'm not really concerned about security. M$ has also officially released new IE Cumulative Updates for Windows 2000 post EOL (including the latest one) together with a few other security updates.

    I just wanted to say that it's nice to have UltraDefrag support such a wide spectrum of Windows releases. It makes the program kind of unique among all the other different utilities which only focus on the newest versions of Windows :)

    Best regards,
    Tomasz

     
    • Stefan Pendl
      Stefan Pendl
      2012-11-07

      Sure Windows 2000 might not suffer from as much missing API functions as Windows NT 4.0 does, but still there are some that are not supported on a regular Windows 2000 installation.

      Thanks for valuing our effort to make UltraDefrag compatible with older Windows releases.

      --
      Stefan

       
  • dave c
    dave c
    2012-11-15

    6.0.0. beta 1 absolutely ROCKS

    6 year old Win XP SP2 AMD Sempron

    MAJOR difference in OS performance for me when listening to streaming music, if that makes any sense. Previously (with and without V 5) the brower would slow and crash after about an hour, the HDD seemed to be getting badly fragmented.

    Its been 2 hours now (or more) with no problem.

    The defrag window is open right now, Im not seeing the fragmentation and loss of optimisation as before.

     
    Last edit: dave c 2012-11-15
    • Stefan Pendl
      Stefan Pendl
      2012-11-16

      Thanks for the positive feedback, it is good to know that we are on the right track.


      Stefan

       
  • noorman
    noorman
    2012-11-15

    Thanks for that wonderful news.
    I 'm currently using the latest 5.1.1 version of which I 'm very happy too.
    I 've also installed it on my friend's system that was fragmented seriously and helped me to get it 'healthy' again in the shortest time! :-)
    I hope things stay OK and that no major bugs are found.
    I 've been using Ultradefrag for quite some time now, happily.
    It 's always been quicker in its workings as anything I knew before ;-)
    I 'm grateful for this very good program.

    Eric

     
    • Stefan Pendl
      Stefan Pendl
      2012-11-16

      Thanks for the positive feedback, it is our goal to make UltraDefrag as good as possible.


      Stefan

       
  • UAflyer
    UAflyer
    2012-11-22

    I am sure this was asked somewhere, but when is Windows8 support planned.

    V 6 seems to be a big improvement

    Cheers

     
    • Stefan Pendl
      Stefan Pendl
      2012-11-22

      We will support Win8 when we are able to set up a test environment for it.

      Win8 is putting some special hardware requirements on us and we seem to currently don't fulfill these requirements.

      Since we don't earn money for UltraDefrag we will have to wait till we can afford to buy something that fulfills the requirements.


      Stefan

       
      • noorman
        noorman
        2012-11-22

        Always the same with M$; they always demand a minimum of hardware from its users, unlike f.e. Linux! Wouldn't have expected otherwise from them =:(

         
  • Geoff Morris
    Geoff Morris
    2012-11-30

    Using the v6 beta for the first time today. I noticed that it is taking a long time, then discovered that it seems to be set to sort by path (at least, that's the default in "preview").

    I assume that means it's re-sorting the entire disk.

    Is this wise? I mean 1) why isn't there a "no sorting" option (same as v5)? 2) defaulting to "path" seems odd to me - I expect "last modification", ascending would be better for maintaining a clean disk...

     
    • Stefan Pendl
      Stefan Pendl
      2012-12-08

      This may be a good feature request, so the user can specify his default sorting.

      The no sorting option would need discussion with Dmitri, so I will check with him.


      Stefan

       
  • Geoff Morris
    Geoff Morris
    2012-11-30

    Follow-up: after spending a good twenty minutes or so apparently doing nothing and with the "optimised" count creeping up to 0.1%, it suddenly got going - and NOW it's really fast! :-)

    But... it makes a complete mess of my disk, presumably due to the sorting as I suggested. The disk is quite full (9% free) and despite repeating the optimisation several times it seems to keep moving the same blocks from the front of the disk and splitting them to later parts of the disk... then reversing the whole thing again. In the process, it's created holes all over the place. There is NO large block of free space - when there was one (from defragging with v5) before I started!

    P.S. I presume "last modified, ascending" means that the files which change most should end up at the end of the disk...?

    P.P.S. I thought about a better sorting solution: move the least edited files to the very end of the disk, then build up the data (in order of modification) FROM THE BACK, not from the front. That would result in free space at the start of the disk (so any modified files will be put there by Windows), followed by frequently modified files, followed by less frequently all the way down to never modified.

    P.P.P.S I will leave it doing a boot-time defrag (repeated) when I go home from work tonight. Hopefully it will be nice and sorted when I get back on Monday morning...

     
    Last edit: Geoff Morris 2012-11-30
    • Stefan Pendl
      Stefan Pendl
      2012-12-08

      In such a situation quick optimize would leave blocks of already sorted files untouched, so the process will be faster.

      Will check that with Dmitri too.


      Stefan

       
  • Geoff Morris
    Geoff Morris
    2012-12-03

    So this morning it looked much better when I booted up. Judging by the large block of compressed files in the middle, a lot of directory files further up the disk, etc. it seems to have sorted things a bit on Friday night.

    But there are still gaps, and when I run it now it still just shuffles some stuff from the front of the disk to the free space and back again... leaving gaps in the main block and a scattering of files in the free space. Why isn't it filling the gaps and why does it seem to ignore the bulk of the data once the free space is filled? I had cleared and recreated the USN journal, so that isn't causing the issue.

     
    • Stefan Pendl
      Stefan Pendl
      2012-12-08

      Sure the new algorithm needs some fine tuning, so we hope to get much better results for the final release.


      Stefan

       
      • Geoff Morris
        Geoff Morris
        2012-12-10

        No problem. That's why I thought I'd give some feedback... :-)

         
  • Hello guys,

    I'd like to clarify things a bit.

    The main idea of the disk optimization (as I envision it myself) is that you run the full optimization once and wait till it sorts all the files on the disk by the preferred criteria (path, modification time or something else) for the fastest access.

    Then, you run the quick optimization from time to time and it sorts out the recently modified files only taking a great deal shorter time for that.

    I thought about a better sorting solution: move the least edited files to the very end of the disk

    Once we'll switch to this algorithm, the users will complain that the end of the disk is slower than its beginning, so we're moving file in the wrong direction.

    Why isn't it filling the gaps

    It is almost impossible to fill all the gaps precisely due to the variety in file sizes, unmovable files and reluctance to move all the files to and fro.

    We're not trying to kill two birds with one stone. As for me, it's better to leave some gaps on the disk (which will appear anyway when you'll delete or move some files), but make the sorting of files as efficient as possible. I mean, sorting makes access faster because files become closer to each other, so let's simply make 'em closer to each other.


    Dmitri

     
    • Geoff Morris
      Geoff Morris
      2012-12-10

      I agree about the full/quick optimisation (although personally I do a clean-up->backup->defrag every couple of months, so full optimisation works for me as all garbage has been removed each time).

      Once we'll switch to this algorithm, the users will complain that the end of the disk is slower than its beginning, so we're moving file in the wrong direction.

      I was under the impression(*) that "the beginning" of the disk was supposed to have faster access times than "the end"... so it makes sense to have most reading/writing happening there instead. And the OS prefers to put new files at the beginning anyway...

      (*) The last "real" version of Norton SpeedDisk (before Symantec took over and turned all the Norton stuff into invasive bloatware) used a strategy where frequently modified files were at the beginning, frequent access in the middle and archive material at the very end.

      It is almost impossible to fill all the gaps precisely due to the variety in file sizes, unmovable files and reluctance to move all the files to and fro.

      I understand... but v5 leaves an almost contiguous block of data - much better than this beta.

      Thanks for the great product, guys! Looking forward to the final v6... :-)

       
      Last edit: Geoff Morris 2012-12-10
  • Geoff Morris
    Geoff Morris
    2012-12-10

    I also have a question: the sorting... the boot version uses the sorting from the "preview" menu option in the UI, correct? Is it possible to set sort options in the boot-time script?

    And a feature request: no idea if this is possible but it would be great to run UUDefrag during shut-down instead of during boot. Much more convenient to click a button "defrag and shutdown" (then leave the PC to do it's thing) instead of enabling the boot-script, waiting for the defrag during boot, then opening UUDefrag to disable the boot-script again. At the moment I emulate defrag-and-shutdown by enabling boot-script with the shutdown command inside, then clicking on restart and going to bed... but I still have to remember to hit a key during the next boot and disable the script again afterwards.

     
  • The boot time defragmenter reads options from the boot time script, not from the configuration file of GUI. To set sorting method use environment variables, for example, include the following lines to the script:

    set UD_SORTING=M_TIME
    set UD_SORTING_ORDER=DESC

    The default is to sort files by path in ascending order.

    More info on the accepted environment variables can be found in the handbook (press F1 in GUI to open it).


    To disable the boot time defragmenter for the subsequent boots, just add the boot-off command before the shutdown command.

     
    • Geoff Morris
      Geoff Morris
      2012-12-10

      Thanks. For some reason I overlooked the sorting commands in the help. So it was doing path-sorting at boot... that explains a lot!

      The boot-off option is very useful. It wasn't very clear to me before, but now after looking again at the help description I understand how it works.

      Time to change my boot-script...

       
      Last edit: Geoff Morris 2012-12-10
  • v5 leaves an almost contiguous block of data - much better than this beta

    This is exactly because it filled gaps by files in random order. It never tried to sort 'em.

    The new algorithms prefer the fine sorting over the dull gaps filling. To fill gaps as tightly as in v5 we should either don't care about the sorting or move huge bunches of files just to cover the gap before 'em.

     
  • Geoff Morris
    Geoff Morris
    2012-12-12

    Does "modified time, ascending" sort so that "earlier values precede later ones e.g. 1/1/2000 will sort ahead of 1/1/2001"? After using that sort option, the block of data seemed to show a lot of compressed (i.e. old) files near the bottom instead of the top as I expected.

    I noticed you quoted descending time in your example, so I tried that last night - now the compressed files are a quite scattered around the disk, so it's hard to tell.

    Feature request: allow investigation of the disk map (like old Norton SpeedDisk): clicking on a point of the map will show which files' data is stored at that location.

     
    Last edit: Geoff Morris 2012-12-12
  • Yeah, the ascending order means that earlier values will precede.

    The program shows files compressed by NTFS file system, not by a file archivers.

    Maybe we'll add the requested feature later, however it would be not easy.

     
    • Stefan Pendl
      Stefan Pendl
      2012-12-13

      There is an existing feature request Make grid clickable to identify files for this, so you may vote for it.


      Stefan

       
      • Geoff Morris
        Geoff Morris
        2012-12-13

        I have added a + vote... seems I am 100% of the electorate! ;-)

         
        • Stefan Pendl
          Stefan Pendl
          2012-12-14

          Users will have to get used to the new capabilities of SourceForge.

          I think I will add a notice to check out the feature request tracker with the next release.


          Stefan

           
  • Ray Mulder
    Ray Mulder
    2012-12-16

    Hi, just joined the forum
    I have always loved this app
    Just installed Windows 8 and would like to know if there is any specific reason why it should not work on Win 8. I have already used the 5.1.1 version a few times and it seems to be ok, but then I cannot really tell. I have not noted any obvious errors on the drive.
    Quite prepared to be the guinea pig...Have Win 8 Pro
    Thanks
    Ray

     
    Last edit: Ray Mulder 2012-12-16
  • Hi, thanks for your feedback.

    We haven't tested the program as careful on Win8 as on all the previous Windows editions. So, we've decided to declare its official support a bit later when we'll gather enough information on its compatibility.

    If you'll notice any bugs or even little mistakes occurring on Win8, let us know - we'll try our best to fix 'em.

    Best wishes,
    Dmitri