HDDSuperClone

[maximus]

What I am trying to do right now is the approximate equivalent of ddrescueview. It loads a log file and creates the display. It will have most of the basics that I need to learn for GUI programming. You are correct that just a GUI for the functions and options would not suffer the the CPU usage that I mention, so I am not sure how much time I will spend on that aspect of it. Right now with a large log it struggles with resizing the window and scrolling the main window. But if you have been paying attention to how my brain works, I seem to always like to do the hardest and most impossible thing first

The important thing is being able to do a layout, and have drop down menus from the top bar, both of which I have achieved at the basic level. Even if I don't get the CPU usage down for the drawing area, it is still something I need to finish good enough so I know how to do things for the cloning tool. I need to know enough about it so that I can figure out how I need to restructure the program. It could even be possible that I may do away with the command line version of hddsuperclone completely, even for the free version. It just depends on how much I need to restructure it.

 

[pclab]

Hi Maximus

Have you seen the Deepspar interface? It's kind of a MS-DOS interface. What it basically does if to be able to see the functions and the possibility to enable or disable them, even during the imaging process (we need to stop and continue of course).
So, no need to have something to much eye-candy.

 

[maximus]

You are right, it does not need to be the best looking interface, and it probably won't be either. It won't have the real time bitmap viewing for sure, after seeing how much resources it would use (even though I have already made a significant improvement in that respect). But I also don't want it to look like crap.

But you are missing the point. I need to learn how to do the GUI and figure out the limits of what I can do, and the best way is to write a fully functional program from scratch. There is so much code in HDDSuperClone that needs to be modified for the GUI that it is not practical to try while learning the GUI at the same time. So HDDViewer is born as my learning tool. Plus I like ddrescueview and always wondered if I could do something like that. And maybe a part of me also wants to one-up it

 

[maximus]

I would also like to defend my new HDDViewer as an important tool in helping the end user to understand and "see" what is going on with the recovery. I initially had some great ideas as to what I was planning on adding to it, but I think that once I have the main visual aspect down (might be at that point now) and learn all the things I need to know (still working on that), it will stay as mostly a visual tool. I think about the only thing I will add is that when you click on a visual block it will simply show the start and end of that block, and it will be up to you to look it up in the logfile to see more.

Imagine that after a few hours of running HDDSuperClone you have 75% recovered after the first two passes, but then it slows down to a crawl. Using the viewer would give a visual or what is going on. Maybe all the pros would just know (and perhaps not even use HDDSuperClone for such cases), but being able to visualize it is just that much better. And I don't want to rely on exporting a ddrescue style log so it can be viewed with ddrescueview.

 

[maximus]

I released an alpha of HDDViewer a few days ago. I call it an alpha because I had intentions of putting some extra abilities in it, but it currently does what it needs to so I am not in any hurry to make it better. I am attaching a few screenshots of it in action on a 8GB chunk of a 160GB drive. Note that the bad head marking is not possible with the free version of HDDSuperClone, but I wanted to show it here. Also the second two pics are with fast skipping which is also not available in the free version. But the free version does perform the head skipping algorithm which would basically give the same results of the first two normal skipping pics minus marking the bad head.

Pic 1 is with normal skipping after finishing phase 1. Pic 2 is with normal skipping after finishing phase 2. Pic 3 and 4 are the same as 1 and 2 except with fast skipping. They probably won't show in the proper order on here but the files are named so you can tell them apart.

This drive actually does not react well when using soft resets, so this was done with the error control timer on the drive set at 1 (lowest useful value for this drive, 100ms timeout, done with standard ATA commands), so it only took a few minutes to achieve this result. This drive has a weak head with many small single sector errors that slow the recovery down a lot (initial clone with ddrescue took 30 days total without error timer control), but when fully cloned will recover 99.95%.

If you look towards the top of the images you can see there appear to be more reads in the bad head, but by about half way down the pic you will notice that it is leveled out to only a few reads in each section of the bad head (7 per pass for normal, 4 for fast skip). Yes, I am trying to show off the self learning head skipping algorithm with this

Edit: I forgot to add that this visual ability showed some major flaws in the head marking ability that I was not aware of and were very difficult to notice by just looking at a log. I spent a few days on HDDSuperClone to fix that before I could make these pretty pictures :roll:

 

[maximus]

I have seen a few references to DMDE and have just now tried out the free Linux version. It had a DOS style interface and was very slow to respond to commands (maybe because it is the free version?), and locked up on me a few times.I see it has a reference that the paid full DOS version has the ability to use a read timeout. So has anyone had experience with this software? Is this more or less my direct competition? I was not impressed, and would use ddrescue over it any day.

 

[pclab]

I used to work with the Windows Version and it works pretty good.
Not for imaging, mostly for file recovery.

 

[LarrySabo]

I use it and it's superb, especially for data recovery as @pclab said. I use the Windows version and I suspect that's what the majority of DMDE users use, and what potential HDDSuperClone users would prefer to use.

 

[maximus]

I did more research on DMDE after I posted here and do see it is geared more for file recovery. I am more focused on the cloning aspect of things for now. I would also guess that the Windows version doesn't suffer from the issues that the Linux version did in my test. As for imaging, I would never plug a failing drive directly into Windows for recovery... might as well take a hammer to it instead. But I can see the benefit (and popularity) of using Windows for file recovery from an image or clone, although not sure I would even want to plug a clone into Windows unless it only had a small handful of bad sectors.

 

[LarrySabo]

I could be wrong, but I think it opens the image file in read-only mode.