Eraser 6.0.7.1893 remarkably faster than vs. 5?

MarkS

New Member
I recently upgraded to version 6. I noticed that the time it takes to erase a file with 35 passes with vs. 6 is remarkably faster than with vs. 5. The only thing that has changed on my system is the version of Eraser. The same hardware, OS and the files are the same size. I would like to say this is a result of optimizations in the code, but the speed for a 35 pass erase is around the same as a single pass erase. What used to take around 10 minutes now takes less than 1.

I have a sneaking suspicion that 35 passes are not being performed. Has anyone else noticed this?

Running on Windows 7 x64 Ultimate with 8 GB of RAM.
 
What are you erasing? Which file system are you erasing the files on? How big is the file? How did you create the task?

I'd estimate that for a 10-minute task with 35 passes, the file would have to be around 1GB.
 
Leaving aside the fact that you almost certainly don't need 35 passes, I too have noticed that Eraser 6 is quite fast on file erasing with modern machines. Both later versions of Eraser 5 and Eraser 6 could run as 64 bit apps on an x64 OS, but Eraser 6 seems to do rather better in this regard.

If you check the erase with a file recovery program (e.g. Recuva), and your erased files are not discoverable, Eraser has to all intents and purposes done its job.

David
 
On a side note, what technique or process is used to increase the speed of the passes?


I know this sounds like a vague question, but any insights into the matter would be Much Appreciated.
 
None, actually. This is the first time I've heard of such an observation.
 
What we do know is that, compared with Eraser 5, Eraser 6 is a completely different program, written in a different language, and that it uses a different runtime library. As the .NET library now used is the one Microsoft has chosen to promote, it is reasonable to assume that they have put some effort into optimising the code within it. That, of course, is not something the Eraser Team can have any involvement with; like us at one further remove, they just use the product. But it is always possible that, if good use is made of current processor, memory and OS technology, speed gains can occur.

This is only a hypothesis, and even if it is well founded, I would expect that different systems would show different degrees of improvement.

David
 
Joel said:
What are you erasing? Which file system are you erasing the files on? How big is the file? How did you create the task?

Misc directories containing files totaling around 2 MB in size. The task was created from the context menu. I don't know what file system is in use (how do I check?), but I did not reformat before the upgrade, so I don't think this is an issue.

Joel said:
I'd estimate that for a 10-minute task with 35 passes, the file would have to be around 1GB.

10 minutes may be an exaggeration, but the difference in time is not. The speed difference is up to 10x faster.

Odd, I have never heard of .NET ever being referred to as fast or optimized. But that is irrelevant. Disk access can only be optimized to a point in software. It is still a typically hardware intensive application. I would have expected a faster erase if I had upgraded to a faster drive, but that is not the case.

I just want to make sure that if I choose 35 passes that 35 passes are being performed. I am aware that 35 passes is truly unnecessary, but still...
 
MarkS said:
Odd, I have never heard of .NET ever being referred to as fast or optimized. But that is irrelevant. Disk access can only be optimized to a point in software. It is still a typically hardware intensive application. I would have expected a faster erase if I had upgraded to a faster drive, but that is not the case.
Agree about the hardware intensive nature of the program, though this is less true of file/folder erasing than of wiping free space. And I'd certainly be surprised if the performance differential were 10 times. As regards .NET, I was merely surmising. I know that (like anything with a Microsoft mark on it), .NET comes in for a great deal of hate mail, but I can only speak as I find, which is that apps using .NET do not seem to be noticeably slower than those using other runtime libraries. Maybe I'm just lucky.

MarkS said:
I just want to make sure that if I choose 35 passes that 35 passes are being performed. I am aware that 35 passes is truly unnecessary, but still...
When you think about it, it's pretty much impossible to prove one way or the other. What matters to me is that a file I want gone is indeed gone. That at least we can test, to a reasonable degree of confidence.

David
 
MarkS said:
Misc directories containing files totaling around 2 MB in size. The task was created from the context menu. I don't know what file system is in use (how do I check?), but I did not reformat before the upgrade, so I don't think this is an issue.
Well, 2 MB, even on 35 passes, is only 70MB of data, something a disk should be able to accomplish in less than 10 seconds.

MarkS said:
10 minutes may be an exaggeration, but the difference in time is not. The speed difference is up to 10x faster.
In view on the earlier point, this sounds reasonable to expect, IMO.
 
Back
Top