A thought for Colin and MS regarding burn speeds

  • Thread starter Thread starter Bernie
  • Start date Start date
B

Bernie

I was wondering about this problem with burning iso images and what
might causing the frequent problems and came up with an idea of what
might be happening.

It simply didn't make sense to me that burn speed would affect the final
result unless the media was bad and didn't get burnt properly. Then I
figured something else...

When downloading a large file like the beta file it is likely to take a
few hours for most of us on a reasonable connection. At the same time we
are probably using the machine for other things and even if not, other
things are still happening with the O/S. Those other things involve
writing to disk....

I have the idea that the problem is the large download becomes
fragmented. It can even become extremely fragmented in the magnitude of
thousands of fragments for that one file.

You are then burning a "single" file that could be in thousands of
pieces scattered about the drive and this might affect the maximum speed
at which you could burn at as the process has to find all those
fragments pretty fast.

So the cure might be to complete the download and then defrag the file
before burning it. I think you can do this simply by copying it to
another location while doing nothing else with the machine. In the copy
process the fragments are brought together.

Just a thought.
 
Aren't most burners equipped with Overburn which stops writing when the
buffer is empty?.
You might be right ... but I doubt it.
BTW have you done checksum checks on the downloaded file to ensure it is not
corrupted?.
I postulated last week in this ng that the case of the belowmentioned
problems might be iso file corruption, however it was lost in the sea of
triviality found here.
 
It is a thought, but the following is an exception.

I downloaded the 3.6 gig file and saved it to my drive H: - a seldom used
drive (for some of my less important backups, a second copy if you will).
When my computer is being used, 95% of the disk activity is on drive C: and
drive D:

I had to burn at 4x to get a good copy. Why? Because of the disks I had
available at the time. When I obtained higher quality DVD's I was able to
burn the same file at 16x.

--
Regards,

Richard Urban
Microsoft MVP Windows Shell/User
(For email, remove the obvious from my address)

Quote from George Ankner:
If you knew as much as you think you know,
You would realize that you don't know what you thought you knew!
 
Bernie said:
I was wondering about this problem with burning iso images and what
might causing the frequent problems and came up with an idea of what
might be happening.

It simply didn't make sense to me that burn speed would affect the final
result unless the media was bad and didn't get burnt properly. Then I
figured something else...

When downloading a large file like the beta file it is likely to take a
few hours for most of us on a reasonable connection. At the same time we
are probably using the machine for other things and even if not, other
things are still happening with the O/S. Those other things involve
writing to disk....

I have the idea that the problem is the large download becomes
fragmented. It can even become extremely fragmented in the magnitude of
thousands of fragments for that one file.

You are then burning a "single" file that could be in thousands of
pieces scattered about the drive and this might affect the maximum speed
at which you could burn at as the process has to find all those
fragments pretty fast.

So the cure might be to complete the download and then defrag the file
before burning it. I think you can do this simply by copying it to
another location while doing nothing else with the machine. In the copy
process the fragments are brought together.

Just a thought.
Click to expand...

Bernie, I think you have some very good points.
I personally have never experienced any
problems burning .ISO files themselves. The
only problems I have ever encountered were due
to corrupted downloads. I have always burned
..ISO files at full speed.
 
We may know something before RC1. MS has mailed the test media to the 500
or so beta testers who volunteered to try out the various dvd's being used
in the test. I don't know just how they vary but I suspect that both media
and burn method are varied. I have no idea if the results will be published
to the Techbeta folks. But MS is going to some expense to track down an
issue so it is clearly a real concern.
 
Intel Inside:
I agree with you fully about corruption in the download. I have
2 Windows x64 towers running so I downloaded Vista x64 on each tower.
Neither download would install, after trying different burn speeds and brand
media I found Andre' post about the MD5 checksum. The first two downloads
that I had were corrupt, the third download matched the MD5 checksum and
installed without errors.
 
It is wise to check that any burner, CD or DVD, has the latest firmware. The
usual reason given for firmware updates is to improve compatibility with
media.

Older burners, for which manufacturers no longer produce firmware updates,
may struggle with newer media unless the burn speed is restricted. The lower
speed will compensate for minor technical problems with media
characteristics which the burner was not designed to cope with.
 
When I downloaded Vista Beta 2 x86, I burned it at 48x, and it burned fine.

The "problem" with burn speed has nothing to do with Vista, and everything
to do with the DVD burner and the media.

I have a LITE-ON DVDRW SHM-165H65 Lightscribe dvd burner, and I'm using HP
DVD+R Lightscribe dvds, and I have not had any problems.

Todd
 
That is what MS is studying.

Todd said:
When I downloaded Vista Beta 2 x86, I burned it at 48x, and it burned
fine.

The "problem" with burn speed has nothing to do with Vista, and everything
to do with the DVD burner and the media.

I have a LITE-ON DVDRW SHM-165H65 Lightscribe dvd burner, and I'm using HP
DVD+R Lightscribe dvds, and I have not had any problems.

Todd
Click to expand...
 
As I understood it the problem that MS have did not involve corrupted
downloads. If the download was corrupted it wouldn't matter what speed
you burnt at and MS were saying on the Vista get ready site to burn at
the slowest speed. They didn't say anything about corrupted downloads.
Obviously corrupted downloads can occur as happened to you but I don't
think that is what they are investigating.
 
On Sun, 30 Jul 2006 14:36:05 -0400, in
microsoft.public.windows.vista.general "Todd" <[email protected]>
wrote:

|When I downloaded Vista Beta 2 x86, I burned it at 48x, and it burned fine.

No, you did not burn a DVD at 48x.
 
Bernie:
I understand fully what MS is investigating. My reply was based on
my experience with the corruption of a file during download not a corruption
when burning the image. I tried different burn speeds and media prior to
suspecting the download. As in any diagnostic pattern starting with the
obvious and easier solutions is the first direction before going deeper to
find the solution.
 
Have you read the post above by Dennis Pack?.
Makes me wonder if you "500 or so beta testers" are chasing your tails.
 
Intel Inside:
I don't think that the testers are chasing their tails. I
eliminated media and burn speed prior to finding the download corruption.
Media, burn speed and the download itself are three areas of possible
corruption. MS is testing two of the three areas. MD5 and CRC are two ways
of verifying the download. When XPSP2 was being downloaded, after the
download finished a verification was performed before the download box
closed, but that was a much smaller file.
 
That is for MS to decide.

Intel Inside said:
Have you read the post above by Dennis Pack?.
Makes me wonder if you "500 or so beta testers" are chasing your tails.
Click to expand...
 
Bernie said:
I was wondering about this problem with burning iso images and what might
causing the frequent problems and came up with an idea of what might be
happening.

It simply didn't make sense to me that burn speed would affect the final
result unless the media was bad and didn't get burnt properly. Then I
figured something else...

When downloading a large file like the beta file it is likely to take a
few hours for most of us on a reasonable connection. At the same time we
are probably using the machine for other things and even if not, other
things are still happening with the O/S. Those other things involve
writing to disk....

I have the idea that the problem is the large download becomes fragmented.
It can even become extremely fragmented in the magnitude of thousands of
fragments for that one file.

You are then burning a "single" file that could be in thousands of pieces
scattered about the drive and this might affect the maximum speed at which
you could burn at as the process has to find all those fragments pretty
fast.

So the cure might be to complete the download and then defrag the file
before burning it. I think you can do this simply by copying it to another
location while doing nothing else with the machine. In the copy process
the fragments are brought together.

Just a thought.
Click to expand...

Isn't the "download and burn" scenario simply an academic issue? There
isn't this wide spread a problem with any other software that MSFT delivers
via download.

That is, of course, unless, MSFT intends to deliver Vista RETAIL by
download.... Hmmmmm...
 
Considering your comments "But MS is going to some expense to track down an
issue so it is clearly a real concern" I figured you could put in 2c of
effort and find out if corruption during download had been investigated.
Considering your elated status as an MVP I thought you'd be interested
rather than stating "That is for MS to decide".
I wonder why I bother.
 
Back
Top