[Bug 1458230] Re: libburn imposing 0.2x BD-R burn (5-6 hours)

[Thomas Schmitt]

Hi,

Njorl wrote:
> Any possibility for making libburn more idiot-proof?  If the log had
> been able to decompose how the burn speed was determined, I might have
> managed to check up on the bottleneck - and have avoided this drain on
> your time.

First i would have to make myself idiot-proof.
 
In hindsight the pacifier messages give a clear indication:

>> xorrecord ... fs=8m ...
>> xorriso : UPDATE : 10186 of 10189 MB written (fifo 1%) [buf 78%] 0.2x.

If the software fifo is at 1% of 8 MB = 80 kB while there are still 3 MB
to burn, then the input side of xorriso is not fast enough.

The drive buffer stays filled at about 60 to 80 % because the drive sees
no urgent need for writing. It rather waits until the buffer is nearly
full before it puts the data on medium.

----------------------------------------------------------------------

So i now officially declare that the problem was caused by insufficient
data source bandwidth.

USB 1 has at most 12 MBit/s. The shown xorriso run burned 10189 MB in
10286 seconds. That's 990 kByte/s = 7.925 MBit/s net payload.
This matches well enough to consider USB 1 the original cause of the
poor input speed.
In any case, USB 1 is not fast enough for normal DVD or BD burning.
1x DVD = 1385 kB/s , 1x BD = 4495 kB/s.

----------------------------------------------------------------------

Have a nice day :)

Thomas

-- 
You received this bug notification because you are a member of Ubuntu
Burning Team, which is subscribed to libburn in Ubuntu.
https://bugs.launchpad.net/bugs/1458230

Title:
  libburn imposing 0.2x BD-R burn (5-6 hours)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/libburn/+bug/1458230/+subscriptions