Re: [OT] CD-R Requirements: Conspiracy
On 22 Nov 2002 15:42:17 +0100, Stefan Janecek <stefan.janecek@jku.at>
wrote:
>On Thu, 2002-11-21 at 23:37, Pigeon wrote:
>> On Wed, 20 Nov 2002 18:55:41 -0800, Vineet Kumar
>> <debian-user@virtual.doorstop.net> wrote:
>>
>> >This is exactly what burnproof does. You can load the machine; you can
>> >deplete the buffer. Rather than spewing the contents of the empty
>> >buffer to the disc (creating a coaster), it suspends the write and
>> >resumes when the buffer fills some more.
>
>> Just imagine if the electronics of your hard drive spewed garbage to
>> the platter if it ran out of incoming data for a bit. Or if your NIC
>> broadcast random crap while the processor was attending to someone
>> else. What a stupid way to build a device.
>>
>
>*GG* that's bullshit, of course. It is easy not to send data over the
>network while you are busy doing something else. It is not so trivial to
>resume writing to a CDR exactly at the point you stopped...
OK, but what about the hard drive? In both cases the head has to track
over the preceding sectors and switch into write mode when it gets to
the right sector to be written. OK, so one's magnetic, one's optical,
but the head servos are doing the same thing. The spiral track on a CD
means you basically have to "pick the track up" a few rotations behind
where you need to start writing so you've got time to sync in; not a
problem. And it's the same problem as starting a new session on a
multisession CD, or packet writing. It's only gonna be awkward if the
laser drive is so screwed that it burns garbage for the first few
sectors after you put it into write mode.
Pigeon
Reply to: