[MPlayer-users] Re: Which TV card ?

Tuukka Toivonen tuukkat at ee.oulu.fi
Wed Aug 20 19:12:26 CEST 2003


On Wed, 20 Aug 2003, Oliver Fromme wrote:

>I'm sorry, I should have been clearer.
>I'm recording at DVD resolution which is 720 x 576 (for PAL).

I'm sorry too but for my bad memory. Using BTTV card to grab PAL, the
maximum image size is actually slightly larger. I think 768x576 (but I
really need to check this). This is the image size when zooming xawtv to
fill whole screen; it'll have image of that size, with the space between
window edges and the image just plain black. But the image is actually
larger than what TV displays: sometimes there are some funny things at the
extreme left/right edges which TV doesn't show (like overwrapping...).

>I'm using UFS2.  My Samsung 160GB disk has a sequential
>write speed of about 45 MByte/sec at the beginning.  I

7200 RPM (P80) or 5400 RPM (V80)?
I'm planning to buy the V80 tomorrow :)
Let me see...I'll guess it's V80: according to other people, hdparm reports
44.44 MB/sec on that. V60 would be only 35.16 MB/sec, unless it's P40
(older 7200 RPM disk).

UFS2? Why? Is it faster or is there some unrelated reason?

> > [*] Why the heck hard disks with multiple platters don't use RAID
> > internally?
>
>I don't think it would make much sense.  RAID 1 or 5 (for
>redundancy) is useless, because usually the whole drive
>fails, not just one platter.

I mostly agree, though not completely. Few sectors might fail on one
platter before the drive reallocates them.

>And RAID 0 only makes sense
>when the components can be accessed independently, which
>is not the case within one drive, because there's only
>one actuator for the drive heads, so you cannot move the
>heads independently from each other.  So, an operation on

I don't see why that would stop using RAID-0. The actuator could write
every other bit on first platter, every other bit on the second platter, at
the exactly same positions. The data transfer speed to the actuator would
be need to be doubled (or more), but that shouldn't be a problem.

It might be a problem to align the tracks at exactly same positions on each
platter, but I'm surprised if that doesn't get compensated with 2-4 times
faster speed. Well, seek time wouldn't change...



More information about the MPlayer-users mailing list