[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: growisofs + DVD+R + Plextor708A

> Some more feedback. Pav requested that I try the unit in another OS. I
> have installed Windows2000 (Servicepacks and blah) and used Nero. I
> have succesfully burned an image at 8x, taking 63 minutes.

You mean it said it *intended* to record at 8x and recorded at ~1x, as
63 minutes sounds as ~1x, less actually...

> The drive
> underruns a lot and is therefor slow in burning. It does work though.

Just like under Linux:

> >  785645568/4633516032 (17.0%) @3.0x, remaining 17:03
> >  790134784/4633516032 (17.1%) @0.9x, remaining 17:16   <-- speed switch
> >  794886144/4633516032 (17.2%) @1.0x, remaining 17:23
> >  801112064/4633516032 (17.3%) @1.3x, remaining 17:32
> >
> > It does complete succesfully though. mediainfo then shows (on the
> > completed DVD+R):

As for decreasing effective velocity. 3x is commonly observed speed in
lack of DMA and I see no reason why it should drop even further, when
unit increases the spin (that's what happens at "speed switch" point,
increase from 6x to 8x spin). Indeed, as internal buffer capacity is
enough to sustain *several* revolutions, firmware should be able to
minimize the amount of idle revolutions, which has to be taken during
underrun condition. The effective velocity should therefore be somewhat
lower  than the "wire" speed, but it shouldn't just drop by factor of
several N... But life is not always what we expect it to be:-( I would
argue that such poor performance during buffer underrun condition is a
firmware deficiency, but it would hardly help you with this particular
problem, woudn't it? Engaging DMA and hoping for the best is the only
sensible thing to do.

> My first attempt at burning a DVD+R without any options whatsoever to
> growisofs resulted in a 5.9-6.0 speed right away. I'm sorry to say that
> after a couple of hundred MB, the burning process stopped mysteriously,

Do save error outputs as well as check kernel log or explicitly say "no
error messages whatsoever" were produced. I mean avoid such words as
"mysteriously" in reports:-) One way or another it sounds like media
problem. Bad batch maybe? It even might be a counterfeit media. At least
Maxell DVD-R brand was exposed to counterfeit attempt, you can't be sure
that it won't happen again with some other brand. How does media
emballage look like?

As for growisofs failing with "Failed to change write speed:
8310->5540." It's known issue, see
http://lists.debian.org/cdwrite/2004/cdwrite-200402/msg00094.html for
details and temporary workaround, and will be addressed at some point in
some way.

As for

> $ growisofs -dvd-compat -Z/dev/cd0=image
> Executing 'builtin_dd if=image of=/dev/pass1 obs=32k seek=0'
> /dev/pass1: "Current Write Speed" is 6.1x1385KBps.
>          0/4633516032 ( 0.0%) @0x, remaining ??:??
>     262144/4633516032 ( 0.0%) @0.1x, remaining 2945:44
>     ...
>     262144/4633516032 ( 0.0%) @0.0x, remaining 14728:43

Essentially it's OK that recording gets stuck in the beginning (time
required for initialization procedure and inevitably longer than the
time to initially fill the buffer up), but I suppose you waited long
enough (a minute maybe)... I wonder how does dvd+rw-mediainfo output
look like for this media. A.

Reply to: