[FFmpeg-devel] Anyone needs HW or other things that require money, please open a ticket on trac

[Michael Niedermayer]

On Tue, May 13, 2014 at 11:54:36PM +0200, Reimar Döffinger wrote:
> On Tue, May 13, 2014 at 08:26:01PM +0200, Michael Niedermayer wrote:
> > On Tue, May 13, 2014 at 02:18:59PM -0400, Ronald S. Bultje wrote:
> > > Yeah, let's get all devs an avx2-capable laptop. I'd love a new macbook pro
> > > :)
> > 
> > https://trac.ffmpeg.org/ticket/3636#ticket
> > 
> > any other ideas? anyone ?
> 
> To start with a caveat: Hardware on its own is only of limited use.
> Having time to actually make use of it is a bigger problem.
> And I don't open a ticket since the following are too vague and money
> isn't really the primary issue. I'd rather hear ideas from others if
> you have some (particularly on 1 and 2).
> That said, some of the larger annoyances/missing things hardware-wise
> from my point of view:
> 
> 1) Some actually _fast_, well-supported ARM based hardware. Even better
> if 64 bit.
> For most devices, both the CPUs and especially the flash storage
> performance are a bad joke.

this btw reminds me, that subjectively my cubox seems much
more responsive sd card access wise than my panda board.
But i didnt compare them side by side ...


> And the driver code is only available for 10 kernel revisions back and
> makes you weep when you look at it (average time to find a bug just by
> reviewing it, as someone who is clueless about the kernel code: maybe 
> 10 minutes?).
> Possibly doesn't exist.
> 
> 2) Some fast big-endian hardware. The public PPC servers seem to have
> disappeared and since then I only have an ancient and slow Mac Mini.

i wouldnt mind a fast arm and fast big endian box either
though as fate client this isnt essential,
a slow arm board works fine even if one compiles on the device instead
of doing cross compilation

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Let us carefully observe those good qualities wherein our enemies excel us
and endeavor to excel them, by avoiding what is faulty, and imitating what
is excellent in them. -- Plutarch
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <https://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20140514/0da77bf5/attachment.asc>