[extropy-chat] more moore

Eugen Leitl eugen at leitl.org
Wed Sep 1 13:51:55 UTC 2004


On Wed, Sep 01, 2004 at 11:29:07PM +1000, Alejandro Dubrovsky wrote:
> >  H.264 realtime HDTV
> > encoding in studio can take 100 k$ of hardware (if there is such a thing yet,
> > haven't looked), it doesn't matter.
> 
> visiblelight seems to think that $325 will do.  And if you don't like

http://www.visiblelight.com/mall/productview.aspx?pid=582
is a 2U rackmount box which says "please contact us for pricing". This is
usually a sign for professional price tags. (Though probably lower than 100
k$ which I pulled out of /dev/ass).

http://www.visiblelight.com/mall/productview.aspx?pid=491
doesn't do HTDV, though I'm surprised that H.264 realtime encoding in
hardware is already available.

> them, there seem to be many other companies offering alternatives.  or
> am i missing something big here?

Not much, just the HDTV part. And HDTV is just the beginning, current
mid-range displays (though not yet video projectors) already do several megapixels.
 
> Immersive audio has been available for a while (depending on your
> definition of the term), mostly brought on by the soundblaster live and
> whatever its competitor was at the time (i forget).  What happened i
> think is that they realised that noone cared, and few can even tell on
> how accurate the modelling is, so why waste cycles, or more importantly,
> coding time.  

Fraunhofer did sell it's first phased array sound system to the cinemas,
though I do realize that being able to take a stroll through the orchester
pit is not something even audiophiles would buy. 

True immersive audio (without instrument physical modelling, for each
individual organ pipe and string) involves audio wavefront propagation in the
full volumetric game world model, reflection and attenuation included. So you
hear the echos, and feel the fog/snow.

> video is a different story.

Yes, our limit here are displays. There are some 3D LCD displays which take
OpenGL drivers, but no immersive displays, whether individual HUDs or
volumetric ones.
 
> > I meant unmodified (no xfce) default Fedora Core 2 Gnome end user reaction
> > times. It will take at least an order of magnitude faster CPU to render that
> > at XP desktop speed, all other things being equal.
> >  
> 
> I don't run xfce.  Why the focus on fc2?  If the software's properly

I don't have a specific focus, every Xfree86 or X.org displays on speedy CPUs
and reasonably recent nForce/ATI offering performs abysmally as compared to
e.g. Windows display snappiness (everything else, particularly awful
worst-case QoS is another story). It is an architecture issue, but changes in
architecture takes years to decade, so it's effectively static in comparison to hardware
advances.

> screwed (ie waiting for a timeout somewhere), then no amount of hardware
> will help you. Try some others.  My experience with XP has been brief,
> but it didn't feel any faster to me than my usual X/gnome running on a
> dinky 1 gig duron.

FC2 is particulary screwy, but XP is considerably snappier than the fastest
X system I've seen. There will be slow progress (hopefully on FC3 already),
but my point was that faster hardware is a poor man's solution to doing it
right to start with.
 
> Yes.  Much.  Agreed.  Even disabling bells and whistles in fact.
> 
> (Agreed with rest too)

We do agree, and I'm overposting here today. Should be last post, at least on
this thread.

-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a>
______________________________________________________________
ICBM: 48.07078, 11.61144            http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
http://moleculardevices.org         http://nanomachines.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20040901/4333eccd/attachment.bin>


More information about the extropy-chat mailing list