Am Dienstag 15 Februar 2005 18.52 schrieb Steven M. Schultz:
> > Did anyone try to use alternative compilers?
>
> Never got around to doing that. It was simpler (or so it seemed)
> to get a faster system ;)
>
Since the issue was puzzling me, I tried to get the best out of my box. her
On Tue, 15 Feb 2005, Michael Hanke wrote:
> I'm using the recommendation -t 3 by Steven Boswell. The source material is
> very (not to say extremly) noisy.
I think the recommendation for extremely noisy material is '-t 4'
(and I have used 5 for the worst material). That paramet
Am Dienstag 15 Februar 2005 00.42 schrieb Steven M. Schultz:
>
> On Mon, 14 Feb 2005, Michael Hanke wrote:
>
> > I count y4mdenoise an excellent piece of software but I thought that I
make
> > something wrong: On my machine (1.2GHz Athlon) I obtain a denoising rate
of
> > 0.3 frames/sec. (tha
On Mon, 14 Feb 2005, Michael Hanke wrote:
> I count y4mdenoise an excellent piece of software but I thought that I make
> something wrong: On my machine (1.2GHz Athlon) I obtain a denoising rate of
> 0.3 frames/sec. (that is, 5 days for a movie) This is far too slow for
That is a litt
Hi,
Am Samstag 12 Februar 2005 20.47 schrieb Steven M. Schultz:
> The final total elapsed time of the two encoding runs:
>
> dual G5: 34hrs 7min 1.3 frames/sec
> dual Opteron: 15hrs 46 min 2.80 frames/sec
>
I count y4mdenoise an excellent piece of software but I thought that I
On Feb 13, 2005, at 9:13 AM, Steven M. Schultz wrote:
On Sun, 13 Feb 2005 [EMAIL PROTECTED] wrote:
that a movie media made of several layers does not have the same
granularity
(which does obably apply on classic still cameras as well).
The blue layer seems to be made of bigger grains than other,
On Sun, 13 Feb 2005 [EMAIL PROTECTED] wrote:
> Ok no other way than working on the RGB before YUV
Take a look at the utilties/filters in the NetPBM package. Some of
them handle 'streams' (back-to-back) PNM images - I seem to recall
some denoising filters in that package
> You can reconstruct _a_ blue from the U and V planes but it won't
> be the _original_ blue.
That is what I was suspecting ...
Ok no other way than working on the RGB before YUV
> Good Luck!
euphemism :-)
Cheers
E
---
S
On Sun, 13 Feb 2005 [EMAIL PROTECTED] wrote:
> that a movie media made of several layers does not have the same granularity
> (which does obably apply on classic still cameras as well).
> The blue layer seems to be made of bigger grains than other, therefore not
> having the same resolution.
Gday
I think I am gonna be a pain again. Following the setup of my telecine machine
and further readings (yes, I did RTFM on this one), it appears that a movie
media made of several layers does not have the same granularity (which does
obably apply on classic still cameras as well). The blue la
On Sat, Feb 12, 2005 at 11:36:27AM -0800, Steven M. Schultz wrote:
> On Sat, 12 Feb 2005, Roine Gustafsson wrote:
> > It's an urban myth that 64bit is faster than 32bit, like people assume
> > a 2GHz computer is twice as fast as a 1GHz computer.
>
> It's also an urban myth that 64bit is slo
On Sat, Feb 12, 2005 at 07:16:12PM +0100, Roine Gustafsson wrote:
> However, I'll wager the cluster of 32bit beige boxes would be more bang
> for the buck than a honkin' 64 bit Bigiron UltraPower. Atleast for
> rendering.
An off the shelf Opteron or G5 will be a lot cheaper than the same
amount
Steven M. Schultz wrote:
if yes, you may want to hack it so that it locks each thread to a
particular processor, taking into account the data transfers between
I think the benefits of pinning are over rated - and it's not portable
(different systems either do it differently or don't support it
On Sat, 12 Feb 2005, Amaury Jacquot wrote:
> >> The Opteron-250 with its 2.4GHz clock is about *2X* faster
> >> than the
> >> 2GHz G5/PPC system I have! What took 34 hours to run on the G5
> >> (or the
>
> is that G5 PPC dual processor or not ?
Of course. I haven't (wit
On Sat, 12 Feb 2005, Roine Gustafsson wrote:
> Interesting. What was the off-the-shelf price for the Opteron? The
At the time the Opteron-250 cpu was $825. Or were you thinking of
the complete system price? I didn't buy a complete system - just
a motherboard/cpu/memo
On Sat, 12 Feb 2005, Roine Gustafsson wrote:
> But clusters are difficult. They eat lots of electricity and put out
> vast amounts of heat. They are also quite difficult to administrate and
> only specific tasks scale well.
Very true. Some things (dealing with weather data) scale well
A good 64bit CPU (64 bit data) in a 32 bit OS (32 bit address space)
will be faster.
However, I'll wager the cluster of 32bit beige boxes would be more bang
for the buck than a honkin' 64 bit Bigiron UltraPower. Atleast for
rendering.
Weta used a cluster of 2.8Ghz Xeons, for probably very well
Gday Roine
Well ... If I follow, it sounds to be better off with a couple of nice 32 bits
in cluster mode than a 64 ??
E
just end user
> Therefore, applications that do not use the larger addressing space
> won't be any faster on a "64 bit operating system". If anything, it
> just means 2x t
I've been looking into getting just such a machine to replace my dual 1GHz PIII.
How about posting your system specs.
Richard Ray
On Sat, 12 Feb 2005, Steven M. Schultz wrote:
> Howdy -
>
> On Wednesday and Thursday I put together a new dual Opteron-250
> system and installed SuSE 9.
On Feb 12, 2005, at 1:10 PM, Amaury Jacquot wrote:
64bit is generally slower than 32bit. The only benefit of 64bit is
non-segmented addressing of several gigabytes of data. If you don't
need that then 64 bit adressing is just overhead.
no it's not.
64 bit on a 32 bit processor is smaller because
Roine Gustafsson wrote:
On Feb 12, 2005, at 9:54 AM, Steven M. Schultz wrote:
Howdy -
On Wednesday and Thursday I put together a new dual Opteron-250
system and installed SuSE 9.2 (64bit). Uneventful install but there
were a few issues with rebuilding a few apps (all but one have been
Roine Gustafsson wrote:
64bit is generally slower than 32bit. The only benefit of 64bit is
non-segmented addressing of several gigabytes of data. If you don't
need that then 64 bit adressing is just overhead.
On x86-64 they did not only change from 32 bit to 64 bit, but they also
doubled the num
On Feb 12, 2005, at 9:54 AM, Steven M. Schultz wrote:
Howdy -
On Wednesday and Thursday I put together a new dual Opteron-250
system and installed SuSE 9.2 (64bit). Uneventful install but there
were a few issues with rebuilding a few apps (all but one have been
reso
Howdy -
On Wednesday and Thursday I put together a new dual Opteron-250
system and installed SuSE 9.2 (64bit). Uneventful install but there
were a few issues with rebuilding a few apps (all but one have been
resolved - can't get Ogle to build so I'll use Xine inste
24 matches
Mail list logo
