On 25/01/06, Justin The Cynical [EMAIL PROTECTED] wrote:
IMO, for a remote FE, a fast Celeron will work fine for SD broadcast.
I'm running a slave BE/FE system using a Celeron 1.3GHz w/256MB RAM
(Intel 815 chipset). Video is TV-Out from an ATI Radeon 9000 Pro. This
is with a PVR-150 and 2xDVB
Justin The Cynical wrote:
On Tue, January 24, 2006 10:14, Raphael Pooser wrote:
*snip*
In reality, encoding/decoding is computationally intensive, and at the
same time you need bandwidth as these actions involve streaming. since
Celeron is piss poor at floating point and has no
On Tue, January 24, 2006 10:14, Raphael Pooser wrote:
*snip*
In reality, encoding/decoding is computationally intensive, and at the
same time you need bandwidth as these actions involve streaming. since
Celeron is piss poor at floating point and has no bandwidth to access
the RAM, it must
Hmm, I'm running on an old FIC-SD11 Mainboard with a Slot-A Athlon 950. I've
got 384 Megs of RAM
in it. As for encoders/decoders, I'm running a 350 and using the hardware
output. I also have a
500 and all three tuners are working well, now that I got IVTV 0.4.2 installed
(I had the new
On 1/25/06, Justin The Cynical [EMAIL PROTECTED] wrote:
On Tue, January 24, 2006 10:14, Raphael Pooser wrote:
*snip*
In reality, encoding/decoding is computationally intensive, and at the
same time you need bandwidth as these actions involve streaming. since
Celeron is piss poor at
When I started this thread, I was hoping we could get a discussion going
about processor performance beyond just raw cpu speed, especially as it
relates to HDTV material. If cpu speed were the only factor, then a Celeron
3GHz would be better than a P4 2.8GHz, but I doubt that is true.
I guess
Michael T. Dean wrote:
Trey Boudreau wrote:
A useful rule of thumb says that you can buy the same compute hardware
cheaper next week (or next month). If you wait 18 months (a la Mr.
Moore), you'll only end up half broke ;-)
I thought Moore's "law" recently died,
Well Cache is important. Another reason a Celeron will suck compared to
a P4 is the bandwidth between the processor and RAM is crippled. The
Celeron front side bus was always chopped in half compared to a
pentium. The netburst architecture has always been bandwidth hungry, so
you take a
On Tuesday 24 January 2006 12:14 pm, Raphael Pooser wrote:
Well Cache is important. Another reason a Celeron will suck compared to
a P4 is the bandwidth between the processor and RAM is crippled. The
Celeron front side bus was always chopped in half compared to a
pentium. The netburst
Trey Boudreau wrote:
On Sun, Jan 22, 2006 at 05:42:50PM -0700, Brian Wood wrote:
Nobody ever complained about having too much CPU, or RAM, I'd go for
the best I could afford.
Sometimes all the CPU you can afford choice gets you into the more
fan noise than you can stand configuration.
Trey Boudreau wrote:
A useful rule of thumb says that you can buy the same compute hardware
cheaper next week (or next month). If you wait 18 months (a la Mr.
Moore), you'll only end up half broke ;-)
I thought Moore's law recently died, thus the X2 and Core (TM)
Duo's* and soon quad-cores
On 23/01/06, Michael T. Dean [EMAIL PROTECTED] wrote:
*I can't believe Intel is actually calling them Intel (R) Core (TM) Duo
processors ( http://www.intel.com/products/processor/coreduo/ ). And,
why isn't that (TM) after the word Duo? I'm pretty sure that the word
Core is *not* owned by
On 01/24/2006 01:51 AM, Nick wrote:
On 23/01/06, Michael T. Dean [EMAIL PROTECTED] wrote:
*I can't believe Intel is actually calling them Intel (R) Core (TM) Duo
processors ( http://www.intel.com/products/processor/coreduo/ ). And,
why isn't that (TM) after the word Duo? I'm pretty sure
I currently have an XP2600 and 6200 video card that seems to have enough
processing power for HD in Windows, but requires xvmc in Linux. Due to the
stability and usability issues with xvmc, I would like to get my system to
the point where I will not need to use it.
Most of what has been
On Jan 22, 2006, at 5:11 PM, Gary Manning wrote:
I currently have an XP2600 and 6200 video card that seems to have
enough
processing power for HD in Windows, but requires xvmc in Linux.
Due to the
stability and usability issues with xvmc, I would like to get my
system to
the point
On Sun, Jan 22, 2006 at 05:42:50PM -0700, Brian Wood wrote:
Nobody ever complained about having too much CPU, or RAM, I'd go for
the best I could afford.
Sometimes all the CPU you can afford choice gets you into the more
fan noise than you can stand configuration.
As to RAM, you'd do well
On Jan 22, 2006, at 6:03 PM, Trey Boudreau wrote:
On Sun, Jan 22, 2006 at 05:42:50PM -0700, Brian Wood wrote:
Nobody ever complained about having too much CPU, or RAM, I'd go for
the best I could afford.
Sometimes all the CPU you can afford choice gets you into the more
fan noise than you
From: Brian Wood [EMAIL PROTECTED]
Reply-To: Discussion about mythtv mythtv-users@mythtv.org
To: Discussion about mythtv mythtv-users@mythtv.org
Subject: Re: [mythtv-users] Performance from a different perspective
Date: Sun, 22 Jan 2006 17:42:50 -0700
On Jan 22, 2006, at 5:11 PM, Gary Manning
On Sun, Jan 22, 2006 at 06:32:27PM -0700, Brian Wood wrote:
On Jan 22, 2006, at 6:03 PM, Trey Boudreau wrote:
As to RAM, you'd do well to follow my First Law of Computing:
Thou shalt not swap.
Perhaps, but if you find yourself with unused RAM you could always
use it as ramdisks
19 matches
Mail list logo
