A computer components & hardware forum. HardwareBanter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » HardwareBanter forum » General Hardware & Peripherals » Homebuilt PC's
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

Nice - HQ encode, 8-core bulldozer



 
 
Thread Tools Display Modes
  #1  
Old August 11th 17, 12:36 PM posted to alt.comp.hardware.pc-homebuilt
Flasherly[_2_]
external usenet poster
 
Posts: 2,407
Default Nice - HQ encode, 8-core bulldozer

Got around to a decent video encode, something I'd have prior, with
certain likelihood, run into issues, at the least with further audio
processing I also run. Quadcores then, octals now. Distribution
across eight cores can also be seen favorably, a token estimate at 20%
processing demands, on average -- less linearity to any code specifics
(modeling for a multi-core environ), than reliance on AMD's generic
distribution for cores. Not that 20% figure need be so radically
different for resolution across quadcore, either. Temps aren't
especially significant, or also decent.

What it does is become the biggest jump since an I'd abandoned an
Intel Duron single-core -- through and subsequent dualcores, including
maxing those to same-socket quadcores (w/ both AMD and Intel,
same-class MBs). For me that makes this my next-generation update, a
month or two ago, when deciding to update to an octal. As I can't
think of any other (program) usage I'll likely engage, offhand, more
demanding, or so tailored for a wider range of cores.

If it weren't for AMD, for all of what an Intel octal designates --
not much, actually, in any practical sense -- and for all of what is
pretty much left, Intel hyper-threading -- there's not much else,
being AMD hasn't a lot else to show for itself. A recent cheapening
Bulldozers for otherwise a nondescript achievement, at or below quads,
notably for those models [above] in application to make a further
distinction to marketing its Ryzen platform.

The Trickle-Down Effect, as it's at times also called.
  #2  
Old August 11th 17, 02:58 PM posted to alt.comp.hardware.pc-homebuilt
Yousuf Khan[_2_]
external usenet poster
 
Posts: 1,296
Default Nice - HQ encode, 8-core bulldozer

On 11/08/2017 7:36 AM, Flasherly wrote:
Got around to a decent video encode, something I'd have prior, with
certain likelihood, run into issues, at the least with further audio
processing I also run. Quadcores then, octals now. Distribution
across eight cores can also be seen favorably, a token estimate at 20%
processing demands, on average -- less linearity to any code specifics
(modeling for a multi-core environ), than reliance on AMD's generic
distribution for cores. Not that 20% figure need be so radically
different for resolution across quadcore, either. Temps aren't
especially significant, or also decent.


Yeah, I recently started using Windows Movie Maker a lot more than
usual. At the beginning of a project, WMM happens to reencode the entire
video into its own internal Mpeg-4 implementation (WMV), before it even
begins allowing you to edit it. You see a huge reencoding process go on
for half-hour depending on how big the video is. You see every core get
pegged at 100% during this time, which is a wondrous sight if you have 8
cores.

In fact, it has to redo this encoding everytime you restart the project,
so it's best to simply leave the project running in the background, and
to never shutdown your computer, just let it go into standby if you need
to save electricity.

Yousuf Khan

--
Sent from Giganews on Thunderbird on my Toshiba laptop
  #3  
Old August 11th 17, 09:09 PM posted to alt.comp.hardware.pc-homebuilt
Flasherly[_2_]
external usenet poster
 
Posts: 2,407
Default Nice - HQ encode, 8-core bulldozer

On Fri, 11 Aug 2017 09:58:51 -0400, Yousuf Khan
wrote:

Yeah, I recently started using Windows Movie Maker a lot more than
usual. At the beginning of a project, WMM happens to reencode the entire
video into its own internal Mpeg-4 implementation (WMV), before it even
begins allowing you to edit it. You see a huge reencoding process go on
for half-hour depending on how big the video is. You see every core get
pegged at 100% during this time, which is a wondrous sight if you have 8
cores.

In fact, it has to redo this encoding everytime you restart the project,
so it's best to simply leave the project running in the background, and
to never shutdown your computer, just let it go into standby if you need
to save electricity.

Yousuf Khan


I haven't largely needed to peg mine for any considerable length of
time. Other than audio-only manipulations, there's not much else I
can think of readily, that I have, to use each core to its fullest.

There's also a reverse-case proviso, in the BIOS, curiously to disable
cores in "paired" sets, reductively all the way down to a single core
processor. Two pairs, presumably, although I haven't yet engaged or
tried it.

Although audio decodes still have practical limits, which I set out to
find, independently and far before any such condition. A software
decoded playback condition that varies according to how audio is being
processed, for as many effects selected and added into the playback
(transient recovery, vacuum tube harmonics, etc.) Of course, that I
now have more such effects available is the plus side. An improved
sense for effect selections of which ones, overall among many more
capable of overloaded condition, at a minimum will add the most
integrity to improved sound fidelity, far before an "overload" CPU
condition or only in part a raw determinate of CPU limitations.

I'm not sure Intel hyper-threading might react differently to that, a
contingency of efficiency for code, however unique its distribution
means to cores.

In reading reviews of the AMD octal, I ran across one which said,
fully loaded and with the power-boost enabled, (as it is by default in
the BIOS for a 3.2Ghz to 4Ghz condition when met), that the FX-8300
then will draw 285 watts. That's a lot of power and then some...they
also overclock well at farther extremities.
  #4  
Old August 13th 17, 05:12 AM posted to alt.comp.hardware.pc-homebuilt
[email protected]
external usenet poster
 
Posts: 220
Default Nice - HQ encode, 8-core bulldozer

A company I worked for actually built a cluster of Bulldozer Opterons.
Perhaps the only one in the country that did?
They reckoned they went pretty well in Centos 6.
The head IT guy said he was using said CPU at home for video encoding.
  #5  
Old August 23rd 17, 04:38 PM posted to alt.comp.hardware.pc-homebuilt
Yousuf Khan[_2_]
external usenet poster
 
Posts: 1,296
Default Nice - HQ encode, 8-core bulldozer

On 11/08/2017 4:09 PM, Flasherly wrote:
In reading reviews of the AMD octal, I ran across one which said,
fully loaded and with the power-boost enabled, (as it is by default in
the BIOS for a 3.2Ghz to 4Ghz condition when met), that the FX-8300
then will draw 285 watts. That's a lot of power and then some...they
also overclock well at farther extremities.


The Bulldozers were extremely heat-tolerant, that's why they could be
overclocked so heavily, still no current Intel Core or AMD Ryzen
processor has matched the FX-9590 for stock core speeds yet, 4.7 & 5.0
GHz. But of course, you didn't actually get much extra performance even
at those high MHz, as the cores just spun their wheels wildly rather
than do actual work.

--
Sent from Giganews on Thunderbird on my Toshiba laptop
  #7  
Old August 23rd 17, 06:49 PM posted to alt.comp.hardware.pc-homebuilt
Flasherly[_2_]
external usenet poster
 
Posts: 2,407
Default Nice - HQ encode, 8-core bulldozer

On Wed, 23 Aug 2017 11:38:31 -0400, Yousuf Khan
wrote:

In reading reviews of the AMD octal, I ran across one which said,
fully loaded and with the power-boost enabled, (as it is by default in
the BIOS for a 3.2Ghz to 4Ghz condition when met), that the FX-8300
then will draw 285 watts. That's a lot of power and then some...they
also overclock well at farther extremities.


The Bulldozers were extremely heat-tolerant, that's why they could be
overclocked so heavily, still no current Intel Core or AMD Ryzen
processor has matched the FX-9590 for stock core speeds yet, 4.7 & 5.0
GHz. But of course, you didn't actually get much extra performance even
at those high MHz, as the cores just spun their wheels wildly rather
than do actual work.


And, still, not a bad price for all the esoterics of 220W at the upper
end of Bulldozers. Under $150/US. The review I'd mentioned (
lowest-wattage octal Bulldozers) is a bit sketchy now to fully recall:
He'd probably set it up for overclocked, and, second, used a suitable
program, a spread workload across all cores, in order to achieve the
285W extremity claimed reached.

A lot of processing power, no doubt, now at the higher end available.
And reasonable, too. Advanced and efficiently designed for smooth
sailing. As well more of specialized crowd, I'll bet, a "limited"
concern with cloud services and portable devices filling an interim
gap for actually building or learning computers, when simply operating
occasional programs and games is an affair of convenience alongside
social media or communications from hybrid smartphones and a likes.
 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
A quick look at Bulldozer thread scheduling Yousuf Khan[_2_] Intel 10 December 5th 11 05:35 PM
reason for AMD's Bulldozer fiasco? Orson Cart Intel 5 November 3rd 11 01:57 AM
AMD Bulldozer - arrives what decade? Kulin Remailer Intel 2 May 28th 11 04:21 AM
AIW MCE 2004 SW Encode Kit Don Ati Videocards 0 October 25th 04 01:45 AM
Is it possible to have ATI MMC's TV to encode in DivX 5.x format? [email protected] Ati Videocards 2 September 30th 03 05:25 AM


All times are GMT +1. The time now is 08:24 AM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 HardwareBanter.
The comments are property of their posters.