Welcome, Guest. Please login or register.
Amiga Kit Amiga Store Iridium Banner AMIStore App Store A600 Memory

AuthorTopic: lame benchmarks (pun intended)  (Read 8426 times)

0 Members and 1 Guest are viewing this topic.

Offline Piru

  • \' union select name,pwd--
  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 6946
  • Total likes: 0
    • http://www.iki.fi/sintonen/
lame benchmarks (pun intended)
« on: February 02, 2012, 05:50:04 PM »
AmigaONE X1000 has been benchmarked against Mac mini G4 1.5GHz and AmigaONE 500 460EX 1.15GHz with lame (mp3 encoder). The graph shows some interesting results:

X1000 appears to crush the competition, even with only single core enabled.

Except that interestingly the MorphOS results are from a lame build without AltiVec SIMD support. Why is that significant you may ask? The AltiVec accelerated version is significantly faster than the scalar (non-AltiVec) one. See what happens when you run the tests with the proper AltiVec enabled lame:
« Last Edit: February 02, 2012, 07:45:25 PM by Piru »
 

Offline jorkany

Re: lame benchmarks (pun intended)
« Reply #1 on: February 02, 2012, 06:05:15 PM »
An invalid comparison. The PPC Macs are outdated hardware from 2006, the X1000 is brand new hardware! As such the X1000 should be compared to current hardware.

From the A-eon website:
Quote
The X1000 ends the years of AmigaOS being relegated to a ghetto of outdated hardware - great as it was at the time, the world has moved on a long way since the days of Commodore. For the first time in many years, AmigaOS has a genuinely modern hardware platform.
 

Offline kickstart

Re: lame benchmarks (pun intended)
« Reply #2 on: February 02, 2012, 06:52:21 PM »
x1000 need some propaganda, real or false, for excuse the overprice.
a1200 060
 

Offline pampers

Re: lame benchmarks (pun intended)
« Reply #3 on: February 02, 2012, 06:56:07 PM »
Piru - any chance to benchmark it under experimental MorphOS on G5 Mac?
 

Offline jorkany

Re: lame benchmarks (pun intended)
« Reply #4 on: February 02, 2012, 07:08:54 PM »
Quote from: pampers;678795
Piru - any chance to benchmark it under experimental MorphOS on G5 Mac?

Piru, also could you link the source audio file you used in the conversion?
 

Offline Piru

  • \' union select name,pwd--
  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 6946
  • Total likes: 0
    • http://www.iki.fi/sintonen/
Re: lame benchmarks (pun intended)
« Reply #5 on: February 02, 2012, 07:14:35 PM »
Quote from: jorkany;678798
Piru, also could you link the source audio file you used in the conversion?

http://sintonen.fi/temp/AKsack.wav (md5sum: 5f644fbe4fe60211f8f1dde60e55dd53)

To benchmark it, download the file, and then issue the following command:

Code: [Select]
lame AKsack.wav
(with MorphOS make sure you're using the lame_vmx binary if you have altivec!)
« Last Edit: February 02, 2012, 08:00:29 PM by Piru »
 

Offline Hans_

Re: lame benchmarks (pun intended)
« Reply #6 on: February 02, 2012, 07:16:07 PM »
@Piru

According to the lame readme on os4depot, the AmigaOS4 version also lacks altivec support (the porter couldn't get it working properly). The A1-X1000's CPU has an altivec unit.

Hans
http://hdrlab.org.nz/ - Amiga OS 4 projects, programming articles and more. Home of the RadeonHD driver for Amiga OS 4.x project.
 

Offline wawrzon

Re: lame benchmarks (pun intended)
« Reply #7 on: February 02, 2012, 07:39:36 PM »
@piru:
i knew it! what a mess! so, where is the file, clear instructions and finally g5 result? hurry up!
 

Offline Piru

  • \' union select name,pwd--
  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 6946
  • Total likes: 0
    • http://www.iki.fi/sintonen/
Re: lame benchmarks (pun intended)
« Reply #8 on: February 02, 2012, 07:54:09 PM »
Quote from: Hans_;678802
@Piru

According to the lame readme on os4depot, the AmigaOS4 version also lacks altivec support (the porter couldn't get it working properly). The A1-X1000's CPU has an altivec unit.

Hans
The benchmark author claims he used the altivec version (I can only presume he used the older Stephan Rupprecht's version included in the said archive: lame-398-4/bin/lame.g4-3.98.2)

In english this post says: "Of course, why should I use other version while my CPU has AltiVec. Don't ask me about MorphOS, but I assume that they also used AV lame on G4 machines"

As far as I remember the PA6-T AltiVec indeed is slower than the one found from E600 (G4). It's somewhere close to 970 (G5) IIRC. Per clock the e600 altivec is the fastest implementation to date.
« Last Edit: February 02, 2012, 08:03:04 PM by Piru »
 

Offline kas1e

Re: lame benchmarks (pun intended)
« Reply #9 on: February 02, 2012, 07:54:42 PM »
@Piru
Can you please make a graph with all the non-altivec lame tests ? Because currently i can't understand the reall differences beetwen macs and x1000. Only what i see, is that non-altivec version on x1000 are the same by speed as macs with altivec one, but then seems non-altivec version on macs will be slower in 2 times in compare with x1000 one (even powermac ones).
 

Offline krashan

  • Full Member
  • ***
  • Join Date: Jan 2003
  • Posts: 245
  • Country: pl
  • Total likes: 0
  • Gender: Male
  • Hardware designer and programmer
    • Personal homepage
Re: lame benchmarks (pun intended)
« Reply #10 on: February 02, 2012, 08:03:04 PM »
Quote from: kas1e;678807
Can you please make a graph with all the non-altivec lame tests ? Because currently i can't understand the reall differences beetwen macs and x1000. Only what i see, is that non-altivec version on x1000 are the same by speed as macs with altivec one
As Piru linked (and I can confirm as a native Polish speaker), Mufa states that he used AltiVec accelerated LAME on X1000. Then on Piru graph all results except of AmigaOne 500 of course, are for AltiVec accelerated LAME.
 

Offline Piru

  • \' union select name,pwd--
  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 6946
  • Total likes: 0
    • http://www.iki.fi/sintonen/
Re: lame benchmarks (pun intended)
« Reply #11 on: February 02, 2012, 08:05:26 PM »
Quote from: jorkany;678790
The PPC Macs are outdated hardware from 2006

The 10+ year old 500MHz PowerMac is beating the AmigaOne 500, too. Oh the humanity!
 

Offline vox

Re: lame benchmarks (pun intended)
« Reply #12 on: February 02, 2012, 08:15:23 PM »
So we know two things:
- Altivec use is great and speeds up some multimedia demanding tasks 2x
- X1000 is champion in same fair category of non altivec use, in this case. Good, so its worthwhile after all as "new miggy champ"
Future Acube and MOS supporter, fi di good, nothing fi di unprofessionals. Learn it harder way! http://www.youtube.com/user/rasvoja and https://www.facebook.com/rasvoja
 

Offline kas1e

Re: lame benchmarks (pun intended)
« Reply #13 on: February 02, 2012, 08:16:10 PM »
@Krashan
Quote
As Piru linked (and I can confirm as a native Polish speaker), Mufa states that he used AltiVec accelerated LAME on X1000. Then on Piru graph all results except of AmigaOne 500 of course, are for AltiVec accelerated LAME.

Hardly can belive than matherboard with 1800mhz cpu, can be the same by speed on the lame tests as mac mini with 1.4. Imho mufa do something wrong in that terms.. Or, it will be epic fail, but still, i think that non-altivec version was used even he say that he use altivec one. Can you ask him what he download, from where, and how he run it and so on.
 

Offline vox

Re: lame benchmarks (pun intended)
« Reply #14 on: February 02, 2012, 08:23:17 PM »
Quote from: Piru;678812
The 10+ year old 500MHz PowerMac is beating the AmigaOne 500, too. Oh the humanity!


Its clear AMCC CPUs are not performance wise and meant for integrated combos and not desktops. Or software is just not optimized for them since they should have multimedia instructions of their own but not as standard as Altivec (have 24 additional digital signal processing instructions). However, SAM460 does make good use of fast RAM, SATA and PCI-E and will try to balance the loss there to make it overall feel the same / faster execept in any CPU intensive tasks.

Previous benches like Ragemem tests at AW,nt showed it, but not that 500Mhz G4 can beat SAM 460 http://httwww.amigaworld.net/modules/newbb/viewtopic.php?forum=14&topic_id=32815&viewmode=thread&order=0

That is a bit surprising!
Future Acube and MOS supporter, fi di good, nothing fi di unprofessionals. Learn it harder way! http://www.youtube.com/user/rasvoja and https://www.facebook.com/rasvoja