Click Here
home features news forums classifieds faqs links search
6005 members 
Amiga Q&A /  Free for All /  Emulation /  Gaming / (Latest Posts)
Login

Nickname

Password

Lost Password?

Don't have an account yet?
Register now!

Support Amigaworld.net
Your support is needed and is appreciated as Amigaworld.net is primarily dependent upon the support of its users.
Donate

Menu
Main sections
Home
Features
News
Forums
Classifieds
Links
Downloads
Extras
OS4 Zone
IRC Network
AmigaWorld Radio
Newsfeed
Top Members
Amiga Dealers
Information
About Us
FAQs
Advertise
Polls
Terms of Service
Search

IRC Channel
Server: irc.amigaworld.net
Ports: 1024,5555, 6665-6669
SSL port: 6697
Channel: #Amigaworld
Channel Policy and Guidelines

Who's Online
20 crawler(s) on-line.
 22 guest(s) on-line.
 0 member(s) on-line.



You are an anonymous user.
Register Now!
 nikosidis:  20 mins ago
 pixie:  25 mins ago
 BSzili:  36 mins ago
 matthey:  38 mins ago
 MEGA_RJ_MICAL:  1 hr 30 mins ago
 The_Shadow:  2 hrs 19 mins ago
 Fl@sh:  2 hrs 26 mins ago
 Trekiej:  2 hrs 35 mins ago
 DiscreetFX:  2 hrs 55 mins ago
 simplex:  3 hrs 37 mins ago

/  Forum Index
   /  Amiga OS4 Hardware
      /  X1000 - It's arrived...
Register To Post

Goto page ( Previous Page 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 Next Page )
PosterThread
stevieu 
Re: X1000 - It's arrived...
Posted on 3-Feb-2012 16:51:21
#261 ]
Cult Member
Joined: 23-Apr-2003
Posts: 647
From: England, UK

@Fab

Quote:

Here are my results from a Pegasos 2 and a Mac mini (that happens to have an uptime of 36 days), so i wouldn't mind other results to consolidate the values anyway. :)

Pegasos 2 / G4:
BENCHMARKs: VC: 153.620s VO: 0.021s A: 0.000s Sys: 2.313s =155.955s
BENCHMARK%: VC: 98.5032% VO: 0.0136% A: 0.0000% Sys: 1.4832% = 100.0000%

Mac mini 1.5GHz:
BENCHMARKs: VC: 84.626s VO: 0.009s A: 0.000s Sys: 0.679s = 85.314s
BENCHMARK%: VC: 99.1935% VO: 0.0111% A: 0.0000% Sys: 0.7954% = 100.0000%



And a bit off-topic, but on the Mac mini, with skiploopfilter=all, i get:
BENCHMARKs: VC: 69.425s VO: 0.009s A: 0.000s Sys: 0.679s = 70.113s
BENCHMARK%: VC: 99.0183% VO: 0.0134% A: 0.0000% Sys: 0.9683% = 100.0000%

Considering the clip duration is 69 seconds, the mac mini result gets rather interesting: it can actually decode the video stream in realtime. Of course, with additional audio AAC decoding and video blit, it wouldn't make it, but it's not *that* far away (the blitting routine is quite optimized in MPlayer).

And after a test with display enabled (using overlay output, -vo cgx_overlay):
BENCHMARKs: VC: 69.429s VO: 17.409s A: 0.000s Sys: 0.688s = 87.526s
BENCHMARK%: VC: 79.3235% VO: 19.8905% A: 0.0000% Sys: 0.7861% = 100.0000%



To be a bit different, I'll post a picture. Here are the X1000 results I got from each executable I have. The top result from each 'MUI MPlayer' executable uses -skiploopfilter=none and the bottom,-skiploopfilter=all.



Steve

Last edited by stevieu on 03-Feb-2012 at 04:56 PM.

_________________
A1200T - OS4.0,OS3.9: 603e PPC 200mhz,060 50mhz, 256mb ram, FastATA MK-III, BVision, 160gb,20gb HDDs

A1200 - OS3.1: Blizzard IV 030, 64mb ram, 400mb HDD

OS4.x - Flying the AMIGA flag

 Status: Offline
Profile     Report this post  
Fab 
Re: X1000 - It's arrived...
Posted on 3-Feb-2012 17:16:51
#262 ]
Super Member
Joined: 17-Mar-2004
Posts: 1178
From: Unknown

@stevieu

Ah, thank you. So this gives us 69s vs 84s with altivec. This is a more reasonable. :)

Now the problem is i don't see where the result from Tommy comes, unless he used generic version with skiploopfilter=all. :)

With a proper overlay driver and p96pip output and assuming you get a good cpu->vram speed (or through DMA otherwise), you might actually be able to play it realtime, with altivec and skiploopfilter=all.

Last edited by Fab on 03-Feb-2012 at 05:20 PM.
Last edited by Fab on 03-Feb-2012 at 05:18 PM.
Last edited by Fab on 03-Feb-2012 at 05:18 PM.

 Status: Offline
Profile     Report this post  
andres 
Re: X1000 - It's arrived...
Posted on 3-Feb-2012 17:59:23
#263 ]
Regular Member
Joined: 3-Nov-2008
Posts: 329
From: Firenze (Italy)

It would be interesting to have the same test suggested by Fab for AmigaOS/Pegasos 2 G4@1GHz, possibly with the same MPlayer version used by stevieu (69s).

Last edited by andres on 03-Feb-2012 at 06:00 PM.
Last edited by andres on 03-Feb-2012 at 05:59 PM.

_________________
A1200/020+68882 - 6 MB RAM - AmigaOS 3.0

Home Recording Audio

 Status: Offline
Profile     Report this post  
TheKorn 
Re: X1000 - It's arrived...
Posted on 3-Feb-2012 18:37:12
#264 ]
Regular Member
Joined: 15-Oct-2008
Posts: 169
From: Texas

@andres

Here is another Peg2 result using Fab's suggestion (mplayer -benchmark -nosound -ao null -vo null -lavdopts skiploopfilter=none file)


BENCHMARKs: VC: 171.536s VO: 0.019s A: 0.000s Sys: 1.634s = 173.189s
BENCHMARK%: VC: 99.0456% VO: 0.0107% A: 0.0000% Sys: 0.9437% = 100.0000%

_________________
Talos BlackBird (awaiting a Miggy like OS)
Amiga 4000 3.9 / Pegasos II 4.1 F.E.

 Status: Offline
Profile     Report this post  
sundown 
Re: X1000 - It's arrived...
Posted on 3-Feb-2012 20:09:04
#265 ]
Elite Member
Joined: 30-Aug-2003
Posts: 5119
From: Right here...

@Rose

Quote:
GFX driver DOESN'T affect on raytrace rendering speed since it's used only for output, not on actual rendering.

Yes, I understand that & I would love to run that benchmark, but Andy (broadblues) has told me the gui code in that file requires a card with miniGL support. There is no way to run it yet, unless someone has a 9250 PCIe x8 or x16 card installed. I do have an older PCI 9250, but the PCI bus is slower then the PCIe bus. I did find some PCIe 9250's on ebay, but short on money for awhile to buy one.

_________________
Hate tends to make you look stupid...

 Status: Offline
Profile     Report this post  
sundown 
Re: X1000 - It's arrived...
Posted on 3-Feb-2012 20:14:22
#266 ]
Elite Member
Joined: 30-Aug-2003
Posts: 5119
From: Right here...

@Fab

Quote:
To be sure altivec was indeed used

I was told that altivec was not "turned on" yet on the x1000. Maybe there's a simple way to test that?

_________________
Hate tends to make you look stupid...

 Status: Offline
Profile     Report this post  
Tuxedo 
Re: X1000 - It's arrived...
Posted on 3-Feb-2012 20:30:22
#267 ]
Elite Member
Joined: 28-Nov-2003
Posts: 2326
From: Perugia, ITALY

@Fab

here my Peg2 G4@1131 mui-mplayer altivec:

BENCHMARKs: VC: 149.981s VO: 0.018s A: 0.000s Sys: 1.586s = 151.585s
BENCHMARK%: VC: 98.9416% VO: 0.0120% A: 0.0000% Sys: 1.0463% = 100.0000%


@sundown

if AltiVec wasnt on that maybe was a good point...
However I think that the easiest way to test that was to run lame and lmaeG$ from the os4depot archive so ve can see if effectively they give us differnete or similar times...

also trying to load mui-mplayer altivec and generic...

However the steveu test seems to tell us that altivec as on...

Last edited by Tuxedo on 03-Feb-2012 at 08:31 PM.

_________________
Simone"Tuxedo"Monsignori, Perugia, ITALY.

 Status: Offline
Profile     Report this post  
Rob 
Re: X1000 - It's arrived...
Posted on 3-Feb-2012 21:46:28
#268 ]
Elite Member
Joined: 20-Mar-2003
Posts: 6040
From: S.Wales

@TheKorn

That's 11 seconds quicker than the same test on my 1Ghz XE.

 Status: Offline
Profile     Report this post  
sundown 
Re: X1000 - It's arrived...
Posted on 3-Feb-2012 22:55:55
#269 ]
Elite Member
Joined: 30-Aug-2003
Posts: 5119
From: Right here...

@Tuxedo

Quote:
However the steveu test seems to tell us that altivec as on...

Could be.

os4 update 5, FC
File=WhereDoAllTheJunkiesComeFrom.mp3, using lame

7.RAM Disk:la/bin> lame ram:w.mp3 wav
ID3v2 found. Be aware that the ID3 tag is currently lost when transcoding.
LAME 3.98.4 32bits (http://www.mp3dev.org/)
Using polyphase lowpass filter, transition band: 16538 Hz - 17071 Hz
Encoding ram:w.mp3 to wav
Encoding as 44.1 kHz j-stereo MPEG-1 Layer III (11x) 128 kbps qval=3
Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
9467/9473 (100%)| 0:40/ 0:40| 0:40/ 0:40| 6.1456x| 0:00
-------------------------------------------------------------------------------
kbps LR MS % long switch short %
128.0 3.6 96.4 88.5 6.9 4.6

Using lame.g4

7.RAM Disk:la/bin> lamea ram:w.mp3 wav
ID3v2 found. Be aware that the ID3 tag is currently lost when transcoding.
LAME 3.98.2 32bits (http://www.mp3dev.org/)
Using polyphase lowpass filter, transition band: 16537 Hz - 17071 Hz
Encoding ram:w.mp3 to wav
Encoding as 44.1 kHz j-stereo MPEG-1 Layer III (11x) 128 kbps qval=3
Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
9468/9473 (100%)| 0:23/ 0:23| 0:23/ 0:23| 10.462x| 0:00
-------------------------------------------------------------------------------
kbps LR MS % long switch short %
128.0 3.6 96.4 88.5 6.9 4.6

Last edited by sundown on 03-Feb-2012 at 11:00 PM.

_________________
Hate tends to make you look stupid...

 Status: Offline
Profile     Report this post  
stevieu 
Re: X1000 - It's arrived...
Posted on 4-Feb-2012 1:17:47
#270 ]
Cult Member
Joined: 23-Apr-2003
Posts: 647
From: England, UK

@Fab

No problem.

I just ran the test, as you said.

Steve

_________________
A1200T - OS4.0,OS3.9: 603e PPC 200mhz,060 50mhz, 256mb ram, FastATA MK-III, BVision, 160gb,20gb HDDs

A1200 - OS3.1: Blizzard IV 030, 64mb ram, 400mb HDD

OS4.x - Flying the AMIGA flag

 Status: Offline
Profile     Report this post  
stevieu 
Re: X1000 - It's arrived...
Posted on 4-Feb-2012 1:19:16
#271 ]
Cult Member
Joined: 23-Apr-2003
Posts: 647
From: England, UK

@Tuxedo

The picture shows the results from the generic version and the altivec enabled version.

Steve

Last edited by stevieu on 04-Feb-2012 at 04:01 AM.
Last edited by stevieu on 04-Feb-2012 at 01:22 AM.
Last edited by stevieu on 04-Feb-2012 at 01:21 AM.

_________________
A1200T - OS4.0,OS3.9: 603e PPC 200mhz,060 50mhz, 256mb ram, FastATA MK-III, BVision, 160gb,20gb HDDs

A1200 - OS3.1: Blizzard IV 030, 64mb ram, 400mb HDD

OS4.x - Flying the AMIGA flag

 Status: Offline
Profile     Report this post  
gregthecanuck 
Re: X1000 - It's arrived...
Posted on 4-Feb-2012 5:26:58
#272 ]
Cult Member
Joined: 30-Dec-2003
Posts: 846
From: Vancouver, Canada

Any X1000 owners care to test out the OGR-NG/RC5-72 projects from distributed.net? Run the benchmark - all cores - just for grins.

Would be interested in seeing how it handles the X1000 and if it recognizes the Altivec.

The latest version is v2.9109.518 and available here.

Thanks

 Status: Offline
Profile     Report this post  
sundown 
Re: X1000 - It's arrived...
Posted on 4-Feb-2012 7:52:09
#273 ]
Elite Member
Joined: 30-Aug-2003
Posts: 5119
From: Right here...

@gregthecanuck

dnetc

_________________
Hate tends to make you look stupid...

 Status: Offline
Profile     Report this post  
pavlor 
Re: X1000 - It's arrived...
Posted on 4-Feb-2012 8:02:31
#274 ]
Elite Member
Joined: 10-Jul-2005
Posts: 9223
From: Unknown

@sundown

With or without AltiVec? (It would be interesting to see difference)

 Status: Offline
Profile     Report this post  
sundown 
Re: X1000 - It's arrived...
Posted on 4-Feb-2012 8:41:41
#275 ]
Elite Member
Joined: 30-Aug-2003
Posts: 5119
From: Right here...

@pavlor

From the readme;

The OS4 PPC version requires:
- A 603e or higher PPC processor.
- AmigaOS 4.0 pre-release or newer.
- AltiVec cores require a G4 and kernel 51.12 or higher.
- The Roadshow TCP/IP stack supplied with OS4.

There is only one file to run. From the lame test I did, it seem AltVec is working.

Its almost as fast as my dual core 1,5GHz laptop, would be faster if both core were working on the x. The laptop shows both cores working.

Last edited by sundown on 04-Feb-2012 at 08:45 AM.

_________________
Hate tends to make you look stupid...

 Status: Offline
Profile     Report this post  
g_kraszewski 
Re: X1000 - It's arrived...
Posted on 4-Feb-2012 8:48:14
#276 ]
Regular Member
Joined: 3-Sep-2010
Posts: 341
From: Unknown

@pavlor

Both "KOGE 3.1 Hybrid" and "KKS7540" cores use AltiVec. In my opinion dnetc test is another proof, that PA6T processor is not a rocket science and its core is slower "per MHz" compared to Freescale's e600:

RC5-72: X1000 at 1.8 GHz 10.2 Mkeys/s, Mac mini at 1.42 GHz 15.0 Mkeys/s
OGR-NG: X1000 at 1.8 GHz 23.2 Mnodes/s, Mac mini at 1.42 GHz 30.54 Mkeys/s

Then it seems to me X1000 wins memory bound tasks (thanks to its superior memory transfer speeds), but looses CPU bound tasks. Dnetc is a pure CPU bound one. LAME tests (mostly CPU bound task) and H.264 decoding tests (balance between CPU and memory bound) seems to confirm it. The other end is RAM: to RAM: copying test (mostly memory bound), where X1000 beats all G4 hardware.

_________________
RastPort

 Status: Offline
Profile     Report this post  
pavlor 
Re: X1000 - It's arrived...
Posted on 4-Feb-2012 9:20:25
#277 ]
Elite Member
Joined: 10-Jul-2005
Posts: 9223
From: Unknown

@g_kraszewski

Quote:
Then it seems to me X1000 wins memory bound tasks (thanks to its superior memory transfer speeds), but looses CPU bound tasks.


Yes, It seems G4/e600 has better AltiVec implementation.

Results in RC5/OGR show that PA6T 1.8 GHz is on the level of 970(FX?) 1.6 GHz when using AltiVec in this test.

However, PA6T is powerful also for some CPU intensive tasks - I was told it can run DosBox with more than 20000 CPU cycles (for comparison 7447 1 GHz can reach 7-8000 CPU cycles).

To summarize:
PA6T has splendid memory performance (caches, RAM)
PA6T can play HD video (as we saw from the Fabs benchmark)
PA6T has worser Altivec implementation than G4 (PA6T 1.8 GHz is on the level of G4 1.5 GHz or worser)
PA6T excells in some CPU intensive tasks (DosBox), but not in all (Blender).

Now Im convinced to buy X1000 for Christmas, depending on my financial situation of course. I need also space in my lair...

 Status: Offline
Profile     Report this post  
gregthecanuck 
Re: X1000 - It's arrived...
Posted on 4-Feb-2012 9:46:41
#278 ]
Cult Member
Joined: 30-Dec-2003
Posts: 846
From: Vancouver, Canada

@sundown

Thanks very much for running those tests. They make a good starting point for pure CPU performance evaluations.


@g_kraszweski

The PA6T core is very good when you consider the power draw required to get those results (max 7W@2GHz for *both* cores according to Wikipedia) vs. max 30W@1.42Ghz for the single core 7447A. That is close to 10% of the power draw of the 7447A (per core).

The engineers likely made tradeoffs on caching and/or pipelining to help get the power draw down and those tradeoffs are definitely exposed here.

However, those tradeoffs are balanced against a massively superior memory interface and I/O subsystem. Overall system performance should be better as a result. More benchmarks are needed, more time for the O/S and applications to mature are needed as well. The PA6T is relatively uncharted territory. And who knows, perhaps there are some compiler optimizations that could be made to make more effective use of the CPU.

It is a shame this CPU family didn't continue. It would have kicked Intel's @ss.

However it appears Freescale took notice and their current and upcoming multicore parts (T4240) bear a remarkably similar architecture. Low power consumption and substantial core improvements, both in MIPS/MHz and overall floating point performance.






 Status: Offline
Profile     Report this post  
pavlor 
Re: X1000 - It's arrived...
Posted on 4-Feb-2012 9:50:56
#279 ]
Elite Member
Joined: 10-Jul-2005
Posts: 9223
From: Unknown

@gregthecanuck

Quote:
The engineers likely made tradeoffs on caching and/or pipelining to help get the power draw down and those tradeoffs are definitely exposed here.


Exactly. I ever wanted to know how fast X1000 is in "real" applications. Sources on the web were scarce and not much helpful. Now we unveil nearly every aspect of the performance of this great CPU - thanks to the X1000 project.

 Status: Offline
Profile     Report this post  
itix 
Re: X1000 - It's arrived...
Posted on 4-Feb-2012 9:53:56
#280 ]
Elite Member
Joined: 22-Dec-2004
Posts: 3398
From: Freedom world

Oops, never mind :)

Last edited by itix on 04-Feb-2012 at 09:55 AM.

_________________
Amiga Developer
Amiga 500, Efika, Mac Mini and PowerBook

 Status: Offline
Profile     Report this post  
Goto page ( Previous Page 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 Next Page )

[ home ][ about us ][ privacy ] [ forums ][ classifieds ] [ links ][ news archive ] [ link to us ][ user account ]
Copyright (C) 2000 - 2019 Amigaworld.net.
Amigaworld.net was originally founded by David Doyle