Is this machine assembled/manufactured somewhere in Europe?

@orb85750

Quote:

Is this machine assembled/manufactured somewhere in Europe?

Probably in UK (Varisys is manufacturer). Parts are from other countries.


and of course...

Welcome!

@Fab

Is in this command line

Quote:

mplayer -benchmark -nosound -ao null -vo null -lavdopts skiploopfilter=none {file}

the "-lavdopts" a mandatory switch? What does it do exactly? Bacause I wanted to run the same test on my AROS boxes, but AROS' MPlayer doesn't seem to understand this -lavdopts thing. Btw, on my development system I get about 71 seconds in a virtual machine. Not too bad...

@paolone

It makes sure the loopfilter won't be disabled. It's not by default, but some people might have put the skiploopfilter=all or so in their config file, which would have compromised the results.

And this option should definitely work on AROS too. This option exists for years in MPlayer, and since the AROS port is based on the MorphOS port, i don't see any reason why it wouldn't work.

What CPU does your virtual machine run on?

Last edited by Fab on 04-Feb-2012 at 05:57 PM.

@Fab

Thanks for your answer. I will try again (maybe I had mispelled the command line). I know the history of AROS MPlayer, I simply wasn't sure that option was included as well, but anyway thanks for your kind explanation. Physical processor running my AROS virtual machines is a 1.30 GHz core2duo U7300 from Intel. There is much overhead due to VMware Workstation 8 (which is not as fast as former releases, although update 8.0.1 has made things much better) and to Icaros services running in the background (I should disable them before running tests, yes). I should do some testing also on my 1.6 GHz Atom netbook and my Athlon64 X2 5200+ desktop machine, and post results.

@myself and Fab

Ooook, I've just repeated the test with the right command line and it worked fine. This is the result under VMware, whilist physical processor is the 1.30 GHz Core2Duo U7300:

BENCHMARKs: VC: 42.689s VO: 10.619s A: 0.000s Sys: 0.580s = 53.888s
BENCHMARK%: VC: 79.2179% VO: 19.7062% A: 0.0000% Sys: 1.0759% = 100.0000%

not that bad...

This is from my netbook (1.6 Ghz Atom processor)

BENCHMARKs: VC: 72.710s VO: 17.958s A: 0.000s Sys: 0.727s = 91.396s
BENCHMARK%: VC: 79.5553% VO: 19.6489% A: 0.0000% Sys: 0.7959% = 100.0000%


...and this from my Athlon64 X2 5200+ (2.6 Ghz) Icaros desktop computer:

BENCHMARKs: VC: 28.315s VO: 7.185s A: 0.000s Sys: 0.398s = 35.897s
BENCHMARK%: VC: 78.8774% VO: 20.0146% A: 0.0000% Sys: 1.1080% = 100.0000%

Last edited by paolone on 04-Feb-2012 at 11:47 PM.

finally finished with moving A1 -> X1000. System feels sure faster, specially when browsing with muiOWB on facebook. Now its actually enjoyable :) Couldnt get mplayer to work. Seems problem with pip so i changed to -vo cgx_wpa and it worked. So now i can watch 720p movies. 1080p cant play that good. More to try out and see what works and not.

A good thing is hd-rec and usb midi works like before so i can continue to make music. Soon ill sell my old A1G4Xe.

@Fab

Quote:

by Fab on 4-Feb-2012 13:27:18 @pavlor Quote: PA6T can play HD video (as we saw from the Fab´s benchmark) Well, no, it can't. The clip duration was 69s. So there's no time left to actually decode audio, blit the video and output the audio data (except if you disable loopfilter). And without overlay, the blitting stage would take substancially more time, too. Should be enough for 720p, though (like macs are).

Will improved video card drivers for OS4.2 in the future improve video playback or streaming video performance? Or is this totally dependent on the PA6T and the design of the X1000's motherboard and components that make it unable to process and display 1080p video content? (I know you can't predict how much video card drivers can be improved in the future, but is the current state of OS4 video card drivers one of the limiting factors preventing 1080p video playback?)

Last edited by amigadave on 05-Feb-2012 at 04:47 AM.
Last edited by amigadave on 05-Feb-2012 at 04:46 AM.
Last edited by amigadave on 05-Feb-2012 at 04:45 AM.

_________________
Amiga! The computer that inspired so many, to accomplish so much, but has ended up in the hands of . . . . . . . . . .

@Tuxedo

Quote:

Tuxedo wrote: @A1XK users can you plz try that pictures iwith my app LoVIew plz? http://photojournal.jpl.nasa.gov/jpeg/PIA08353.jpg http://www.v12-gt.com/var/v12gt/storage/images/galerie/galeries-photos/galeries-gt/alfa-romeo/alfa-romeo-8c-spider-blanc-jante/60416-1-fre-FR/Alfa-Romeo-8C-Spider-blanc-jante.jpg plz remember to add "USEWPA" ToolType since AOS4.x picture.datatype cant handle so big pictures... My Peg2@1131 takes 11.3 secs to load and display moon and about 5.2 secs loading the car... Plz dont compare with MOS since MOS 2D subsystem was proved to be fast and than not a valid reference(here oin Peg2 was almost twice faster loadinfg the moon ) Sop an only AmigaOS4.x bench... Thank you.

Loading and display the moon takes here under 5 sec

Last edited by tommysammy on 05-Feb-2012 at 06:38 AM.

_________________
Amiga600/Vampire2/PrismaMegaMix

@amigadave

Quote:

Will improved video card drivers for OS4.2 in the future improve video playback or streaming video performance?

I think both drivers and MPlayer could be improved.

However, as FAB wrote, 1080p may work with skiploopfilter option (worser quality of some movies).

@tommysammy

Quote:

Loading and display the moon takes here under 5 sec

Nice!

What about benchmarks I proposed in post 300? (I know I´m a little bit persistent in this DosBox area...)

@amigadave

With the introduction of PCIe ports connecting graphics card directly to SoC there isn't a lot you can do to introduce bottlenecks in a motherboard design, really. So it's entirely down to processing power. The graphics card could assist if there were drivers, but last time I tried to get the people who develop media players for AmigaOS to look into GPU assisting CPU on video playback the answer was "that will never help as the SAM I have has a slow PCI bus".

Well, what do we see now? The hardware is there and the software is still lagging behind. Maybe set up a bounty to have some GPU assistance on video playback now that we have the bandwidth but not the processing power?

That would sure help a LOT of things.

Actually, having assistance from the second core would help as well. Give developers a way to use the second core with a simple scheduler that takes standalone assembly functions if the second core exists but is not used by OS.

We need to think outside the box here. I could whip up a dual-core design around 1.5GHz with x8 pcie 2.0 (same throughput as original x16 PCIe in a few months. But without some way of utilizing the second core and the GPU that would just be too slow and too many people would throw #### at me. So I won't. For now.

But to get back to the point at hand: We need to offload mpeg 4 functions to the GPU. The sooner the better.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@tommysammy

that was a really nice result :)
that was about 60% faster tha on my system so about 70% faster that on G4@1GHZ :)
I've to try it out with the Updsate5 we it will be relased for "standard" OS4.x hw to test if the time was changing due to system opts also here, but for the moment I think it was anice result :)

@olegil

For sure we NEED GPU decoding for HD movies! We have the hw now and NEED the software!
I think that a bounty for that was needed!
Or better....since AFAIR mplayer was single core in original we may need a vlc that can use multi.cores and have integrated the GPU assistance(if portable)...

_________________
Simone"Tuxedo"Monsignori, Perugia, ITALY.

@tommysammy

Quote:

tommysammy wrote: @Tuxedo Quote: Tuxedo wrote: @A1XK users can you plz try that pictures iwith my app LoVIew plz? http://photojournal.jpl.nasa.gov/jpeg/PIA08353.jpg http://www.v12-gt.com/var/v12gt/storage/images/galerie/galeries-photos/galeries-gt/alfa-romeo/alfa-romeo-8c-spider-blanc-jante/60416-1-fre-FR/Alfa-Romeo-8C-Spider-blanc-jante.jpg plz remember to add "USEWPA" ToolType since AOS4.x picture.datatype cant handle so big pictures... My Peg2@1131 takes 11.3 secs to load and display moon and about 5.2 secs loading the car... Plz dont compare with MOS since MOS 2D subsystem was proved to be fast and than not a valid reference(here oin Peg2 was almost twice faster loadinfg the moon ) Sop an only AmigaOS4.x bench... Thank you. Loading and display the moon takes here under 5 sec

Loading and display the car needs 3,5 sec

_________________
Amiga600/Vampire2/PrismaMegaMix

Quote:

tommysammy wrote: @tommysammy Quote: tommysammy wrote: @Tuxedo [quote] Tuxedo wrote: @A1XK users can you plz try that pictures iwith my app LoVIew plz? http://photojournal.jpl.nasa.gov/jpeg/PIA08353.jpg http://www.v12-gt.com/var/v12gt/storage/images/galerie/galeries-photos/galeries-gt/alfa-romeo/alfa-romeo-8c-spider-blanc-jante/60416-1-fre-FR/Alfa-Romeo-8C-Spider-blanc-jante.jpg plz remember to add "USEWPA" ToolType since AOS4.x picture.datatype cant handle so big pictures... My Peg2@1131 takes 11.3 secs to load and display moon and about 5.2 secs loading the car... Plz dont compare with MOS since MOS 2D subsystem was proved to be fast and than not a valid reference(here oin Peg2 was almost twice faster loadinfg the moon ) Sop an only AmigaOS4.x bench... Thank you. Loading and display the moon takes here under 5 sec

Loading and display the car needs 3,5 sec[/quote]
The pictures are loading and displaying with picshow more faster

_________________
Amiga600/Vampire2/PrismaMegaMix

With WarpView in Softwaremode the pictures need only under 3sec for loading and displaying

_________________
Amiga600/Vampire2/PrismaMegaMix

@olegil

The newer gfx cards have built in 1080p decoding. But i guess the wallet needs to be big to get amd to write drivers/give us required docs.

_________________
I'm an antique. Don't light my fuse

@Antique

Unless we know some Amigan (ex is enough) at AMD and could convice him to do the needed binary...

(+arrange prebuilt dev system + bounty for the driver)

+IIRC, in future it should be possible to use Gallium3D to access the GPU's ebgines to accelerate video decoding.

Last edited by KimmoK on 05-Feb-2012 at 11:38 AM.
Last edited by KimmoK on 05-Feb-2012 at 11:35 AM.

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

@paolone

AROS/x86 natively (AMD Phenom II X4 955, 3.2 GHz):

BENCHMARKs: VC: 37.467s VO: 9.164s A: 0.000s Sys: 0.764s = 47.394s
BENCHMARK%: VC: 79.0539% VO: 19.3349% A: 0.0000% Sys: 1.6113% = 100.0000%

Looking at your values: Could it be the Phenom is still running at reduced speed here (powersaving)?

_________________
amiga-news.de