|
Home > Archive > Cheap Linux Hardware > August 2006 > Radeon 9200 is slower than on-board intel 815 graphic card
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Radeon 9200 is slower than on-board intel 815 graphic card
|
|
|
| Hallo
This fact seems quite strange to me.
Some details:
I do not play games, I mainly watch movies that are avi 8Mbps files with 50 fps. On built in 815 intel graphics card They are played a little bit too slow and on radeon 9200 they are even a bit slower. This is very strange to me.
I think that HD is not a problem:
hdparm -t /dev/hdc tells that drive speed is 30MB/s so it is far enough for those movies.
Machine: PIII 1Ghz, built in graphics card intel 815 that takes 64 MB memory if it is enabled.
I am running FC5 with standard ATI drivers that are in that distribution. It's glxinfo tool claims that DRI support is enabled for radeon 9200.
My question is which way to go to improove performance of the graphics card.
What do You think about installing proprietary radeon drivers from AVI web site?
Are there any other ways to improove its performance?
Greetings
Pawel
| |
| Henrik Carlqvist 2006-07-18, 7:15 pm |
| Pawel <null7@wp.pl> wrote:
> I do not play games, I mainly watch movies that are avi 8Mbps files with
> 50 fps. On built in 815 intel graphics card They are played a little bit
> too slow and on radeon 9200 they are even a bit slower. This is very
> strange to me.
What program are you using to play the movies? What method does the
program use to display the video?
> I am running FC5 with standard ATI drivers that are in that
> distribution. It's glxinfo tool claims that DRI support is enabled for
> radeon 9200.
>
> My question is which way to go to improove performance of the graphics
> card. What do You think about installing proprietary radeon drivers from
> AVI web site? Are there any other ways to improove its performance?
I use the opensource drivers from X.org with my Radeon 9200 and have no
problem watching avi or mpeg files. I usually use mplayer to watch the
files. I also have DRI enabled and mplayer works fine with the gl output
driver (-vo gl), mplayer seems not to work so well with the gl2 output
driver for me (-vo gl2). However, the best results is with the xv output
driver (-vo xv). This is alot faster than the x11 output driver.
Are you using the xv output driver? What does xvinfo say on your system?
It was a long time since I tried the binary ATI driver, but if I remember
right it did work worse not only for 3D but also for xvideo.
regards Henrik
--
The address in the header is only to prevent spam. My real address is:
hc8(at)uthyres.com Examples of addresses which go to spammers:
root@variousus.net root@localhost
| |
|
|
Hallo Henrik
Henrik Carlqvist <Henrik.Carlqvist@deadspam.com> writes:
> Pawel <null7@wp.pl> wrote:
>
> What program are you using to play the movies? What method does the
> program use to display the video?
I use mplayer with direct rendering enabled and -vo xv. DRI is enabled
for Radeon 9200.
>
>
> I use the opensource drivers from X.org with my Radeon 9200 and have no
> problem watching avi or mpeg files. I usually use mplayer to watch the
> files. I also have DRI enabled and mplayer works fine with the gl output
> driver (-vo gl), mplayer seems not to work so well with the gl2 output
> driver for me (-vo gl2). However, the best results is with the xv output
> driver (-vo xv). This is alot faster than the x11 output driver.
>
> Are you using the xv output driver? What does xvinfo say on your system?
Yes I use xv output driver.
>
> It was a long time since I tried the binary ATI driver, but if I remember
> right it did work worse not only for 3D but also for xvideo.
>
> regards Henrik
> --
> The address in the header is only to prevent spam. My real address is:
> hc8(at)uthyres.com Examples of addresses which go to spammers:
> root@variousus.net root@localhost
I wonder about recompiling mplayer to i686, maybe this will help.
Pawel
| |
| Henrik Carlqvist 2006-07-19, 7:32 am |
| Pawel <null7@wp.pl> wrote:
> I wonder about recompiling mplayer to i686, maybe this will help.
When compiling mplayer usually you don't have to tell any configure script
which platform you shall optimize for. Mplayer does a rather good job in
automatically detecting your current platform and optimize the compilation
for that platform. The drawback of this is that you can't use this
compiled mplayer on another machine with an older CPU.
I have some more suggestions about your 9200 card. I have optimized the
performance of my card with the following in xorg.conf:
Section "Device"
Identifier "Powercolor Radeon 9200"
Driver "radeon"
#VideoRam 262144
# Insert Clocks lines here if appropriate
Option "AGPMode" "4"
Option "EnablePageFlip" "on"
EndSection
Note the option lines with AGPMode and EnablePageFlip. I have X.org 6.8
which didn't support AGPMode faster than 4. If you have a more recent
X.org you might want to try to set AGPMode to 8 instead.
regards Henrik
--
The address in the header is only to prevent spam. My real address is:
hc8(at)uthyres.com Examples of addresses which go to spammers:
root@variousus.net root@localhost
| |
|
|
Hallo Henrik
Henrik Carlqvist <Henrik.Carlqvist@deadspam.com> writes:
> Pawel <null7@wp.pl> wrote:
>
> When compiling mplayer usually you don't have to tell any configure script
> which platform you shall optimize for. Mplayer does a rather good job in
> automatically detecting your current platform and optimize the compilation
> for that platform. The drawback of this is that you can't use this
> compiled mplayer on another machine with an older CPU.
I realize that. On the other hand I am using FC5 and there is a page
http://rpm.greysector.net/mplayer that has ready to use mplayer rpm binaries.
They are built for i386. There is also possibility of rebuilding it with
customized options (rebuild info on that page). The only thing I consider
is to change target option. This way I would have universal (it, is, i386)
version on their page while my build would be tergetted to i686.
>
> I have some more suggestions about your 9200 card. I have optimized the
> performance of my card with the following in xorg.conf:
>
> Section "Device"
> Identifier "Powercolor Radeon 9200"
> Driver "radeon"
> #VideoRam 262144
> # Insert Clocks lines here if appropriate
> Option "AGPMode" "4"
> Option "EnablePageFlip" "on"
> EndSection
I did as You suggested, both with 4 and 8 mode option. No change
unfortunatelly.
the result of xvinfo is the following:
prompt# xvinfo
X-Video Extension version 2.2
screen #0
Adaptor #0: "ATI Radeon Video Overlay"
number of ports: 1
port base: 73
operations supported: PutImage
supported visuals:
depth 16, visualID 0x23
depth 16, visualID 0x24
depth 16, visualID 0x25
depth 16, visualID 0x26
depth 16, visualID 0x27
depth 16, visualID 0x28
depth 16, visualID 0x29
depth 16, visualID 0x2a
depth 16, visualID 0x2b
depth 16, visualID 0x2c
depth 16, visualID 0x2d
depth 16, visualID 0x2e
depth 16, visualID 0x2f
depth 16, visualID 0x30
depth 16, visualID 0x31
depth 16, visualID 0x32
number of attributes: 22
"XV_DEVICE_ID" (range 0 to -1)
client gettable attribute (current value is 493)
"XV_LOCATION_ID" (range 0 to -1)
client gettable attribute (current value is 494)
"XV_INSTANCE_ID" (range 0 to -1)
client gettable attribute (current value is 495)
"XV_DUMP_STATUS" (range 0 to 1)
client settable attribute
"XV_SET_DEFAULTS" (range 0 to 1)
client settable attribute
"XV_AUTOPAINT_COLORKEY" (range 0 to 1)
client settable attribute
client gettable attribute (current value is 0)
"XV_COLORKEY" (range 0 to -1)
client settable attribute
client gettable attribute (current value is 30)
"XV_DOUBLE_BUFFER" (range 0 to 1)
client settable attribute
client gettable attribute (current value is 1)
"XV_OVERLAY_ALPHA" (range 0 to 255)
client settable attribute
client gettable attribute (current value is 255)
"XV_GRAPHICS_ALPHA" (range 0 to 255)
client settable attribute
client gettable attribute (current value is 255)
"XV_ALPHA_MODE" (range 0 to 1)
client settable attribute
client gettable attribute (current value is 0)
"XV_BRIGHTNESS" (range -1000 to 1000)
client settable attribute
client gettable attribute (current value is 0)
"XV_CONTRAST" (range -1000 to 1000)
client settable attribute
client gettable attribute (current value is 0)
"XV_SATURATION" (range -1000 to 1000)
client settable attribute
client gettable attribute (current value is 0)
"XV_COLOR" (range -1000 to 1000)
client settable attribute
client gettable attribute (current value is 0)
"XV_HUE" (range -1000 to 1000)
client settable attribute
client gettable attribute (current value is 0)
"XV_RED_INTENSITY" (range -1000 to 1000)
client settable attribute
client gettable attribute (current value is 0)
"XV_GREEN_INTENSITY" (range -1000 to 1000)
client settable attribute
client gettable attribute (current value is 0)
"XV_BLUE_INTENSITY" (range -1000 to 1000)
client settable attribute
client gettable attribute (current value is 0)
"XV_SWITCHCRT" (range 0 to 1)
client settable attribute
client gettable attribute (current value is 0)
"XV_GAMMA" (range 100 to 10000)
client settable attribute
client gettable attribute (current value is 1000)
"XV_COLORSPACE" (range 0 to 1)
client settable attribute
client gettable attribute (current value is 0)
maximum XvImage size: 2048 x 2048
Number of image formats: 8
id: 0x41424752 (RGBA)
guid: 52474241-0000-0010-8000-00aa00389b71
bits per pixel: 32
number of planes: 1
type: RGB (packed)
depth: 32
red, green, blue masks: 0xff0000, 0xff00, 0xff
id: 0x0
guid: 52474200-0000-0010-8000-00aa00389b71
bits per pixel: 24
number of planes: 1
type: RGB (packed)
depth: 24
red, green, blue masks: 0xff0000, 0xff00, 0xff
id: 0x54424752 (RGBT)
guid: 52474254-0000-0010-8000-00aa00389b71
bits per pixel: 16
number of planes: 1
type: RGB (packed)
depth: 16
red, green, blue masks: 0x7c00, 0x3e0, 0x1f
id: 0x32424752 (RGB2)
guid: 52474200-0000-0010-8000-00aa00389b71
bits per pixel: 16
number of planes: 1
type: RGB (packed)
depth: 16
red, green, blue masks: 0xf800, 0x7e0, 0x1f
id: 0x32595559 (YUY2)
guid: 59555932-0000-0010-8000-00aa00389b71
bits per pixel: 16
number of planes: 1
type: YUV (packed)
id: 0x59565955 (UYVY)
guid: 55595659-0000-0010-8000-00aa00389b71
bits per pixel: 16
number of planes: 1
type: YUV (packed)
id: 0x32315659 (YV12)
guid: 59563132-0000-0010-8000-00aa00389b71
bits per pixel: 12
number of planes: 3
type: YUV (planar)
id: 0x30323449 (I420)
guid: 49343230-0000-0010-8000-00aa00389b71
bits per pixel: 12
number of planes: 3
type: YUV (planar)
>
> Note the option lines with AGPMode and EnablePageFlip. I have X.org 6.8
> which didn't support AGPMode faster than 4. If you have a more recent
> X.org you might want to try to set AGPMode to 8 instead.
>
> regards Henrik
> --
> The address in the header is only to prevent spam. My real address is:
> hc8(at)uthyres.com Examples of addresses which go to spammers:
> root@variousus.net root@localhost
regards, Pawel
| |
| Henrik Carlqvist 2006-07-20, 7:25 am |
| Pawel <null7@wp.pl> wrote:
> http://rpm.greysector.net/mplayer that has ready to use mplayer rpm binaries.
> They are built for i386.
If so it might be worth a try to rebuild from source to get better
performance. On the other hand, I suppose that you did use the same
mplayer for i386 with the 815 graphics?
How much slower is the 9200 and how did you measure this? Did you check
how much CPU is consumed with top while mplayer is running?
>
> I did as You suggested, both with 4 and 8 mode option. No change
> unfortunatelly.
Strange, that should speed graphics up, not only 3D. Did you restart X
after editing xorg.conf?
> the result of xvinfo is the following:
Your results looks similar to what I get from xvinfo.
regards Henrik
--
The address in the header is only to prevent spam. My real address is:
hc8(at)uthyres.com Examples of addresses which go to spammers:
root@variousus.net root@localhost
| |
|
|
Hallo Henrik
Henrik Carlqvist <Henrik.Carlqvist@deadspam.com> writes:
> Pawel <null7@wp.pl> wrote:
>
> If so it might be worth a try to rebuild from source to get better
> performance. On the other hand, I suppose that you did use the same
> mplayer for i386 with the 815 graphics?
I use the same binary (i386) on both the cards. You are right, R9200 should
be faster anyway
>
> How much slower is the 9200 and how did you measure this? Did you check
> how much CPU is consumed with top while mplayer is running?
It is empirical impression, no technical comparisons.
The movies I am checking the performance on are 50fps (8mbps stream).
On another machine (P IV, 2,4 GHZ), they are displayed very fluent
(because of 50fps), while on my home machine (P III 1Ghz, 256 RAM) an built
in intel 815, the motion is not so fluent. When using R9200, motion is even
worse. Empirically, I would say 10 % worse, than built in card. Performance
of HD can be omitted, because when playing back movie 2 times in the row,
whole the second play goes from cache, with no HD interaction.
>
>
> Strange, that should speed graphics up, not only 3D. Did you restart X
> after editing xorg.conf?
I logged out, and logged in. X server is restarted when logging out
(I hope, I am right).
>
>
> Your results looks similar to what I get from xvinfo.
>
> regards Henrik
> --
> The address in the header is only to prevent spam. My real address is:
> hc8(at)uthyres.com Examples of addresses which go to spammers:
> root@variousus.net root@localhost
regards, Pawel
| |
| Henrik Carlqvist 2006-07-20, 1:14 pm |
| Pawel <null7@wp.pl> wrote:
>
> It is empirical impression, no technical comparisons.
With top you could see how much CPU mplayer is consuming. If it consumes
100% cpu it needs to be speeded up. Also, mplayer usually warns if your
system is to slow to play an avi file. If you want to se what that warning
looks like you can probably try -vo x11.
> The movies I am checking the performance on are 50fps (8mbps stream).
> On another machine (P IV, 2,4 GHZ), they are displayed very fluent
> (because of 50fps), while on my home machine (P III 1Ghz, 256 RAM) an
> built in intel 815, the motion is not so fluent. When using R9200,
> motion is even worse. Empirically, I would say 10 % worse, than built in
> card. Performance of HD can be omitted, because when playing back movie
> 2 times in the row, whole the second play goes from cache, with no HD
> interaction.
If for some reason the xv driver isn't fast enough you could also try the
-vo gl driver as you have DRI working.
>
> I logged out, and logged in. X server is restarted when logging out (I
> hope, I am right).
I'm not so sure that X is really restarted when logging out. You can
verify which AGP mode you are using by looking at the log file from X:
fgrep AGP /var/log/Xorg.0.log
Among other lines I see the following:
(--) Chipset ATI Radeon 9200 5961 (AGP) found
(**) RADEON(0): Option "AGPMode" "4"
(--) RADEON(0): Chipset: "ATI Radeon 9200 5961 (AGP)" (ChipID = 0x5961)
(II) RADEON(0): AGP card detected
(**) RADEON(0): Using AGP 4x mode
(II) RADEON(0): AGP Fast Write disabled by default
(II) RADEON(0): [agp] Mode 0x1f004a0f [AGP 0x8086/0x2578; Card 0x1002/0x5961]
Above I can verify that it has read my option AGPMode 4 and that it also
really is using 4x AGP. However, as your motherboard is an old PIII
motherboard it is probably not capable of AGP 8 and maybe not even AGP 4.
You might have to try AGP 2 instead. Try different settings to xorg.conf
and check how they worked by studying the log file. On my Slackware system
the log file is called /var/log/Xorg.0.log, on your FC system it might be
called something else.
If X isn't restarted only after a logout (and I don't think it is) you can
probably force a restart by pressing ctrl-alt-backspace.
regards Henrik
--
The address in the header is only to prevent spam. My real address is:
hc8(at)uthyres.com Examples of addresses which go to spammers:
root@variousus.net root@localhost
| |
| Henrik Carlqvist 2006-07-21, 1:14 pm |
| Pawel <null7@wp.pl> wrote:
> You are right, R9200 should be faster anyway
I hope that you will be able to get your R9200 up to speed. Today I will
go on vacation and will not be able to post messages for some time.
However, I will continue to follow this thread when I get back.
regards Henrik
--
The address in the header is only to prevent spam. My real address is:
hc8(at)uthyres.com Examples of addresses which go to spammers:
root@variousus.net root@localhost
| |
|
|
Hallo Henrik
As You suggested I went through xorg options. Except those You mentioned,
I found 2 more:
Option "AGPFastWrite" "boolean"
Enable AGP fast write.
(used only when DRI is enabled)
The default is off.
Option "DisplayPriority" "string"
Used to prevent flickering or tearing problem caused
by display buffer underflow.
AUTO -- Driver calculated (default).
BIOS -- Remain unchanged from BIOS setting.
Use this if the calculation is not correct
for your card.
HIGH -- Force to the highest priority.
Use this if you have problem with above options.
This may affect performance slightly.
The default value is AUTO.
"DisplayPriority" seems to be especially interesting.
My final xorg.conf radeon section looks like following:
Section "Device"
Identifier "Videocard0"
Driver "radeon"
VendorName "Videocard vendor"
Option "AGPMode" "8"
Option "EnablePageFlip" "on"
Option "AGPFastWrite" "On"
Option "DisplayPriority" "HIGH"
BoardName "ATI Technologies Inc RV280 [Radeon 9200 SE]"
EndSection
pieces of Xorg.0.log file looks like following:
(**) RADEON(0): AGP 8x mode is configured
(**) RADEON(0): Enabling AGP Fast Write
(II) RADEON(0): Render acceleration enabled
(II) RADEON(0): Direct rendering enabled
(II) RADEON(0): Using XFree86 Acceleration Architecture (XAA)
(II) RADEON(0): Page flipping enabled
Unfortunatelly animation of movie is not better. It looks in the same way,
it is teared.
top consumption is:
4300 root 16 0 34984 15m 8856 R 56.0 6.2 0:02.08 gmplayer
2347 root 16 0 152m 7888 4664 R 22.9 3.1 0:04.63 Xorg
I will also reencode my movies to 4Mbps and see how will it go.
regards, Pawel
| |
|
| Have a nice vacation,
and thank You for Your help.
I am also going for vacation since tommorow . I am back on aug 6.
I will inform You about progress in improoving the perfromance.
regards, Pawel
| |
| Henrik Carlqvist 2006-07-25, 7:16 pm |
| Pawel <null7@wp.pl> wrote:
> As You suggested I went through xorg options. Except those You mentioned,
> I found 2 more:
> Option "AGPFastWrite" "boolean"
> Enable AGP fast write.
> (used only when DRI is enabled)
> The default is off.
I have tried that option on my system, but found that it caused my system
to become unstable. Maybe it works better for you.
> Option "DisplayPriority" "string"
> Used to prevent flickering or tearing problem caused by
> display buffer underflow.
> AUTO -- Driver calculated (default). BIOS -- Remain
> unchanged from BIOS setting.
> Use this if the calculation is not correct for
> your card.
> HIGH -- Force to the highest priority.
> Use this if you have problem with above options.
> This may affect performance slightly.
> The default value is AUTO.
I haven't tried that option myself, at least I can't remember that I have
tried it.
> "DisplayPriority" seems to be especially interesting.
>
> My final xorg.conf radeon section looks like following: Section "Device"
> Identifier "Videocard0"
> Driver "radeon"
> VendorName "Videocard vendor"
> Option "AGPMode" "8"
I searched the net for specifications of intel 815 chipset. It seems as if
the 815 chipset is capable of AGP 4x, however it is not capable of AGP 8x.
> BoardName "ATI Technologies Inc RV280 [Radeon 9200 SE]"
Ouch, if you have a 9200SE it might explain some of your performance
problems. The 9200SE has 166 MHz memory on a 64-bit bus. The 9200 has 200
MHz memory on a 128-bit bus. This means that the memory bandwidth of the
9200SE is only 2.6 GB/s compared to 6.4 GB/s on a 9200.
> pieces of Xorg.0.log file looks like following:
> (**) RADEON(0): AGP 8x mode is configured
> (**) RADEON(0): Enabling AGP Fast Write
> (II) RADEON(0): Render acceleration enabled
> (II) RADEON(0): Direct rendering enabled
> (II) RADEON(0): Using XFree86 Acceleration Architecture (XAA)
> (II) RADEON(0): Page flipping enabled
You did configure AGP 8x, but your motherboard is probably only capable of
AGP 4x, it would be interesting to see your line from the log-file that
reports which AGP speed you really got, my line looks like this:
(**) RADEON(0): Using AGP 4x mode
> My final xorg.conf radeon section looks like following:
> Section "Device"
> Identifier "Videocard0"
> Driver "radeon"
> VendorName "Videocard vendor"
> Option "AGPMode" "8"
> Option "EnablePageFlip" "on"
> Option "AGPFastWrite" "On"
If you get any problems with X stability ( I don't remember if it was only
X, only OpenGL or if it was the entire computer that locked up ) it might
be because of the AGPFastWrite line.
> Unfortunatelly animation of movie is not better. It looks in the same
> way, it is teared.
Even though you have a 9200 SE the AGPMode and EnablePageFlip should
increas performance. My guess now is that AGPMode hasn't given any effect
because it was specified to high (8 instead of 4).
> top consumption is:
> 4300 root 16 0 34984 15m 8856 R 56.0 6.2 0:02.08 gmplayer
> 2347 root 16 0 152m 7888 4664 R 22.9 3.1 0:04.63 Xorg
Please avoid running programs as root unless absolutely necessary. A small
mistake as root user can cause big problems. With a normal user only that
user is affected by most small mistakes.
> I will also reencode my movies to 4Mbps and see how will it go.
If it would be to any help and if you have the possibility to put a test
video file where I can download it I could test it on my system. However,
it would not be fair to compare to your system as I have a P4 2.6 GHz with
a 9200 card. The interesting thing here is to see how the 9200SE compares
to the 815 built in graphics.
At http://users.erols.com/chare/video.htm many graphics chipsets are
described.
Maybe the 815 has very good memory bandwidth for graphics, at
http://www.xbitlabs.com/articles/ch...y/intel810.html it is
explained why the built in 810 graphics have better memory bandwidth than
the AGP port. If the 810 chipset with PC100 is capable of 800 MB/s, then
maybe 815 with PC133 is capable of a little bit more than 1GB/s. This is
about four times as fast as AGP 1x, twice as fast as AGP 2x and about as
fast as AGP 4x. If so, if you get AGP 4x working maybe you could expect
about the same memory bandwidth as with your built in 815 graphics, but
not much more memory bandwidth. The graphics memory bandwidth is probably
the most important part for xvideo performance.
Even though the memory bandiwidth isn't higher at least the 9200SE should
have better OpenGL performance. Than performance could be measured with a
simple benchmark like glxgears or a more serious benchmark like the ones
built in into games like Quake or Unreal Tournament.
regards Henrik
--
The address in the header is only to prevent spam. My real address is:
hc8(at)uthyres.com Examples of addresses which go to spammers:
root@variousus.net root@localhost
| |
|
|
Hallo Henrik
Henrik Carlqvist <Henrik.Carlqvist@deadspam.com> writes:
> Pawel <null7@wp.pl> wrote:
>
> I have tried that option on my system, but found that it caused my system
> to become unstable. Maybe it works better for you.
>
>
> I haven't tried that option myself, at least I can't remember that I have
> tried it.
>
>
> I searched the net for specifications of intel 815 chipset. It seems as if
> the 815 chipset is capable of AGP 4x, however it is not capable of AGP 8x.
>
>
> Ouch, if you have a 9200SE it might explain some of your performance
> problems. The 9200SE has 166 MHz memory on a 64-bit bus. The 9200 has 200
> MHz memory on a 128-bit bus. This means that the memory bandwidth of the
> 9200SE is only 2.6 GB/s compared to 6.4 GB/s on a 9200.
>
>
> You did configure AGP 8x, but your motherboard is probably only capable of
> AGP 4x, it would be interesting to see your line from the log-file that
> reports which AGP speed you really got, my line looks like this:
>
> (**) RADEON(0): Using AGP 4x mode
>
>
> If you get any problems with X stability ( I don't remember if it was only
> X, only OpenGL or if it was the entire computer that locked up ) it might
> be because of the AGPFastWrite line.
>
>
> Even though you have a 9200 SE the AGPMode and EnablePageFlip should
> increas performance. My guess now is that AGPMode hasn't given any effect
> because it was specified to high (8 instead of 4).
>
>
> Please avoid running programs as root unless absolutely necessary. A small
> mistake as root user can cause big problems. With a normal user only that
> user is affected by most small mistakes.
>
>
> If it would be to any help and if you have the possibility to put a test
> video file where I can download it I could test it on my system. However,
> it would not be fair to compare to your system as I have a P4 2.6 GHz with
> a 9200 card. The interesting thing here is to see how the 9200SE compares
> to the 815 built in graphics.
>
> At http://users.erols.com/chare/video.htm many graphics chipsets are
> described.
>
> Maybe the 815 has very good memory bandwidth for graphics, at
> http://www.xbitlabs.com/articles/ch...y/intel810.html it is
> explained why the built in 810 graphics have better memory bandwidth than
> the AGP port. If the 810 chipset with PC100 is capable of 800 MB/s, then
> maybe 815 with PC133 is capable of a little bit more than 1GB/s. This is
> about four times as fast as AGP 1x, twice as fast as AGP 2x and about as
> fast as AGP 4x. If so, if you get AGP 4x working maybe you could expect
> about the same memory bandwidth as with your built in 815 graphics, but
> not much more memory bandwidth. The graphics memory bandwidth is probably
> the most important part for xvideo performance.
>
> Even though the memory bandiwidth isn't higher at least the 9200SE should
> have better OpenGL performance. Than performance could be measured with a
> simple benchmark like glxgears or a more serious benchmark like the ones
> built in into games like Quake or Unreal Tournament.
>
> regards Henrik
> --
> The address in the header is only to prevent spam. My real address is:
> hc8(at)uthyres.com Examples of addresses which go to spammers:
> root@variousus.net root@localhost
I just recompiled mplayer to i686 architecture, chnaged to AGP 4x,
but unfortunatelly animation is still teared.
I decreased data stream to 1Mbps (from 8Mbps).
Quality is poor but I still can see that animation is teared(?!strange).
I will try to play with directfb video filter.
Regards,
Pawel
| |
| Henrik Carlqvist 2006-08-09, 7:16 pm |
| Pawel <null7@wp.pl> wrote:
> I just recompiled mplayer to i686 architecture, chnaged to AGP 4x,
Where you able to verify that it really got 4x? Did you find something
like this in Xorg.0.log:
(**) RADEON(0): Using AGP 4x mode
?
At http://www.free3d.org/ there are results from a simple benchmark:
cat /proc/pci | grep VGA || lspci | grep VGA | \
colrm 1 4 ; cat /proc/cpuinfo | \ egrep "model name|MHz" ; xdpyinfo | \
egrep "version:|dimensions|depth of" ; glxinfo | \
egrep -A2 "direct rendering|OpenGL vendor" ; glxgears & sleep 25 ; \
killall glxgears
There are two systems with 9250 cards which both get about 2600 fps. You
can compare your result with these numbers, but the glxgears benchmark is
not a very good benchmark and it doesn't say anything at all about xvideo
performance.
> but unfortunatelly animation is still teared.
> I decreased data stream to 1Mbps (from 8Mbps).
> Quality is poor but I still can see that animation is teared(?!strange).
> I will try to play with directfb video filter.
MPlayer usually warns if it can't get high enough RTC interrupt frequency.
Have you done:
echo 1024 > /proc/sys/dev/rtc/max-user-freq
?
regards Henrik
--
The address in the header is only to prevent spam. My real address is:
hc8(at)uthyres.com Examples of addresses which go to spammers:
root@variousus.net root@localhost
| |
|
|
Hallo Henrik
Henrik Carlqvist <Henrik.Carlqvist@deadspam.com> writes:
> Pawel <null7@wp.pl> wrote:
>
> Where you able to verify that it really got 4x? Did you find something
> like this in Xorg.0.log:
>
> (**) RADEON(0): Using AGP 4x mode
Yes, I found similar info in /var/log/messages.
>
> ?
>
> At http://www.free3d.org/ there are results from a simple benchmark:
>
> cat /proc/pci | grep VGA || lspci | grep VGA | \
> colrm 1 4 ; cat /proc/cpuinfo | \ egrep "model name|MHz" ; xdpyinfo | \
> egrep "version:|dimensions|depth of" ; glxinfo | \
> egrep -A2 "direct rendering|OpenGL vendor" ; glxgears & sleep 25 ; \
> killall glxgears
I copied it and will use it. By the way I will compare my built in intel
with radeon.
>
> There are two systems with 9250 cards which both get about 2600 fps. You
> can compare your result with these numbers, but the glxgears benchmark is
> not a very good benchmark and it doesn't say anything at all about xvideo
> performance.
>
>
> MPlayer usually warns if it can't get high enough RTC interrupt frequency.
> Have you done:
>
> echo 1024 > /proc/sys/dev/rtc/max-user-freq
>
> ?
Regards
Pawel
| |
|
| Hallo Henrik
Henrik Carlqvist <Henrik.Carlqvist@deadspam.com> writes:
> Pawel <null7@wp.pl> wrote:
>
> Where you able to verify that it really got 4x? Did you find something
> like this in Xorg.0.log:
>
> (**) RADEON(0): Using AGP 4x mode
glxperf.sh:
OpenGL renderer string: Mesa DRI R200 20041207 AGP 4x x86/MMX/SSE TCL
It can be obtained from /var/log/messages:
Aug 12 15:51:00 localhost kernel: agpgart: Found an AGP 2.0 compliant device at 0000:00:00.0.
Aug 12 15:51:00 localhost kernel: agpgart: Putting AGP V2 device at 0000:00:00.0 into 4x mode
Aug 12 15:51:00 localhost kernel: agpgart: Putting AGP V2 device at 0000:01:00.0 into 4x mode
I noticed is AGP 8x configuration information inconsistency
Note: AGP mode info is different in glxperf.sh script and in /var/log/messages.
Current config:
Option "AGPMode" "8"
glxperf.sh:
OpenGL renderer string: Mesa DRI R200 20041207 AGP 8x x86/MMX/SSE TCL
/var/log/messages:
ug 12 20:03:57 localhost kernel: agpgart: Found an AGP 2.0 compliant device at 0000:00:00.0.
Aug 12 20:03:57 localhost kernel: agpgart: Putting AGP V2 device at 0000:00:00.0 into 1x mode
Aug 12 20:03:57 localhost kernel: agpgart: Putting AGP V2 device at 0000:01:00.0 into 1x mode
For AGP 4x info is consistent. Conclusion: 4x is the highest available for my hardware.
>
> ?
>
> At http://www.free3d.org/ there are results from a simple benchmark:
>
> cat /proc/pci | grep VGA || lspci | grep VGA | \
> colrm 1 4 ; cat /proc/cpuinfo | \ egrep "model name|MHz" ; xdpyinfo | \
> egrep "version:|dimensions|depth of" ; glxinfo | \
> egrep -A2 "direct rendering|OpenGL vendor" ; glxgears & sleep 25 ; \
> killall glxgears
Here are my results:
bash-3.1$ ~/Scripts/glxperf.sh
0.0 VGA compatible controller: ATI Technologies Inc RV280 [Radeon 9200 SE] (rev 01)
model name : Pentium III (Coppermine)
cpu MHz : 994.404
X.Org version: 7.0.0
dimensions: 1024x768 pixels (302x232 millimeters)
depth of root window: 24 planes
direct rendering: Yes
server glx vendor string: SGI
server glx version string: 1.2
--
OpenGL vendor string: Tungsten Graphics, Inc.
OpenGL renderer string: Mesa DRI R200 20041207 AGP 4x x86/MMX/SSE TCL
OpenGL version string: 1.3 Mesa 6.4.2
4993 frames in 5.0 seconds = 998.429 FPS
bash-3.1$ ~/Scripts/glxperf.sh
2.0 VGA compatible controller: Intel Corporation 82815 CGC [Chipset Graphics Controller] (rev 04)
model name : Pentium III (Coppermine)
cpu MHz : 994.209
X.Org version: 7.0.0
dimensions: 1024x768 pixels (302x232 millimeters)
depth of root window: 24 planes
direct rendering: No
server glx vendor string: SGI
server glx version string: 1.2
--
OpenGL vendor string: Mesa project: www.mesa3d.org
OpenGL renderer string: Mesa GLX Indirect
OpenGL version string: 1.2 (1.5 Mesa 6.4.2)
570 frames in 5.1 seconds = 111.648 FPS
>
> There are two systems with 9250 cards which both get about 2600 fps. You
> can compare your result with these numbers, but the glxgears benchmark is
> not a very good benchmark and it doesn't say anything at all about xvideo
> performance.
>
>
> MPlayer usually warns if it can't get high enough RTC interrupt frequency.
> Have you done:
>
> echo 1024 > /proc/sys/dev/rtc/max-user-freq
Yes I have
>
> ?
I saw another interesting option in man mplayer related to directfb video driver:
buffermode=single|double|triple
Double and triple buffering give best results if you
want to avoid tearing issues. Triple buffering is more
efficient than double buffering as it does not block
MPlayer while waiting for the vertical retrace. Single
buffering should be avoided (default: single).
I will try to use it
>
> regards Henrik
> --
> The address in the header is only to prevent spam. My real address is:
> hc8(at)uthyres.com Examples of addresses which go to spammers:
> root@variousus.net root@localhost
regards,
Pawel
| |
| Henrik Carlqvist 2006-08-14, 7:15 pm |
| Pawel <null7@wp.pl> wrote:
> glxperf.sh:
> OpenGL renderer string: Mesa DRI R200 20041207 AGP 4x x86/MMX/SSE TCL
That looks a lot like my result:
OpenGL renderer string: Mesa DRI R200 20030328 AGP 4x x86/MMX/SSE2 TCL
> Here are my results:
> bash-3.1$ ~/Scripts/glxperf.sh
> 0.0 VGA compatible controller: ATI Technologies Inc RV280 [Radeon 9200
> SE] (rev 01) model name : Pentium III (Coppermine) cpu MHz :
> 994.404
> X.Org version: 7.0.0
> dimensions: 1024x768 pixels (302x232 millimeters) depth of root
> window: 24 planes
> direct rendering: Yes
> server glx vendor string: SGI
> server glx version string: 1.2
> --
> OpenGL vendor string: Tungsten Graphics, Inc. OpenGL renderer string:
> Mesa DRI R200 20041207 AGP 4x x86/MMX/SSE TCL OpenGL version string: 1.3
> Mesa 6.4.2
>
> 4993 frames in 5.0 seconds = 998.429 FPS
That is less than half my fps. I don't know for sure if the big difference
is because of our difference in CPU (I have a 2.6 GHz P4) or because you
have a 9200SE and I have a 9200.
> bash-3.1$ ~/Scripts/glxperf.sh
> 2.0 VGA compatible controller: Intel Corporation 82815 CGC [Chipset
> Graphics Controller] (rev 04) model name : Pentium III (Coppermine)
> cpu MHz : 994.209
> X.Org version: 7.0.0
> dimensions: 1024x768 pixels (302x232 millimeters) depth of root
> window: 24 planes
> direct rendering: No
> server glx vendor string: SGI
> server glx version string: 1.2
> --
> OpenGL vendor string: Mesa project: www.mesa3d.org OpenGL renderer
> string: Mesa GLX Indirect OpenGL version string: 1.2 (1.5 Mesa 6.4.2)
> 570 frames in 5.1 seconds = 111.648 FPS
The above would indicate that your 9200SE is about 9 times as fast as the
built in graphics for OpenGL. However, this comparision is not fair as the
test with the built in graphics was run without DRI enabled.
>
> Yes I have
> I saw another interesting option in man mplayer related to directfb
> video driver:
>
> buffermode=single|double|triple
> Double and triple buffering give best results
> if you want to avoid tearing issues. Triple
> buffering is more efficient than double
> buffering as it does not block MPlayer while
> waiting for the vertical retrace. Single
> buffering should be avoided (default: single).
>
> I will try to use it
I wish you luck with this setting, I'm out of ideas now.
regards Henrik
--
The address in the header is only to prevent spam. My real address is:
hc8(at)uthyres.com Examples of addresses which go to spammers:
root@variousus.net root@localhost
| |
|
|
Hallo Henrik
Henrik Carlqvist <Henrik.Carlqvist@deadspam.com> writes:
[vbcol=seagreen]
> OpenGL renderer string: Mesa DRI R200 20030328 AGP 4x x86/MMX/SSE2 TCL
>
>
> That is less than half my fps. I don't know for sure if the big difference
> is because of our difference in CPU (I have a 2.6 GHz P4) or because you
> have a 9200SE and I have a 9200.
>
>
> The above would indicate that your 9200SE is about 9 times as fast as the
> built in graphics for OpenGL. However, this comparision is not fair as the
> test with the built in graphics was run without DRI enabled.
>
>
Playing with framebuffer did final the job. Additionally I had to increase
video stream from 8mbps to 15mbps. I am guessing that it caused smaller
overhead on decompression of the movie.
Now my movies go very fluent.
Henrik, thank You once more for Your great help!
Best regards,
Pawel
| |
| Henrik Carlqvist 2006-08-23, 1:20 am |
| Pawel <null7@wp.pl> wrote:
>
> Playing with framebuffer did final the job.
Nice to hear that you finally solved the problem. By "framebuffer", do you
mean the buffermode setting above or did you use something like "-vo fbdev"?
regards Henrik
--
The address in the header is only to prevent spam. My real address is:
hc8(at)uthyres.com Examples of addresses which go to spammers:
root@variousus.net root@localhost
| |
|
|
Hello Henrik.
Henrik Carlqvist <Henrik.Carlqvist@deadspam.com> writes:
[vbcol=seagreen]
I mean "-vo directfb"
Best regards,
Pawel
|
|
|
|
|