on os x (10.39), and using pd 0.39.2 extended test 7, i'm getting this weird behaviour with the gui redraw.
using a 20x20 pixel canvas as a cursor, and sending a [pos xy< message to change the cursor's position, the gui redraw is really blocky.
BUT, if i hold down a key on the keyboard (any key seems to work), the gui redraws suddenly work really smoothly, and the 20x20 pixel canvas glides across the screen beautifully.
the same thing happens with other gui objects (number boxes etc) too.
attached is a patch that demonstates this.
basically, the nice smooth action when i hold down a key is what i am after, but i don't want to have to hold down a key to do that.
actually, just playing round with the audio settings is having an effect here. even though compute audio is turned off.
it seems that the rate of gui redraw is connected to the audio latency, and somehow you can override that by holding down a key on the keyboard.
hmm.. kind of annoying this...why can't pd do simple graphics stuff that was possible on commodore 64?
On 28/01/2007, at 10.32, hard off wrote:
using a 20x20 pixel canvas as a cursor, and sending a [pos xy< message to change the cursor's position, the gui redraw is really blocky.
Pardon my lack of knowledge, but what does "blocky" mean? Here (pd-
extended-test7 mac intel 10.4.8) the square canvas scrolls
horizontally what seam like smooth 1 pixel per time unite.
On Jan 28, 2007, at 11:05 AM, Steffen wrote:
On 28/01/2007, at 10.32, hard off wrote:
using a 20x20 pixel canvas as a cursor, and sending a [pos xy<
message to change the cursor's position, the gui redraw is really blocky.Pardon my lack of knowledge, but what does "blocky" mean? Here (pd- extended-test7 mac intel 10.4.8) the square canvas scrolls
horizontally what seam like smooth 1 pixel per time unite.
I am on Pd-0.39.2-extended-test7 on Mac/PowerPC and it seems smooth
to me too. I can see know difference whether I am holding down a key
or not, or wether the DSP is on or not.
One thing, you are updating every 5 ms. Your screen probably has a
refresh rate of around 12.5ms (80Hz) or less, so that's 2.5 position
updates per draw update. That could make things less that smooth, I
suppose.
.hc
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/ listinfo/pd-list
Using ReBirth is like trying to play an 808 with a long stick. - David Zicarelli
'blocky' = the the sqaure canvas is jumping about 50 pixels at a time, stopping for 100ms or so, and then juming again. if i hold down a key, this blocky motion stops, and the square canvas slides 1 pixel at a time.
i can't see why the screen refresh rate would have anything to do with this problem.
there is definately a problem here. i am not hallucinating!
I am not saying it doesn't exist, I am just putting ideas out there.
I can't reproduce it though. Can you make a video of it somehow?
.hc
On Jan 28, 2007, at 5:41 PM, hard off wrote:
'blocky' = the the sqaure canvas is jumping about 50 pixels at a time, stopping for 100ms or so, and then juming again. if i hold down a key, this blocky motion stops, and the square canvas slides 1 pixel at a time.
i can't see why the screen refresh rate would have anything to do with this problem.
there is definately a problem here. i am not hallucinating!
Access to computers should be unlimited and total. - the hacker ethic
Yes, try to capture a short video screen capture!
There are a number of free tools, but I can't remember their names right now.
~Kyle
On 1/29/07, Hans-Christoph Steiner hans@eds.org wrote:
I am not saying it doesn't exist, I am just putting ideas out there. I can't reproduce it though. Can you make a video of it somehow?
.hc
On Jan 28, 2007, at 5:41 PM, hard off wrote:
'blocky' = the the sqaure canvas is jumping about 50 pixels at a time, stopping for 100ms or so, and then juming again. if i hold down a key, this blocky motion stops, and the square canvas slides 1 pixel at a time.
i can't see why the screen refresh rate would have anything to do with this problem.
there is definately a problem here. i am not hallucinating!
Access to computers should be unlimited and total. - the hacker ethic
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
On Mon Jan 29, 2007 at 02:55:12PM -0600, Kyle Klipowicz wrote:
Yes, try to capture a short video screen capture!
be sure to capture at about 100 fps, so it doesnt introduce much jitter / quantization error of its own, so we can tell whats _really_ going on in your patch
There are a number of free tools, but I can't remember their names right now.
~Kyle
On 1/29/07, Hans-Christoph Steiner hans@eds.org wrote:
I am not saying it doesn't exist, I am just putting ideas out there. I can't reproduce it though. Can you make a video of it somehow?
.hc
On Jan 28, 2007, at 5:41 PM, hard off wrote:
'blocky' = the the sqaure canvas is jumping about 50 pixels at a time, stopping for 100ms or so, and then juming again. if i hold down a key, this blocky motion stops, and the square canvas slides 1 pixel at a time.
i can't see why the screen refresh rate would have anything to do with this problem.
there is definately a problem here. i am not hallucinating!
Access to computers should be unlimited and total. - the hacker ethic
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
--
http://theradioproject.com http://perhapsidid.blogspot.com
(((())))(()()((((((((()())))()(((((((())()()())()))) (())))))(()))))))))))))(((((((((((()()))))))))((()))) ))(((((((((((())))())))))))))))))))__________ _____())))))(((((((((((((()))))))))))_______ ((((((())))))))))))((((((((000)))oOOOOOO
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
on GUI redrawing:
I wonder if its possible to optimize the underlaying TK TCL without much effort like rewriting parts of pd. I mean just by tweaking some config files. Is there a TK TCL Shell out which uses OpenGL or other hardware acceleration?
I encounter few audio dropouts while tweaking polygons of datastructs, apart from it the GUI runs well but I would like to use more of the structs abd their graphical representation.
Cheers,
Malte
Hallo, Malte Steiner hat gesagt: // Malte Steiner wrote:
I wonder if its possible to optimize the underlaying TK TCL without much effort like rewriting parts of pd. I mean just by tweaking some config files. Is there a TK TCL Shell out which uses OpenGL or other hardware acceleration?
Usually all 2D-operations already are hardware accelerated on any card that is newer than, say, seven or eight years. ;)
I encounter few audio dropouts while tweaking polygons of datastructs, apart from it the GUI runs well but I would like to use more of the structs abd their graphical representation.
The problem is that there's too much communication going on between the main Pd process and the GUI at times. Matju is trying a fix with DesireData, a temporaty solution is inside of Pd since I think 0.37 where Miller introduced throttling for some GUI objects.
But the simple fix still isn't enough as everyone can easily see by moving around two or three active [vu] objects. Or by playing with lots of data structures visible. (Hidden data structures are fine, btw.)
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__
Usually all 2D-operations already are hardware accelerated on any card that is newer than, say, seven or eight years. ;)
Not to the full extent when you use a recent card. The manufacturers spend more time to improve the 3D performance than the 2D so meanwhile even for 2D operations OpenGL is preferable, when you have correct running drivers.
Cheers,
Malte
hallo pidier
i'm using Core 2 Duo mac book , and i don't know if the flowing
problem is a os x (10.39 pd 0.39.2 extended test 7 or not:
it doesn't load to array properly, when a sound file is longer than 2
Min.
Message | read -resize -maxsize le+08 |
error: soundfiler_read: truncated to 4000000 elements warning: array array1: clearing save-in-patch flag error: usage: read [flags] filename tablename... flags: -skip <n> -nframes <n> -resize -maxsize <n> ... -raw <headerbytes> <channels> <bytespersamp> <endian (b, l, or n)>.
it works with a old extended version on G4... ;-( is there any tip for me?
(\ /) [- . - 猫earweego_sonicvision www.earweego.net
Hi Echo!
Tip: Don't use files longer than 2 mins :)
Best way is to use [sfread~] instead.
Andy
On Wed, 31 Jan 2007 00:37:54 +0100 Echo Ho pidi@echoho.at wrote:
hallo pidier i'm using Core 2 Duo mac book , and i don't know if the flowing
problem is a os x (10.39 pd 0.39.2 extended test 7 or not:it doesn't load to array properly, when a sound file is longer than 2
Min. Message | read -resize -maxsize le+08 |error: soundfiler_read: truncated to 4000000 elements warning: array array1: clearing save-in-patch flag error: usage: read [flags] filename tablename... flags: -skip <n> -nframes <n> -resize -maxsize <n> ... -raw <headerbytes> <channels> <bytespersamp> <endian (b, l, or n)>.
it works with a old extended version on G4... ;-( is there any tip for me?
(\ /) [- . - ___earweego_sonicvision www.earweego.net
Hmm, sounds like a bug in the Intel build, this should work. File a
bug in the tracker:
http://puredata.org/dev/bugtracker
.hc
On Jan 30, 2007, at 6:37 PM, Echo Ho wrote:
hallo pidier i'm using Core 2 Duo mac book , and i don't know if the flowing
problem is a os x (10.39 pd 0.39.2 extended test 7 or not:it doesn't load to array properly, when a sound file is longer than
2 Min. Message | read -resize -maxsize le+08 |error: soundfiler_read: truncated to 4000000 elements warning: array array1: clearing save-in-patch flag error: usage: read [flags] filename tablename... flags: -skip <n> -nframes <n> -resize -maxsize <n> ... -raw <headerbytes> <channels> <bytespersamp> <endian (b, l, or n)>.
it works with a old extended version on G4... ;-( is there any tip for me?
(\ /) [- . - 猫earweego_sonicvision www.earweego.net
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/ listinfo/pd-list
If you are not part of the solution, you are part of the problem.
-
carmen -
Echo Ho -
Frank Barknecht -
Hans-Christoph Steiner -
hard off -
Kyle Klipowicz -
Malte Steiner -
padawan12 -
Steffen