User talk:Dora Gustafson - Revision history

Dora Gustafson: /* llSetMemoryLimit */

2012-02-29T14:07:00Z

llSetMemoryLimit

← Older revision		Revision as of 07:07, 29 February 2012
Line 56:		Line 56:
	== llSetMemoryLimit ==		== llSetMemoryLimit ==

	Thanks for adding the caveat. Small question: if you previously set it to say 63KiB, and then set it to 2KiB and that failed, it would reset it to 64KiB? If that is the case, I'll add it as a sub-caveat. Thanks for the detailed response to my previous query, it's excellent. -- '''[[User:Strife_Onizuka\|Strife]]''' <sup><small>([[User talk:Strife_Onizuka\|talk]]\|[[Special:Contributions/Strife_Onizuka\|contribs]])</small></sup> 21:59, 28 February 2012 (PST)		Thanks for adding the caveat. Small question: if you previously set it to say 63KiB, and then set it to 2KiB and that failed, it would reset it to 64KiB? If that is the case, I'll add it as a sub-caveat. Thanks for the detailed response to my previous query, it's excellent. -- '''[[User:Strife_Onizuka\|Strife]]''' <sup><small>([[User talk:Strife_Onizuka\|talk]]\|[[Special:Contributions/Strife_Onizuka\|contribs]])</small></sup> 21:59, 28 February 2012 (PST)<br>
			You are right in your assumption; The limit will not be reset to max, it will be untouched. I made this test script:
			<lsl>
			string ballast="use some memory";
			default
			{
			state_entry()
			{
			integer i=64;
			list heavy=[];
			while (i--) heavy+=[ballast];
			llOwnerSay("No limit set ever\nMemory used: "+(string)llGetUsedMemory()+" Free memory: "+(string)llGetFreeMemory());
			llSetMemoryLimit( llGetUsedMemory()/2);
			llOwnerSay("Memory limit set to "+(string)(llGetUsedMemory()/2)+"\nMemory used: "+(string)llGetUsedMemory()+" Free memory: "+(string)llGetFreeMemory());
			llSetMemoryLimit( 2*llGetUsedMemory());
			llOwnerSay("Memory limit set to "+(string)(2*llGetUsedMemory())+"\nMemory used: "+(string)llGetUsedMemory()+" Free memory: "+(string)llGetFreeMemory());
			llSetMemoryLimit( llGetUsedMemory()/2);
			llOwnerSay("Memory limit set to "+(string)(llGetUsedMemory()/2)+"\nMemory used: "+(string)llGetUsedMemory()+" Free memory: "+(string)llGetFreeMemory());
			}
			}</lsl>
			Which produce this output:
			No limit set ever
			Memory used: 5636 Free memory: 59900

			Memory limit set to 2928
			Memory used: 5856 Free memory: 59680

			Memory limit set to 11712
			Memory used: 5856 Free memory: 5856

			Memory limit set to 2928
			Memory used: 5856 Free memory: 5856
			I made a change in the Caveats.
			[[User:Dora Gustafson\|Dora Gustafson]] 06:07, 29 February 2012 (PST)

Strife Onizuka: /* llSetMemoryLimit */ new section

2012-02-29T05:59:50Z

llSetMemoryLimit: new section

← Older revision		Revision as of 22:59, 28 February 2012
Line 53:		Line 53:
	I have not tested other prim types and not linked sets, just a single BOX_TYPE prim		I have not tested other prim types and not linked sets, just a single BOX_TYPE prim
	--[[User:Dora Gustafson\|Dora Gustafson]] 13:05, 16 December 2010 (UTC)		--[[User:Dora Gustafson\|Dora Gustafson]] 13:05, 16 December 2010 (UTC)

			== llSetMemoryLimit ==

			Thanks for adding the caveat. Small question: if you previously set it to say 63KiB, and then set it to 2KiB and that failed, it would reset it to 64KiB? If that is the case, I'll add it as a sub-caveat. Thanks for the detailed response to my previous query, it's excellent. -- '''[[User:Strife_Onizuka\|Strife]]''' <sup><small>([[User talk:Strife_Onizuka\|talk]]\|[[Special:Contributions/Strife_Onizuka\|contribs]])</small></sup> 21:59, 28 February 2012 (PST)

Dora Gustafson: /* llGetCenterOfMass */

2010-12-17T11:44:05Z

llGetCenterOfMass

← Older revision		Revision as of 04:44, 17 December 2010
Line 29:		Line 29:
	First in newly created prim:		First in newly created prim:
	Box cut in 1/4, non-physic: 0.000000 // primposition returned for center of mass, NOT correct		Box cut in 1/4, non-physic: 0.000000 // primposition returned for center of mass, NOT correct
	Physic box cut in 1/4: 0.706897 // square root(2), correct		Physic box cut in 1/4: 0.706897 // square root(0.5), correct
	Physic box cut in half: 0.500004 // 0.5, correct		Physic box cut in half: 0.500004 // 0.5, correct
	Box cut in half, non-physic: 0.500000 // saved value, correct		Box cut in half, non-physic: 0.500000 // saved value, correct
Line 36:		Line 36:
	Second run:		Second run:
	Box cut in 1/4, non-physic: 0.500000 // saved value, NOT correct		Box cut in 1/4, non-physic: 0.500000 // saved value, NOT correct
	Physic box cut in 1/4: 0.706915 // square root(2), correct		Physic box cut in 1/4: 0.706915 // square root(0.5), correct
	Physic box cut in half: 0.499993 // 0.5, correct		Physic box cut in half: 0.499993 // 0.5, correct
	Box cut in half, non-physic: 0.500001 // saved value, correct		Box cut in half, non-physic: 0.500001 // saved value, correct

Dora Gustafson at 13:05, 16 December 2010

2010-12-16T13:05:53Z

← Older revision		Revision as of 06:05, 16 December 2010
Line 1:		Line 1:
	===llGetCenterOfMass===		===llGetCenterOfMass===
	You seem to have done some testing for [[llGetCenterOfMass]]. I'm curious, if you change the shape of the object in ways that should change the center of mass while it's non-physical, but after it has been physical (so that there is a cached value for the center) does it invalidate the cached value? So if you call llGetCenterOfMass does it then return the invalid value or the value of llGetPos? If you move the entire object does it return a position adjusted cached value for CoM? -- '''[[User:Strife_Onizuka\|Strife]]''' <sup><small>([[User talk:Strife_Onizuka\|talk]]\|[[Special:Contributions/Strife_Onizuka\|contribs]])</small></sup> 18:32, 3 December 2010 (UTC)		You seem to have done some testing for [[llGetCenterOfMass]]. I'm curious, if you change the shape of the object in ways that should change the center of mass while it's non-physical, but after it has been physical (so that there is a cached value for the center) does it invalidate the cached value? So if you call llGetCenterOfMass does it then return the invalid value or the value of llGetPos? If you move the entire object does it return a position adjusted cached value for CoM? -- '''[[User:Strife_Onizuka\|Strife]]''' <sup><small>([[User talk:Strife_Onizuka\|talk]]\|[[Special:Contributions/Strife_Onizuka\|contribs]])</small></sup> 18:32, 3 December 2010 (UTC)
			The answer to the first question is: no. The 'cached' value remains unchanged in non-physical mode.
			The answer to the second is that llGetPos() is only assumed in a newly created prim, one that was never made physical and never computed the center of mass in physical mode.
			When the non-physical virgin prim is moved around llGetCenterOfMass() returns the updated llGetPos() for the new positions.
			When the non-physical prim where llGetCenterOfMass() was computed at least once in physical mode, the cached value persist.
			'''Here is a script I used for testing:'''
			<lsl>default
			{
			touch_end(integer num)
			{
			llSetPrimitiveParams([PRIM_SIZE, <2.0, 2.0, 2.0>, PRIM_TYPE, 0, 0, <0.125, 0.375, 0.0>, 0.0, <0.0, 0.0, 0.0>, <1.0, 1.0, 0.0>, <0.0, 0.0, 0.0>]);
			llOwnerSay("Box cut in 1/4, non-physic: "+(string)llVecDist(llGetCenterOfMass(),llGetPos()));
			llSetStatus( STATUS_PHYSICS, TRUE);
			llSleep(0.5);
			llOwnerSay("Physic box cut in 1/4: "+(string)llVecDist(llGetCenterOfMass(),llGetPos()));
			llSetPrimitiveParams([PRIM_TYPE, 0, 0, <0.375, 0.875, 0.0>, 0.0, <0.0, 0.0, 0.0>, <1.0, 1.0, 0.0>, <0.0, 0.0, 0.0>]);
			llSleep(0.5);
			llOwnerSay("Physic box cut in half: "+(string)llVecDist(llGetCenterOfMass(),llGetPos()));
			llSetStatus( STATUS_PHYSICS, FALSE);
			llSleep(0.5);
			llOwnerSay("Box cut in half, non-physic: "+(string)llVecDist(llGetCenterOfMass(),llGetPos()));
			llSetPrimitiveParams([PRIM_TYPE, 0, 0, <0.0, 1.0, 0.0>, 0.0, <0.0, 0.0, 0.0>, <1.0, 1.0, 0.0>, <0.0, 0.0, 0.0>]);
			llOwnerSay("Full box, non-physic: "+(string)llVecDist(llGetCenterOfMass(),llGetPos()));
			llResetScript();
			}
			}</lsl>
			'''Output from above script:'''
			First in newly created prim:
			Box cut in 1/4, non-physic: 0.000000 // primposition returned for center of mass, NOT correct
			Physic box cut in 1/4: 0.706897 // square root(2), correct
			Physic box cut in half: 0.500004 // 0.5, correct
			Box cut in half, non-physic: 0.500000 // saved value, correct
			Full box, non-physic: 0.500000 // saved value, NOT correct

			Second run:
			Box cut in 1/4, non-physic: 0.500000 // saved value, NOT correct
			Physic box cut in 1/4: 0.706915 // square root(2), correct
			Physic box cut in half: 0.499993 // 0.5, correct
			Box cut in half, non-physic: 0.500001 // saved value, correct
			Full box, non-physic: 0.500001 // saved value, NOT correct

			Third run:
			Box cut in 1/4, non-physic: 0.500001
			Physic box cut in 1/4: 0.707092
			Physic box cut in half: 0.500009
			Box cut in half, non-physic: 0.500001
			Full box, non-physic: 0.500001

			...
			The results are unchanged in 2. 3. 4. ... run except for minor rounding differences

			I have not tested other prim types and not linked sets, just a single BOX_TYPE prim
			--[[User:Dora Gustafson\|Dora Gustafson]] 13:05, 16 December 2010 (UTC)

Strife Onizuka at 18:33, 3 December 2010

2010-12-03T18:33:12Z

← Older revision		Revision as of 11:33, 3 December 2010
Line 1:		Line 1:
			===llGetCenterOfMass===
	You seem to have done some testing for [[llGetCenterOfMass]]. I'm curious, if you change the shape of the object in ways that should change the center of mass while it's non-physical, but after it has been physical (so that there is a cached value for the center) does it invalidate the cached value? So if you call llGetCenterOfMass does it then return the invalid value or the value of llGetPos? If you move the entire object does it return a position adjusted cached value for CoM? -- '''[[User:Strife_Onizuka\|Strife]]''' <sup><small>([[User talk:Strife_Onizuka\|talk]]\|[[Special:Contributions/Strife_Onizuka\|contribs]])</small></sup> 18:32, 3 December 2010 (UTC)		You seem to have done some testing for [[llGetCenterOfMass]]. I'm curious, if you change the shape of the object in ways that should change the center of mass while it's non-physical, but after it has been physical (so that there is a cached value for the center) does it invalidate the cached value? So if you call llGetCenterOfMass does it then return the invalid value or the value of llGetPos? If you move the entire object does it return a position adjusted cached value for CoM? -- '''[[User:Strife_Onizuka\|Strife]]''' <sup><small>([[User talk:Strife_Onizuka\|talk]]\|[[Special:Contributions/Strife_Onizuka\|contribs]])</small></sup> 18:32, 3 December 2010 (UTC)

Strife Onizuka: Created page with 'You seem to have done some testing for llGetCenterOfMass. I'm curious, if you change the shape of the object in ways that should change the center of mass while it's non-phys...'

2010-12-03T18:32:51Z

Created page with 'You seem to have done some testing for llGetCenterOfMass. I'm curious, if you change the shape of the object in ways that should change the center of mass while it's non-phys...'

New page

You seem to have done some testing for [[llGetCenterOfMass]]. I'm curious, if you change the shape of the object in ways that should change the center of mass while it's non-physical, but after it has been physical (so that there is a cached value for the center) does it invalidate the cached value? So if you call llGetCenterOfMass does it then return the invalid value or the value of llGetPos? If you move the entire object does it return a position adjusted cached value for CoM? -- '''[[User:Strife_Onizuka|Strife]]''' <sup><small>([[User talk:Strife_Onizuka|talk]]|[[Special:Contributions/Strife_Onizuka|contribs]])</small></sup> 18:32, 3 December 2010 (UTC)