revochen Mayne

Perrie Juran wrote: Reminds me of the old Klingon Programmer. Huch nobHa'bogh verenganpu''e ylvoqQo'. (Don't trust a ferengi who refunds you.) :smileyvery-happy:

Ela Talaj wrote: Why would anyone consider 1.5 msec execution time a heavy delay? I don't know. But thats different to the issue here as you have to take the execution time for the whole process. Remember it was ~5s without float text but about ~96s (1.5 minutes) with the llSLPPF function.

Dora Gustafson wrote: I love the example you came up with:) it added some life to the forum Yes, doesnt it? According to the thread views it seems to be popular... somehow. I should post my issues more often here, shouldnt I? I wonder if this thread has a chance to get pinned on top of the forum when it just gets enough views?! :matte-motes-wink-tongue: I don't find it out of scope I rather find it any programmers obligation to think :smileysurprised::smileyvery-happy: Yes, but that for the issue has to be documented. And it is now! \o/ Mission accomplished :matte-motes-nerdy: :matte-motes-wink-tongue:

There might be very rare situation for such a heavy process like mine measuring a regions landscape in combination with a float text but they exist (otherwise we wont have the chance to discuss about it here ). I rather tend to declare it being out of scope for a scripter to take care of the cpu shared time when doing function calls within iterations. Or do you always check if a function is getting its right portion of cpu time? Oô

Your redundant wiki is missing a note about script delay when excessive using float text in a while loop! :matte-motes-evil-invert: :matte-motes-wink-tongue:

When the documentation sais no delay, then i dont expect a delay, no matter if its a native function delay or caused by the simulators cpu shared process time. Or do you expect rain and taking an umbrella with you if the weather report sais it will be a sunny day and 80°F/30°C?

Nova Convair wrote: ...and the delay of this functions I don't see as a delay since it will not have an effect in a real script. It will only have an effect in a synthetical benchmark like the OP's example. Hello Nova! I'm sorry to say but you're wrong. My script was so real as can be when i measured the regions landscape. As i knew that 65K iterations might takes some time i wanted to use a float text to show the progress. Then when i figured the delay i tried to monitor process times. Updating float text within simple loops doesnt seems to be an issue if its just about a few duzends iterations or less. But it is in cases like mine.

Qwalyphi Korpov wrote: "It is generally not useful to update the floating text 1400 times per second." Uhm, this might going to become my signature-of-the-day! *ROFL

Qie Niangao wrote: This is a pretty fun thread. Hey! This is serious as can be! :matte-motes-wink-tongue: Qie Niangao wrote: I'd just add that llSLPPF has a built-in disadvantage in that it takes a list argument, so the script spends time consing up that list to pass it to the function, whereas llSetTest is passed atomic, non-list arguments. Yes that sounds logical but just take a look at the wiki what it sais about llSLPPF: "This function is identical to llSetLinkPrimitiveParams except without the delay." But this doesnt seems to be true as there is a delay anyway and scripters shall be aware of as it might improves their scripts when they avoid trying to update text a few thousend times each second. :matte-motes-nerdy: :matte-motes-wink-tongue:

Hello Miranda! Thank You, thats a great solution and only takes 6.2s to process which is a burst related to the 96s befor. I also still dont agree with Lepre, steph and Q about its a useless hint and most people are already aware about. I even think about making another hint about at the wikis flow control/while page now.Or would that be to excessive? :smileylol:

Dora Gustafson wrote: Seriously, what sane person would update llSetText more than 2000 times per second??? Me did Oô

Qwalyphi Korpov wrote: On the other hand it might be good to add a tiny note WIKI about llSetText: "It is generally not useful to update the floating text 1400 times per second." Good idea! Got this added on the wiki/llSetText caveats part: Measurements showed a high impact of process time when doing numerous iterations in a while loop. For approx. 65K iterations the process times are ca. 5s without float text, 24s with llSetText and 96s when using llSetLinkPrimitiveParamsFast/PRIM_TEXT. Thats why its not adviced to make excessive use of a float text within such iterations.

Hi again! =) Yes, he already replied on my jira report and i got a small note about added on the llSetLPPFast wiki page. Just within the small yellow attention box at the caveats. Important:.... Also don't expect llSetLinkPrimitiveParamsFast being faster than llSetText when setting a float text property within a loop like at "% loading" status bars. Measurements showed llSetText being 3-4 times faster. That also might applies for similar functions. @ LepreKhaun Whats the profit or sense of creating and editing redundant wikis and articles? Oô

All i am is being tired so guess i have to use llSleep now Nite =)

Yes would be good to have a note about somewhere in the wiki. Those fast and heavy processes are rarely i guess but possible when i think of a "...%loading" float text and if one know that it might takes a minute instead of 5s when using float text they might resist in using it. Got a bug report filed on jira but i have no idea how or even if i'm also able to write in the wiki and got a note about on it. https://jira.secondlife.com/browse/BUG-3295

Yes thats right but its rather about the delay in general. Using the llSetLinkPPFast will result in a 20 times slower process which effects any item showing a fast process progress as float text. Using landscape measuring was just an example i did today and figured the issue. And i would rather expect the llSLPPFast function being faster than llSetText as well. Btw i was on a quiet homestead sim and not sure how it behaves on other simulator types yet.

Greetings! I was measuring a regions landscaping by script and observed a very heavy delay when just using an additional float text function or the llSetLinkPrimitiveParamsFast within a while loop. It really confused me as i didnt expect such results when i tried again and stoped time by using the llResetTime for each different process type. Well the process makes quite a lot of tasks (256*256) but can just one function effect the process such bad? The measured times for are: 4.6s without float text used 24.4s with llSetText used 96.6s with using llSetLinkPrimitiveParamsFast (!) How can that be? Here is the code for, just uncomment one of the float text functions in line 37/38 and figure. It really seems to be about the functions themself. If you just put random text into it will have the same delay effect. default { touch_start(integer total_number) { llResetTime(); integer size = 256; float water_level = (integer)llWater(ZERO_VECTOR); integer row = 0; integer column = 0; integer water_count; integer land_count; integer hill_count; integer cell_size = 1; vector current_pos = llGetPos(); integer process_count = (size/cell_size)*(size/cell_size); while(column<=size&&row<=size) { column+=cell_size; if(column>=size) { column = 0; row+=cell_size; } vector pos = <column,row,0>; float ground_level = llGround(pos-current_pos); if(ground_level<water_level) ++water_count; else if(ground_level < 50) ++land_count; else ++hill_count; --process_count; //llSetLinkPrimitiveParamsFast(0,[PRIM_TEXT,(string)(process_count/ ((size/cell_size)*(size/cell_size)/100)),<1,1,1>,1]); //llSetText((string) (process_count/ ((size/cell_size)*(size/cell_size)/100)),<1,1,1>,1); } float time = llGetTime(); llOwnerSay("water count:"+(string)water_count+"\nland count:"+(string)land_count+"\nhill count:"+(string)hill_count+"\ncall time:"+(string)time+"s"); } }

Had to figure on the first test thats its not reliable for my purpose. I had the baby in crawling position and want it to change into stand or sit which is just a 90 degree spin on its local X axis. When applying llLookAt it also turned on its Z axis and thats just what i want to avoid.

Oh ok, i will give it a try again then.

And i just figured the objects local forward axis is the Y, left/right is X and up axis is Z. So i guess i rather have to use llRot2Left instead of llRot2Fwd... but how?! óò It's so frustrating search way to long for a solution on and the wiki wasnt much of a help about. It even left more questions than it could answer. Also messing with quaternions is a pain. When running a few test i got values like <-0,0,1> which was totally new for me that there is a negative zero (-0) existing! Oô So to bother you with my complains but i just had to, not feeling better tho. *sighs*

Hello and thank you for your reply but this also doesnt seems to resolve my problem. llLookAt will point the X axis towards a target vector but all i want is the the Z axis going straight up. Innula Zenovka wrote: How does bumping the doll move it in the first place, though, unless it's a physical object? The doll was just an example although its a baby im working on. The baby is changing its positions to either stand, sit, crawl or lay and for thats what i need the proper orientation for based on its current rotation. I dont want to hard-code it and thought there is a function in LSL to rely on. I believe its about the llRot2Up and llRot2Fwd but i tried anything back and forth without success but more confusions about.

Ok sorry, the thing i'm searching for seems to be the objects upwards direction. Imagine a roly-poly doll, which always stands up by its own to achieve its up direction. So when the doll gets bumped and changing its x + y angle, how do i get it to stand up again as a non physical object?

Hi Doro =) Yes, you are right. It was supposed to be a rotation type but the code doesnt solves my problem anyway. I even wonder why its using a llRot2Fwd function when i want to get its up direction? Thats the code im using in a test pyramide prim and just want it to rotate up based on its current rotation. default{ touch_start(integer total_number) { rotation rot = llGetRot(); vector currentRot = llRot2Fwd(rot); vector rotToUp = llRot2Up(rot); rotation setRot = llRotBetween(currentRot,rotToUp); llSetRot(rot*setRot); }} I'm messing with it for a few days now and start to pull out my missing hair soon. óò

Hello! I need some help with setting an objects forward and up rotation based on its current rotation set. I tried using llRot2Up but wasnt able to figure out how to use it proper. The wiki wasnt a help with as its providing very least informations about. I tried ... vector currentRot = llGetRot(); vector rotToUp = llRot2Up(currentRot); llSetRot(currentRot*llEuler2Rot(rotToUp)); ... but it doesnt seems to do what i need. Anyone may point me to the right direction please?

Happy New Year! We're looking for new tenants on our sim Ocelot Valley. Its for residential and commercial use, except clubs. The conditions for a 1/4 parcel are: 8.128sqm, terraforming allowed, 3.720 prims, 22.000L$ each month. More parcel sizes are available. http://slurl.com/secondlife/Ocelot%20Valley/139/124/22 If you're interested in or for more informations IM to Revochen Mayne Thank You! =)