Expensive Vanilla Scripts

Litcube · Post by **Litcube** » Sat, 9. Jul 11, 21:12

Halconnen wrote: Edit: @Litcube: Where is the LeadShipWingMonitor in your download from? I seem to have a byte-identical version with same changedate from some other source (Modded install here, too), and am curious where I got it from.

Wierd, dude. I wrote it. I haven't included it in any of my other packages either.

What do you have installed?

Re Glen's fix: I doubt there's a way to fix those missions performance wise. Ironically, in my mission fix (in my sig), I had already removed those missions. Mizuchi brought that to my attention after I had them posted for about a year. I figure Glens method is probably the best. MD just blows.

Mizuchi · Post by **Mizuchi** » Sat, 9. Jul 11, 22:14

Litcube wrote:I had already removed those missions.

And nothing of value was lost.

Halconnen · Post by **Halconnen** » Sat, 9. Jul 11, 23:08

Litcube wrote:Wierd, dude. I wrote it. I haven't included it in any of my other packages either.

I have SRM+LowHull+CMOD+MRM+Lucike's stuff+Much of Gazz' stuff+low poly roids+shrunken Alrin roid+Logain Abler's stuff+Phanon Corp on this.

I also added RRF but that doesn't seem to play nice, since I've been crashing randomly since. Probably going to remove that again.

But going through the archives I used, I can't find the script file in them.

Which is what made me ask the question in the first place, I am honestly confused where it came from. I just dragged and dropped the files from your zip into my folder after renaming the one .pck, and it asked me to overwrite that file, at which point I went "waitwhat".

Litcube · Post by **Litcube** » Sat, 9. Jul 11, 23:12

I'm pretty sure you're confusing which file is overwritten. There is an overwrite, but it's the move.follow.template, not he LeadMonitor.

joelR · Post by **joelR** » Sat, 9. Jul 11, 23:15

Halconnen wrote:
Litcube wrote:Wierd, dude. I wrote it. I haven't included it in any of my other packages either.
I have SRM+LowHull+CMOD+MRM+Lucike's stuff+Much of Gazz' stuff+low poly roids+shrunken Alrin roid+Logain Abler's stuff+Phanon Corp on this.

I also added RRF but that doesn't seem to play nice, since I've been crashing randomly since. Probably going to remove that again.

But going through the archives I used, I can't find the script file in them.

Which is what made me ask the question in the first place, I am honestly confused where it came from. I just dragged and dropped the files from your zip into my folder after renaming the one .pck, and it asked me to overwrite that file, at which point I went "waitwhat".

Oh no....Its happening again.....

Mizuchi · Post by **Mizuchi** » Sat, 9. Jul 11, 23:26

joelR wrote:Oh no....Its happening again.....

[ external image ]

creidiki · Post by **creidiki** » Sat, 9. Jul 11, 23:28

My list looks like his with 12 or so more stuck on at the end >_>

Halconnen · Post by **Halconnen** » Sun, 10. Jul 11, 01:45

Litcube wrote:I'm pretty sure you're confusing which file is overwritten. There is an overwrite, but it's the move.follow.template, not he LeadMonitor.

'fraid I'll quote myself for emphasis:

I just dragged and dropped the files from your zip into my folder after renaming the one .pck

With which I meant I renamed the vanilla !move.follow.template.pck to !move.follow.template.bak, then added the new files on top. Me doing that is the main reason I got confused by the popup asking about the overwrite. And I remember looking at the 31051 byte size of the Monitor afterwards.

Thinking about it, though, I am not sure. Maybe it's the RunScriptWhenFinished that got overwritten. Does that sound more plausible?

Litcube · Post by **Litcube** » Sun, 10. Jul 11, 02:03

Halconnen wrote:Thinking about it, though, I am not sure. Maybe it's the RunScriptWhenFinished that got overwritten. Does that sound more plausible?

Yes. Almost all my scripts have that. Including Phanon Corporation.

joelR · Post by **joelR** » Sun, 10. Jul 11, 02:23

Litcube wrote:
Halconnen wrote:Thinking about it, though, I am not sure. Maybe it's the RunScriptWhenFinished that got overwritten. Does that sound more plausible?
Yes. Almost all my scripts have that. Including Phanon Corporation.

Whew. That went better than expected.

joelR · Post by **joelR** » Sun, 10. Jul 11, 06:01

glenmcd have you made a mod to remove all missions? I wanted to try something.....I actually rarely do missions anyhow. Personally I prefer making my fortune other ways and wanted to see if there would be much of a performance increase if no missions at all spawned...

glenmcd · Post by **glenmcd** » Sun, 10. Jul 11, 06:26

joelR wrote:glenmcd have you made a mod to remove all missions? I wanted to try something.....I actually rarely do missions anyhow. Personally I prefer making my fortune other ways and wanted to see if there would be much of a performance increase if no missions at all spawned...

Yes I have, and it's already included in the .zip file linked to above in previous page. Better yet, I spent all day yesterday doing benchmarks of each and every generic mission seperately, including fps decrease over a track consisting of ten sectors one way and then reverse, followed by timing how long it took to fly between two sectors (hold down tab in Jaguar) that I'd completely emptied, even of asteroids, stations and unneeded gates, followed by killing all ships in universe for one more benchmark. This last test was good for determining how strongly associated with ships the performance decrease is/was. The result of that one I think will surprise you in at least one case. Columns F-H I used to prioritise the results, so I could feed this into a rebalancing of the relative spawnings of generic missions. My intention here is/was to very slightly decrease the frequency of those missions with worst performance, in addition to completely removing the convoy ones. Also, instead of simply disabling the convoy missions, I've made them a non-event. The original mod increased the frequency of all missions other than convoy, to "make up the numbers". I figured as some of the others were a noticable performance hit (but not in same league as convoy), why increase the frequency of their spawning? This is still a work in progress and I can see that it will require a number of days of tedious testing. The way the RNG works in TC (regular long runs of things) makes it difficult for testing some things.

Feel free to do your own tests, use mine, sell it to your nephew or whatever.

http://www.glenmcd.com/Generic_missions_benchmarks.xls

joelR · Post by **joelR** » Sun, 10. Jul 11, 07:00

glenmcd wrote:
joelR wrote:glenmcd have you made a mod to remove all missions? I wanted to try something.....I actually rarely do missions anyhow. Personally I prefer making my fortune other ways and wanted to see if there would be much of a performance increase if no missions at all spawned...
Yes I have, and it's already included in the .zip file linked to above in previous page. Better yet, I spent all day yesterday doing benchmarks of each and every generic mission seperately, including fps decrease over a track consisting of ten sectors one way and then reverse, followed by timing how long it took to fly between two sectors (hold down tab in Jaguar) that I'd completely emptied, even of asteroids, stations and unneeded gates, followed by killing all ships in universe for one more benchmark. This last test was good for determining how strongly associated with ships the performance decrease is/was. The result of that one I think will surprise you in at least one case. Columns F-H I used to prioritise the results, so I could feed this into a rebalancing of the relative spawnings of generic missions. My intention here is/was to very slightly decrease the frequency of those missions with worst performance, in addition to completely removing the convoy ones. Also, instead of simply disabling the convoy missions, I've made them a non-event. The original mod increased the frequency of all missions other than convoy, to "make up the numbers". I figured as some of the others were a noticable performance hit (but not in same league as convoy), why increase the frequency of their spawning? This is still a work in progress and I can see that it will require a number of days of tedious testing. The way the RNG works in TC (regular long runs of things) makes it difficult for testing some things.

Feel free to do your own tests, use mine, sell it to your nephew or whatever.

http://www.glenmcd.com/Generic_missions_benchmarks.xls

Interesting. I'll see how it plays out on my computer.

Which one of the director files is it? There were 3 in the download from page 4.

glenmcd · Post by **glenmcd** » Sun, 10. Jul 11, 07:14

joelR wrote:Which one of the director files is it?

3.01 Generic Missions_nogenericmissions.xml

It's easy to spot. It's the one I've added just now

joelR · Post by **joelR** » Sun, 10. Jul 11, 07:24

glenmcd wrote:
joelR wrote:Which one of the director files is it?
3.01 Generic Missions_nogenericmissions.xml

It's easy to spot. It's the one I've added just now

Is there another link im not aware of? Im using the link on page 4 of this thread and Im not seeing that in the .zip file

glenmcd · Post by **glenmcd** » Sun, 10. Jul 11, 07:39

joelR wrote:
glenmcd wrote:
joelR wrote:Which one of the director files is it?
3.01 Generic Missions_nogenericmissions.xml

It's easy to spot. It's the one I've added just now
Is there another link im not aware of? Im using the link on page 4 of this thread and Im not seeing that in the .zip file

I'm leaving that one as is and have created another:
http://www.glenmcd.com/GenericMissionsTestMods.zip
Sorry, I realize I hadn't shared the new file URL here as yet.

The edits are super easy to spot. Look for a table of missions, where I've added either "*1" or "*0" to the end of an expression that sets the chances of..
By making it zero, you effectively disable. So if you want to customise that's how you can easily do it. Note if you leave only one mission you'll get a LOT of them.

joelR · Post by **joelR** » Sun, 10. Jul 11, 14:19

glenmcd wrote:
joelR wrote:
glenmcd wrote:
joelR wrote:Which one of the director files is it?
3.01 Generic Missions_nogenericmissions.xml

It's easy to spot. It's the one I've added just now
Is there another link im not aware of? Im using the link on page 4 of this thread and Im not seeing that in the .zip file
I'm leaving that one as is and have created another:
http://www.glenmcd.com/GenericMissionsTestMods.zip
Sorry, I realize I hadn't shared the new file URL here as yet.

The edits are super easy to spot. Look for a table of missions, where I've added either "*1" or "*0" to the end of an expression that sets the chances of..
By making it zero, you effectively disable. So if you want to customise that's how you can easily do it. Note if you leave only one mission you'll get a LOT of them.

So there's no way to alter the chance of spawning any given mission?

I was thinking about leaving the buy ship going but that seems kind of crazy if I end up getting a bajillion of them.

glenmcd · Post by **glenmcd** » Sun, 10. Jul 11, 15:14

joelR wrote:So there's no way to alter the chance of spawning any given mission?

I was thinking about leaving the buy ship going but that seems kind of crazy if I end up getting a bajillion of them.

There certainly is. Just before my "*0" or "*1"'s, there's a number in range 40 to 100. Adjust down to make the mission spawn less frequently in comparison to other missions. I'm not certain but it looks like if you multiplied all by ten, it would change nothing.

I've had another thought on how to go about this. Use dynamic spawn chances. Ignore a mission type and it will go away (almost). Keep accepting a particular mission type regularly and you'll continue to see it spawn as regularly as it does now. If you change, so does it. The only missions that I'd make a bit special are the CPU overloaders, escort convoy and destroy convoy. Still start them at default, but diminish their chances 2 to 3 times faster than other mission types.

There needs to be strict limits on the upper bound of generic mission spawns, so that they don't exceed current (average) levels for each mission type. Otherwise it becomes a cheat. Any mission that isn't accepted at all may diminish to extremely rare spawns. After all it's always been your choice to reject any and all generic missions. At game start, default chances will be as per vanilla. I think if done this way, most players would hardly notice the change. They would notice if and when they use this for a long time and then suddenly go back to the current static system.
Another benefit of going this way is that when you go looking for suitable missions, you will likely find the one(s) you want sooner. Not because there will be more of them, but because you won't need to wade through dozens of crap (judged by your history of rejections) ones every time. See an icon? There's a good chance that it will be something of interest

joelR · Post by **joelR** » Sun, 10. Jul 11, 19:17

The dynamic mission spawning is a great idea.
So just to be clear, the number has to be between 40 and 100 when adjusting the chance of a particular mission spawning in sector?

So I assume that means the absolute minimum chance is 40%?

HotSake · Post by **HotSake** » Sun, 10. Jul 11, 22:05

Is it a percentage chance or a weighted system? I think that value, compared to other missions, will tell you the relative frequency of spawning. i.e. a mission at 80 spawns twice as often as a mission at 40, not 80% of the time. To get the true chance of spawning, you'd first have to sum all those values, then divide the particular mission's priority by the total. That gives you its proper chance to be selected, considering the weighting.

In other words, imagine adding together all those values and rolling a die with as many faces as the total. A mission with a low priority is only picked on, say, 40 rolls out of your 500-sided die, while a mission with high priority is picked on 100 rolls. So in this example, the first mission has a 1 in 12.5 chance to be picked when a new mission is created, while the second has a 1 in 5 chance.

egosoft.com

egosoft.com

Expensive Vanilla Scripts