Difference between revisions of "User talk:CyberSoul"

From Ultronomicon
Jump to navigation Jump to search
(→‎Mineral-Rich Worlds and Bio-Rich Worlds: Note relevant discussion elsewhere.)
 
(2 intermediate revisions by 2 users not shown)
Line 30: Line 30:
 
:I would also recommend formatting the infodump with wiki-table syntax. I tried my hand at one of them, [[List of top early biological planets]], to see how it would look. One of the big benefits of this IMO is the sortable columns which give the reader a few more options to extract the information he/she is looking for. The formatting could most likely still use some aesthetic tweaking.
 
:I would also recommend formatting the infodump with wiki-table syntax. I tried my hand at one of them, [[List of top early biological planets]], to see how it would look. One of the big benefits of this IMO is the sortable columns which give the reader a few more options to extract the information he/she is looking for. The formatting could most likely still use some aesthetic tweaking.
 
:Also, if you're interested in how far down the rabbit hole you can go with the risk assessment of the planets, you might be interested in some previous (though now somewhat dormant) discussions on the topic: [[Talk:Planet landing risk-reward formula]]. --[[User:Fyzixfighter|Fyzixfighter]] 21:40, 5 April 2011 (UTC)
 
:Also, if you're interested in how far down the rabbit hole you can go with the risk assessment of the planets, you might be interested in some previous (though now somewhat dormant) discussions on the topic: [[Talk:Planet landing risk-reward formula]]. --[[User:Fyzixfighter|Fyzixfighter]] 21:40, 5 April 2011 (UTC)
 +
::I like the wiki table format, but will start with getting the raw data there. Yeah, I think that risk-reward analysis got too complicated for its own good, but it's definitely a great read for the inner math geek. Still, its surprising how complex the problem gets in a hurry. The comparatively simple algorithm I use to generate the planet risk ratings is the Python code below: --[[User:CyberSoul|CyberSoul]] 03:05, 6 April 2011 (UTC)
 +
 +
<nowiki>THREAT_TEMP_BOUNDS = [100, 200, 400]
 +
THREAT_WEATHER_BOUNDS = [2, 3, 4]
 +
THREAT_TECTONICS_BOUNDS = [2, 4, 6]
 +
THREAT_LEVELS = ["none", "some", "moderate", "high", "extreme"]
 +
def _get_index(self, var, bounds_list):
 +
        templist = [var] + bounds_list
 +
        templist.sort()
 +
        return templist.index(var)
 +
def threat_index(self):
 +
        tempidx = self._get_index(self.temp, THREAT_TEMP_BOUNDS)
 +
        widx =  self._get_index(self.weather, THREAT_WEATHER_BOUNDS)
 +
        tidx = self._get_index(self.tectonics, THREAT_TECTONICS_BOUNDS)
 +
        maxidx = max([tempidx, widx, tidx])
 +
        sumidx = sum([tempidx, widx, tidx])
 +
        if maxidx <= 0:
 +
            threatidx = 0
 +
        elif maxidx <= 1 and sumidx <= 1:
 +
            threatidx = 1
 +
        elif maxidx <= 1 and sumidx <= 3:
 +
            threatidx = 2
 +
        elif maxidx <= 2 and sumidx <= 5:
 +
            threatidx = 3
 +
        else:
 +
            threatidx = 4
 +
        return threatidx
 +
def threat(self):
 +
        threatidx = self.threat_index()
 +
        return THREAT_LEVELS[threatidx]</nowiki>
 +
 +
:All seems good, except I don't think 'NW of Pkunk space' should be present on the table. I'd opt to change "Location" to "Territory" and mention Ilwrath Space, Ur-Quan/Kohr-Ah Space, etc. Bio risk will be a bother for people to fill in, but it shouldn't be too hard to approximately quantify. If the life forms are too difficult to put down without lander upgrades, the risk level is high (Fungal Squids?) or extreme (ZEX's Beauty). --[[User:Shiver|Shiver]] 22:54, 5 April 2011 (UTC)
 +
::While I agree that knowing a planet is in the S part of Arilou space doesn't add a whole lot over just Arilou space, I think the sector information is quite useful in the larger Kohr-Ah space. Algorithmically, I just use the vector direction to determine the sector, so I don't currently detect if a planet is actually near the middle of the space, but that may be a useful improvement. I could also potentially eliminate the sector information for smaller territories. I think quantifying the bio risk is relatively easy, as it seems to correlate well with the speed, durability, and possibly number of hunter type creatures. However, dumping that data out of the program in a useful form is not on my current todo list. Tentatively, squid/Zex = extreme, bees/yiminy = moderate, other hunters = some. --[[User:CyberSoul|CyberSoul]] 03:05, 6 April 2011 (UTC)
 +
 +
Hello CyberSoul.  I've done some code diving and come up with more detailed information on planetary risk.  In particular, where the actual breakpoints are for hotspot hazard, and a bit more in-depth analysis of hostile lifeforms.  I've also got some suggestions on other columns that it would be useful for your program to calculate.  Please see the discussion on [[Talk:List of top mineral planets]].  --[[User:Elestan|Elestan]] 14:44, 15 June 2011 (UTC)
  
 
==Quasi-Degenerate Worlds==
 
==Quasi-Degenerate Worlds==

Latest revision as of 14:44, 15 June 2011

Gem Worlds[edit]

Gem worlds getting their own pages? You know, that's not a bad idea! Welcome to Ultronomicon, newcomer. --Shiver 04:33, 26 March 2011 (UTC)

Thanks for the welcome and support. I saw a partial list of the gem worlds in a walkthrough and figured out how to extract the full list (after a while!). At first, I naively searched high and low for static data somewhere in the codebase :) Plus IIRC, the top 21 mineral planets are all gem planets! Do you like the presentation format? I was thinking about using it in a couple more pages, and it would be easier to change now rather than later. Once the format is ready, I would like to add the same data to the rainbow world page, as some of them are quite hostile while others are calm. CyberSoul 04:58, 26 March 2011 (UTC)
I can't speak for the other admins but that all sounds good to me. This wiki could use more RU and bio collection information. --Shiver 17:42, 26 March 2011 (UTC)

The topic was briefly discussed a long time ago (see the top of Talk:List of planet types) and it seems that the decision was made to replace all the separate pages that existed at the time (ie Water Worlds, Ruby Worlds, Yttric Worlds, Auric Worlds, Lanthanide Worlds) with the single list. If each new page is merely a repeat of the couple sentences found at the list, then I don't think each planet type requires its own page. The one thing that CyberSoul has added, which wasn't there before, are lists of location and information about instances of those planet types. Does that addition warrant the re-creation of the individual pages? I'm not certain, but I currently lean towards no - however, perhaps it would be appropriate to have a List of Ruby Worlds and similar pages with that information in a more wikified type table (something along the lines of User:Jacius/Planets maybe or an expansion of the table at Rainbow World with columns for the additional planetary data). The downside to this proposal for me is where do we draw the line of inclusion, in other words how do we choose which planet types to do lists for - do we just do it for the exotic worlds? --Fyzixfighter 17:58, 29 March 2011 (UTC)

I have a counter-proposal: Discard CyberSoul's recent work and create new pages for mineral-rich worlds and for bio-rich worlds. This information is too useful to leave off the wiki. --Shiver 19:26, 29 March 2011 (UTC)
Starting with the gem types, I had found the information very useful in game, and I could easily identify where to put it by analogy with Rainbow Worlds, so those pages were born. But as my capability to mine and use the data progressed, I sort of agree with Shiver. Indeed, I played with making in game decisions based on the data, moving to using either a planet-base view or a systems-based view (ala RPRG). The planet based view is much better, because player time is an important factor, and armed with a map you can find a good route to quickly hit all the good planets in multiple nearby systems. So a list of top X planets by mineral and bio yield is most likely a good idea, and X=100 seems to work well when I look at the resulting lists. Mid-to-late game those are great lists, and I can generate them quite easily. But they are not as good early game, when you don't have the technology to legitimately access hostile worlds, and you can't afford to go half-way across the map to find good planets. Thus I developed the capability to filter by risk and location, so I also propose lists for the top low-risk planets near Sol, top 50 mineral and top 30 bio seem to work well. -- CyberSoul 03:44, 30 March 2011 (UTC)

Mineral-Rich Worlds and Bio-Rich Worlds[edit]

Mineral and bio list pages look good. I'm not sure about using planets over systems for early resource gathering when fuel is scarce, but then again the act of scouring an entire system to find 1-3 good planets is not so much fun. The gem world pages will be deleted soon, but I'll hold off on that a while longer. --Shiver 15:06, 30 March 2011 (UTC)

Yes, I was also surprised that the planet data was so much more useful after trying both. However, if you know where the good worlds are ahead of time and plan a good route on a map, the majority of your fuel use come from dropping down the gravity wells, not from actual travel. And then the higher-yielding worlds from the top planet list require less trips down to the surface to get the same RU value. You also end up fitting more RU into your cargo hold at once, saving repeated long trips back to that region of space. --CyberSoul 00:59, 31 March 2011 (UTC)

What about planets in Ilwrath space? Melee veterans can walk right over any number of Avengers, of course, but those are pretty threatening ships to new players. --Shiver 02:01, 31 March 2011 (UTC)

Yes, the Ilwrath concern is definitely real, as my melee skills were only ever mediocre at best, and I just started back after a long while. I've been using similar data to what I uploaded to play through the early game several different ways, with the idea to bootstrap as fast as possible to decent minerals and tech. This last time I went straight at Ilwrath space (Epsilon/Gamma/Beta Volantis), and I only made it out alive after a few different tries. --CyberSoul 05:58, 31 March 2011 (UTC)
It's probably best not to tell players to go digging through Ilwrath space, then. I've always been able to mine Delta Centauri without Ilwrath contact, but I don't know how many battlegroups can appear in that system. If Delta Centauri can theoretically end up packed full of Avengers, I wouldn't recommend entering that system either. Beyond that, I'm not really sure how to rework the early mineral world, early bio world pages. --Shiver 16:06, 31 March 2011 (UTC)
I've been torn on how to handle this, as on the one hand the data is very useful to some players, particularly the bio worlds near earth, which seem more concentrated in Ilwrath space. On the other hand, some planets are hazardous to newer players. I think that a good solution is another column, pointing out the planets that lie in hazardous territory. --CyberSoul 03:54, 2 April 2011 (UTC)
Put the extra column up, I guess. We can always decide to re-do the page later. --Shiver 19:49, 2 April 2011 (UTC)

I'm playing through the game again to check 0.7 RC for bugs and using your early planet recommendations right now. Many of your Tectonics and Weather ratings are off. In particular, your highest rank early bio world (Beta Volantis I-b) is 3 Tectonics / 4 Weather and a bit of a killing field. I don't think newbies should be hunting critters on it. --Shiver 21:19, 3 April 2011 (UTC)

Four weather wouldn't have made the page, as weather is nastier than tectonics. 3 weather/4 tectonics is the nastiest rating that can make the list. When you are mining minerals, those planets kill crew, but usually not at a rate that you will lose a lander (i.e. they have similar risk to Sol's Mercury). However, I must admit that hunting bio seems nastier than mining, because you spend more time on the surface to fill the lander, and because the lifeforms themselves can sometimes kill crew. Also, the risk factors don't include any lifeform risk (i.e. fungal squid are nasty without tech upgrades), as that would be hard to quantify and isn't included in the data dump I did from the game so far. So one thing I could do, it to use more stringent risk thresholds when constructing the early bio list than constructing the mineral list. I think lightning is the nastiest killer here, so I would change the weather threshold to 2 weather, which has no lightning. To get a better idea of where you think the right risk level for tectonics should be, can you try Alpha Wolf IVa and Delta Centauri IIa? --CyberSoul 06:57, 4 April 2011 (UTC)
No really, go and check the planet in-game. The numbers in the source code for tectonics and weather somehow get bumped up in most instances; almost everything is one point higher than it says. Delta Centauri II-a: It's safe enough. The bees are slow and don't take too many hits. Alpha Wolf IV-a: Maybe too dangerous. There are fast, durable critters that pursue the lander and 4 tectonics will definitely hit the lander a few times while the player is trying to kite them. --Shiver 14:39, 4 April 2011 (UTC)
Yes, weather and tectonics are 0-based in the data dump and 1-based when reported to the player, so they are all exactly 1 lower. I adjusted the risk for this long ago, re-uploading the gem pages with a comment about them being 0-based. I went back to the 3 worlds in question, and if you escape out as soon as you lose half the lander crew, the damage rate is never high enough that you lose a lander. Ignoring bio on Beta-Volantis Ib and just straight mining it, I didn't even need to artificially escape. Yet, I agree that the damage rate on Beta-Volantis is higher than on either Alpha Wolf and Delta Centauri, enough to distinguish them somehow. Another factor to consider is that there are so few decent bio worlds near earth already that getting rid of the worlds like Beta-Volantis just makes it a top 20 list instead of a top 30 list. --CyberSoul 06:13, 5 April 2011 (UTC)

Current proposed changes that I working on coding/automating: 1) bump weather and tectonics by 1 in the tables to match the expected representation 2) add a moderate risk category, Beta-Volantis Ib would go here, while Alpha Wolf IVa and Delta Centauri IIa would stay in the some risk category. 3) Add a column commenting on the planets location, which would indicate when it was in Ilwrath space, for instance. Optional changes: add a bio-risk column, which could be manually filled in by wiki users/admins. --CyberSoul 06:15, 5 April 2011 (UTC)

I would also recommend formatting the infodump with wiki-table syntax. I tried my hand at one of them, List of top early biological planets, to see how it would look. One of the big benefits of this IMO is the sortable columns which give the reader a few more options to extract the information he/she is looking for. The formatting could most likely still use some aesthetic tweaking.
Also, if you're interested in how far down the rabbit hole you can go with the risk assessment of the planets, you might be interested in some previous (though now somewhat dormant) discussions on the topic: Talk:Planet landing risk-reward formula. --Fyzixfighter 21:40, 5 April 2011 (UTC)
I like the wiki table format, but will start with getting the raw data there. Yeah, I think that risk-reward analysis got too complicated for its own good, but it's definitely a great read for the inner math geek. Still, its surprising how complex the problem gets in a hurry. The comparatively simple algorithm I use to generate the planet risk ratings is the Python code below: --CyberSoul 03:05, 6 April 2011 (UTC)
THREAT_TEMP_BOUNDS = [100, 200, 400]
THREAT_WEATHER_BOUNDS = [2, 3, 4]
THREAT_TECTONICS_BOUNDS = [2, 4, 6]
THREAT_LEVELS = ["none", "some", "moderate", "high", "extreme"]
def _get_index(self, var, bounds_list):
        templist = [var] + bounds_list
        templist.sort()
        return templist.index(var)
def threat_index(self):
        tempidx = self._get_index(self.temp, THREAT_TEMP_BOUNDS)
        widx =  self._get_index(self.weather, THREAT_WEATHER_BOUNDS)
        tidx = self._get_index(self.tectonics, THREAT_TECTONICS_BOUNDS)
        maxidx = max([tempidx, widx, tidx])
        sumidx = sum([tempidx, widx, tidx])
        if maxidx <= 0:
            threatidx = 0
        elif maxidx <= 1 and sumidx <= 1:
            threatidx = 1
        elif maxidx <= 1 and sumidx <= 3:
            threatidx = 2
        elif maxidx <= 2 and sumidx <= 5:
            threatidx = 3
        else:
            threatidx = 4
        return threatidx
def threat(self):
        threatidx = self.threat_index()
        return THREAT_LEVELS[threatidx]
All seems good, except I don't think 'NW of Pkunk space' should be present on the table. I'd opt to change "Location" to "Territory" and mention Ilwrath Space, Ur-Quan/Kohr-Ah Space, etc. Bio risk will be a bother for people to fill in, but it shouldn't be too hard to approximately quantify. If the life forms are too difficult to put down without lander upgrades, the risk level is high (Fungal Squids?) or extreme (ZEX's Beauty). --Shiver 22:54, 5 April 2011 (UTC)
While I agree that knowing a planet is in the S part of Arilou space doesn't add a whole lot over just Arilou space, I think the sector information is quite useful in the larger Kohr-Ah space. Algorithmically, I just use the vector direction to determine the sector, so I don't currently detect if a planet is actually near the middle of the space, but that may be a useful improvement. I could also potentially eliminate the sector information for smaller territories. I think quantifying the bio risk is relatively easy, as it seems to correlate well with the speed, durability, and possibly number of hunter type creatures. However, dumping that data out of the program in a useful form is not on my current todo list. Tentatively, squid/Zex = extreme, bees/yiminy = moderate, other hunters = some. --CyberSoul 03:05, 6 April 2011 (UTC)

Hello CyberSoul. I've done some code diving and come up with more detailed information on planetary risk. In particular, where the actual breakpoints are for hotspot hazard, and a bit more in-depth analysis of hostile lifeforms. I've also got some suggestions on other columns that it would be useful for your program to calculate. Please see the discussion on Talk:List of top mineral planets. --Elestan 14:44, 15 June 2011 (UTC)

Quasi-Degenerate Worlds[edit]

Quasi-Degenerate worlds might not deserve their own page if the highest mineral yield on any of them is only 1625 RUs. I don't feel too strongly about this, and will wait to see what other admins say. --Shiver 20:23, 27 March 2011 (UTC)

I won't defend it much, but they do carry only exotic minerals, so 1625 RU fits in just 65 cargo space. But Pellucids have only exotics too, and I would agree that they don't deserve a page (unless all types do). "Only" and 1625 RU together puzzles me though, as the 90% percentile for mineral yield on a single planet is 913 RU. Wow, and just for fun the 50% percentile is only 150 RU. --CyberSoul 00:21, 28 March 2011 (UTC)