Documenting alchemizing UV coincidences (in development)
Mar, 06/09/2015 - 06:06
In order to shed some light on a theory I have developed I would like to ask you the community to help me out.
http://forums.spiralknights.com/en/node/110267
Theory: When you alchemize an item they UVs you receive can be influenced by what levels you heat them in.
Testing: When Heating an item for the experiment be sure to heat it on one single floor (and difficulty) to maximize any effect it may have on the supposed chances.
Positive results: Alchemize and item up and acquire a UV that benefits the situation that it was heated in.
Example: Farm first floor of Occurance at Owlite Keep and when alchemizing receive UVs that include Stun Curse and Shadow Resists as well as damage bonuses against undead for weaponry.
Negative results: Alchemize an item up and acquire a UV that is not relevant to the situation that it was heated in.
Example: Farm first floor of Occurance at Owlite Keep and when alchemizing receive UVs that DO NOT include Stun Curse and Shadow Resists as well as damage bonuses against undead for weaponry.
Be sure to include:
Date Alchemized
Item Alchemized (from X to Y)
UV acquired (if none acquired still report)
Level heated in and difficulty you selected
NOTE: The item in question must solely be heated in one level
Be sure to mail me in-game @Holy-Nightmare so I can add you to a list of participants and be sure to tell me your plans for items and where you are planning to test.
Floors suggested:
Alone in the Dark / It came from Below
Shock/Undead
Terminal Meltdown
Fire/Slime
Occurance at Owlite Keep
Stun/Undead
Weight of Darkness
Poison/Undead
Sewer Stash
Poison/Slime
Whipping and Mishandling
Undead
High Temperature Hostages
Fire/Beast
Sovereign slime
Slime
Rise or Fall
Shock/Beast
Blades of the Fallen
Stun/Fiend
Faith in armor
Construct
Mar, 06/09/2015 - 07:08
#2
@ Bopp
True, having counts of no UV at all would help out in the statistics. There might be some other hidden effect to farming single level, maybe a higher or lower rate of UV acquisition, if anything it would give us another variable to use.
Posting the exact UV you received would be more helpful as it would be more raw data.
Yes it would be nice to have a list of participants too
Mar, 06/09/2015 - 07:51
#3
response
Posting the exact UV you received would be more helpful as it would be more raw data.
Yes, definitely you should have your participants report the exact UVs that they get. How about text like this...
Report these five pieces of information:
1. Date, such as "2015-06-09".
2. Item, such as "Silent Nightblade from Nightblade".
3. Level, such as "first level of Alone in the Dark".
4. Difficulty, such as "Elite".
5. Outcome, such as "Fiend Medium".
Important: Once you've decided to collect to collect a piece of data for us, then you must report your outcome, even if it is "boring". For example, if you get no UV at all, then report that.
Are there mission levels that have different versions with different themes? If so, then it's important that your participant heat in a single theme. Maybe you should explicitly forbid that case.
Mar, 06/09/2015 - 08:05
#4
@ Bopp
I like your proposed layout of information
For the sake of the research I would have to say that you must farm a single floor. Themes have a way of being misleading
(*cough* Arenas *cough*)
I might also have to exclude some floors due to their wide variety of monsters (for example: IronClaw Munitions Factory have Slimes, Constructs, and Gremlins). Missions like the first floor of alone in the Dark have 90% undead, a Trojan here or there, and for some reason Polyps can spawn in this mission (not sure if bug or OOO being trolly).
Basically for a floor to be accepted it has to have massive amounts of one type of monster
Mar, 06/09/2015 - 09:05
#5
but I mean
For the sake of the research I would have to say that you must farm a single floor. Themes have a way of being misleading
(*cough* Arenas *cough*)
Right, but that's not enough to solve the problem. For example, there are two possibilities for the first floor Rotting Metropolis. I think that they have the same mix of monsters and statuses, so saying "first floor of Rotting Metropolis" is probably fine. But maybe there are other missions where the floors come in truly different versions. I think that you need to check the gate map of each mission.
For the sake of argument, suppose that the first floor of The Gauntlet had a shock version and a fire version. And you got a report saying that someone had grinded that floor and got a shock+2 UV on their armor. Well, did they grind the shock version, the fire version, or a mix? It matters, right?
Mar, 06/09/2015 - 10:06
#6
@ Bopp
"Well, did they grind the shock version, the fire version, or a mix? It matters, right?"
Naturally.
I will probably spend some time researching which mission first floors have the most consistent monster content and limit this experiment to those.
Mar, 06/09/2015 - 18:57
#7
great
Then we agree. Cheers.
Mié, 06/10/2015 - 08:36
#8
↓
I believe the only depth that has only one monster type is devilish drudgeries. Shock themed one appears 1st depth of the gauntlet.
Mié, 06/10/2015 - 08:46
#9
right
Yeah, I knew that the first level of The Gauntlet was always shock. It was just a hypothetical example.
I don't think that Holy-Nightmare needs to restrict to a single monster family in the whole level. The level's monster theme seems like an adequate level of detail. For example, if the level is marked construct, then record that, even if there are some gremlins in there.
Dom, 06/14/2015 - 20:42
#10
Question
If this theory is true then what determines the uvs charge time reduction, and attack speed increase for weapons?
Mar, 06/16/2015 - 11:05
#11
@ Red
Those may be totally up to chance, then again certain types of enemies might actually give Effort Values towards that
ASI might be given out by fast enemies like Kats, Wolvers, Thwackers, and Greavers
CTR might be given from the slower enemies with long windups on their attacks like Lumbers, Giant Lichen colonies, and Trojans
Speculations can be made and now that the BK event is almost over I can get back to work on this theory (IE determining which levels/missions should be used in this test, setting up spreadsheets, etc)
Mar, 06/16/2015 - 11:30
#12
I dont think thats how it
I dont think thats how it works. CTR or ASI cant be up to chance, while all the other ones can be coincidences.
Mar, 06/16/2015 - 11:51
#13
I'm watching O.O
I'm sure we will eventually see how things work (or don't work) with implied patterns in supplied data, once we have a huge amount to analyze anyway. I will be watching this thread for results, and am looking forward to being able to contribute to some sort of public table/spreadsheet.
Let's fail to reject the null hypothesis!
Mié, 06/24/2015 - 13:04
#14
~~~
I think we're turning this into pokemon here. We don't need EV training.
I'd like to participate in this, though. Sent you mail more indepth, of course. I'll be starting a spreadsheet.
Working on spreadsheet, current information here.
Time of Creation - 6/24/2014
Item - Frostbreaker Shield
Level Upgraded - N/A
Level Farmed - Blades of the Fallen (T1)
Heat Percentage - 100% (Always 3rd choice)
Number of Deaths - 1
Heat Bonus? Double Level Up? - Yes. 25% Heat Bonus at Level 2. Double Level Up to Level 4. Heat Bonus at 5.
Weapon Only/Primarily Used - Sealed Sword
Difficulty - Elite
Special Notes - Server Reboot at Level 5.
Just wanted to show you the stuff I'm currently jotting down. You can pull out whatever you want once I finish the 9 or more I'm planning to do. Of course, I'll be giving a link to the actual spreadsheet, not just splatting it down here on one post.
Here it is. I'm not sure how much I'll be able to do, since I really don't have CR to play with at this cost. (60 CE for orbs + 1000 CR for 50 Dim Crystals + 600 CR Alchemy Cost) Even though that's kind of low, it'll be adding up for each I do, and thus my planned 9 would be 540 CE + 14.4k CR. That's a lot, which equals roughly 700 CE. Perhaps when I get a considerable amount more to play around with, but not at the moment.
Vie, 06/26/2015 - 05:04
#15
↓
@Menish
When I started my forge project, I used boss token reward items to heat. That's why there's around 200-300 more datapoints on 3* equip for that data (20-30 items worth). I just looped heating/farming quicksilver pieces at IMF. Once one was done heating, I had the tokens to get the next.
Also, your spreadsheet isn't currently set for public access (you can modify it in the share options to "anyone with link can view").
Lun, 07/06/2015 - 18:52
#16
1. 2015-07-06 (July 6th)
2. Flourish (2*)
3. First level of "Sewer Stash". I farmed it about 30-50 times total prior to crafting
4. Elite
5. Slime High (LOL)
Note that this was the SECOND Flourish I crafted at the time. I think this theory may be true.
Lun, 07/13/2015 - 09:01
#17
@ Menish
Your spreadsheet is private and needs to be made visible without anyone needing permission to see.
____________________________
Update: Data is being compiled
Current gear being worked on: Plated Falcon Pathfinder (with low shock and pierce resist) in Sewer stash first floor
Not many players read this forum, although those that do tend to be fairly serious. So you might get a very small but high-quality data set out of this post. Consider putting this same post into your General Discussion thread, to attract more attention.
To avoid observation bias, I recommend that you firmly emphasize the following point. A player participating in this study must decide ahead of time to participate. That is, s/he must say to herself, "I am going to heat this item now by repeatedly running a single level, and after I craft I will report my outcome to Holy-Nightmare, regardless of what it is."
Most (about 90%) of the outcomes will be no UV at all. Have players report that too, not because it matters to your study, but just because it helps enforce the discipline described in my previous paragraph.
Don't have participants report "the favorable UVs that would benefit the item in that level". Figure those out yourself. Why? Because you want to make their job as easy and simple as possible, to encourage them to actually do it.
Of course, take or leave these suggestions as you like. They're just my recommendations, based on a lot of collaborative data collection. Good luck.