These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

Science & Industry

 
  • Topic is locked indefinitely.
 

PI Extractor Discrepancy
Author
Previous Topic Next Topic
Dragoon Ree
Infinite Point
Pandemic Horde
#1 - 2011-12-20 06:37:52 UTC
I've been enormously confused as to why one of my planets always gets off balance and has vastly more of one raw material over the other until i started looking at the numbers and an even further confused. I believe it is broken but if there is a logical explanation please explain this to me.

The current setup is an ice planet producing Viral Agents, that is

Microorganism->Bacteria------>Viral Agents
Planktic Colonies->Biomass-/

In general i try to distribute the number of heads on the extractors so that i receive a near equal amount of the two raw materials, simple enough, but when i check back on it the next day it has a huge stockpile of biomass and planktic colonies with no bacteria or microorganisms. I shift more heads to microorganism extraction and check the next day, even further off course. this continues until i have piles of one and both extractors pulling only microorganisms to get back on track. i believe i have found the cause to be that the survey amount is lying to me.

When i survey on the microorganism extractor it shows I will receive 224,404 total units in the 24 hour window, 9067 per hour with the first cycle yielding around 3000 units in 15 minutes(picture 1). I then install the setup and check the extractor which now shows i will receive 886 units in the first cycle(picture 2), rechecking the survey as its running shows im now getting 64,695 for the 24 hours, almost 1/4th the original number (picutre 3) and the incoming route to the launchpad where the material is being routed corroborates the diminished numbers(picture 4). what is going on here? is the survey number before i install completely meaningless? how do i balance this out to have even product?

http://i95.photobucket.com/albums/l154/Enderdragoon/PIBoggle.jpg
Scrapyard Bob
EVE University
Ivy League
#2 - 2011-12-20 06:50:08 UTC
Train Planetology / Adv Planetology and your heat maps and initial survey numbers will more closely resemble the actual numbers you get after installing the job. (Those 2 skills make the heat map more accurate.)

PL to 3 is not a long train, APL can be trained with PL4, so a lot of people go either:

3/0 - casual PI
4/3 - serious PI, a good middle-ground level
4/4 - super-serious PI, good enough for 95% of cases
5/4 - more time then sense PI folks
Professor Alphane
Les Corsaires Diable
#3 - 2011-12-20 07:17:12 UTC
I didn't know scanning skills effected surveys that way. You learn something new...

Also I would agree good scanning skills are invaluable at assesing planets. I note from your screenies you make 40 per head, finding good extraction sites should double maybe triple that figure even in high sec for a abundant item like MO.

[center]YOU MUST THINK FIRST....[/center] [center]"I sit with the broken angels clutching at straws and nursing our scars.." - Marillion [/center] [center]The wise man watches the rise and fall of fools from afar[/center]
Hundo Kay
Caldari Provisions
Caldari State
#4 - 2011-12-20 13:35:21 UTC
In regards to the heat map of resources, there are several things that you must know.

The first is that underlying the planet is an Absolute Resource Distribution that is the true amount of material available. This seems to be based on a base distribution, plus some random "nuggets" that increase a local area for a short time.

Now along comes our intrepid PI pilot and doing what he does, he scans the planet for "hot spots". The heat map generated is an Apparent Resource Distribution that based on his skills returns a map that may be close to the Absolute Resource Distribution, or may be "off" by several extractor head diameters. In my personal experience, the different level of scanning will return the same "pattern", but will differ in the exact locations of the hot spots.

So based on the Apparent Resource Distribution, our wanna be PI Magnate (the fancy name for business person, not the Amarr scanning frigate with an expanded cargobay for hauling PI) plops down some ECU's and extractor heads and starts moving them around. S/he gets a nice chart and a ton of data about what will be pulled in. Problem is that all these numbers are still based on the Apparent Resource Distribution, and not the Absolute Resource Distribution.

So finally happy with what is estimated, our PI pilot hits the "Install" button (but not the submit). Now our PI pilot can get two pieces of information to see how he is doing.

If he selects the ECU, s/he will get the true units pulled in the first cycle. If he hovers over the ECU he will get a % usage of the link to LP he is using as a buffer. With these two pieces of information and experience, he should be able to see how colse to the true hot spot he was.

If he is happy with the results, he can then hit submit and everything becomes "real". If he is not happy, he can hit cancel and try again, moving the extractor heads "off center" around the hot spot and hoping to get better results.

Another factor is that in HS the Hot Spots are more "pronounced". Like on a scale of 1 to 100, in HS the difference between a Hot Spot and Backgound, might be 25-5 meaning is you miss you might lose 80% of your potential yield. In Null/W-space it might be 80-50, meaning even if you miss you still get almost half the material.

Now even if you nail the hot spot, various things can influence your results.

1) You may have hit a nugget, and it will deplete faster then expected and you will pull less material then initially thought.
2) A nugget may appear after you set the extractors and you'll pull more then expected.
3) different materials deplete at different rates. So even if you perfectly match initial cycles between two materials, you will still end up with an imbalance at the end of a day.

So it will be difficult to balance. Based on my experience, the % usage on the link is the best gauge to the overall productivity of the cycle. If you have the same capacity on the two links between the ECU and LP, and were able to closely match the % usage, I would think you would be close at the end of the day. I have no proof to support this, other than several months of experience.
Hundo Kay
Caldari Provisions
Caldari State
#5 - 2011-12-20 13:39:55 UTC
Scrapyard Bob wrote:
Train Planetology / Adv Planetology and your heat maps and initial survey numbers will more closely resemble the actual numbers you get after installing the job. (Those 2 skills make the heat map more accurate.)

PL to 3 is not a long train, APL can be trained with PL4, so a lot of people go either:

3/0 - casual PI
4/3 - serious PI, a good middle-ground level
4/4 - super-serious PI, good enough for 95% of cases
5/4 - more time then sense PI folks


And at a whopping 4 extra days training, 4/4 over 4/3 is a small investment in time for a serious PI person who is planning to run 6 planets with max CC.
Dragoon Ree
Infinite Point
Pandemic Horde
#6 - 2011-12-21 07:22:33 UTC
Thank you for the wealth of information, I feel enlightened. I am starting my characters on training to 5/4 immediately, i had not realized the importance of these skills. With 5/4 will the apparent resource map be more accurate to the actual?
Tovil Hadras
Ordo Polaris
#7 - 2011-12-21 08:01:34 UTC
More people are having this problem and have already debunked the idea that it is skill related.

This is a bug.

https://forums.eveonline.com/default.aspx?g=posts&t=36505&find=unread
Scrapyard Bob
EVE University
Ivy League
#8 - 2011-12-21 15:22:57 UTC
Until we see a CCP dev response that speciically says "it's a bug" then I'm going to stick with "working as designed" and raising your PL/APL skill levels does make the heat map more accurate along with making the initial survey estimates more accurate.

Now, if you can prove that the "after" number for the total extracted units does not match up with what actually lands in a storage facility / launchpad - then I think you would have uncovered a bug.
Nyrak
Federal Navy Academy
Gallente Federation
#9 - 2011-12-21 17:30:39 UTC
I have noted two scenarios that I have not fully tested but offer here for debate.

First - it seems the extractor heads sometimes will not refresh between the ending of an extraction run and the start of another run if the heads were not moved. That is, the previous run (seems to) subtly influences what the heads would be reading when they remain in their previous position. Then when you submit the extraction order, the actual planetary numbers are pulled and your final amounts are much lower. I've casually noticed that if I had a "hot" run and started to run it dry, I'll note what the extractor head is reading - still quite high. I will move that extractor head to see if I can improve it, but can never obtain that high reading again no matter where I move the head. Then when I submit the run, the overall numbers are closer to what the moved head was reading.

Second - other players affecting the same site seem to influence the numbers too but by how much I do not know. Currently I have a situation where I found a really good hot spot of a raw material and set-up shop. I checked out the area for any Command Centers and found nothing nearby. Lately I have been doing two day runs for a decent balance. The first run produced more than what was quoted. The second run produced more that what was quoted, though not as much as the first run. Repeat this for the third and fourth runs. The fifth run was quoted but produced a bit more than half of the quote. This was when I noticed I had a ~neighbor~ and saw approximately where the extractor heads would be located - on top of mine, based on where his extractor unit was located. While I understand each run will pull less and less overall, the ratio went from gradual to distinct once I had a ~neighbor~.

Again, I do not have numbers but just observations.
Ninyania alCladdyth
McLuvin AstroDynamics
#10 - 2011-12-21 19:05:04 UTC
Nyrak wrote:
... the ratio went from gradual to distinct once I had a ~neighbor~


hmm... potential use of all the janitors, militants, and exotic dancers we've been collecting over the years? Give 'em a couple crates of Small Arms and send 'em on a ground assault.
Invictra Atreides
Toward the Terra
#11 - 2011-12-21 20:03:49 UTC  |  Edited by: Invictra Atreides
This solves the problem: Link

Blog | Tutorials | Youtube "I don’t know everything, I just know what I know."
Professor Alphane
Les Corsaires Diable
#12 - 2011-12-22 03:22:41 UTC
Invictra Atreides wrote:
This solves the problem: Link


Wow thats cool.

Is the number of planetary 'nodes' consatant, because if so are bigger or smaller planets more susceptable to these 'discrepancues'?

[center]YOU MUST THINK FIRST....[/center] [center]"I sit with the broken angels clutching at straws and nursing our scars.." - Marillion [/center] [center]The wise man watches the rise and fall of fools from afar[/center]
Conrad Lionhart
#13 - 2011-12-22 09:35:27 UTC
I have level 5 in planetology and advanced planetology, and my PI extractor readings have become accurate. The values before and after clicking the submit button are the same.

So it is skill related.