These forums have been archived and are now read-only.

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

EVE General Discussion

 
  • Topic is locked indefinitely.
12Next page
 

Project Discovery Analysis Issues

Author
Mike Denard
Stay Frosty.
A Band Apart.
#1 - 2017-07-14 04:22:54 UTC
I'm getting pretty frustrated with how these mechanics work. There's too many areas where random static that doesnt have a discernible pattern or actually doesn't match at all is listed as correct, or areas where identifying a planet is still an analysis failure because it doesnt match up perfectly with what the system has decided is right.

I'd be fine with this if there was some explanation for why some answers are wrong, or how the hell they're identifying planets in some of these areas that dont match up with the in game explanation for exoplanet identification.

Here's one result I've gotten that's pretty mild.
Screenshot
Baboo Yagu
#2 - 2017-07-14 04:25:45 UTC
I'll see your screwed up analysis and raise you this...

https://i.imgur.com/pAHYS3m.png

Hello darkness, my old friend.

Mike Denard
Stay Frosty.
A Band Apart.
#3 - 2017-07-14 06:17:55 UTC
They really need to take a closer look at what they're using to analyze players. This doesnt even make sense.

http://imgur.com/a/BUkhJ
Ghazbaran
Sulima
Literally Triggered
#4 - 2017-07-14 14:57:12 UTC  |  Edited by: Ghazbaran
Algorithm is off for recognizing data points.

I believe we need data tables to be able to better understand changes because we cannot always recognize something by how it looks visually and the vague approximations we get with the graph. Seeing the specific number on the data point would be highly beneficial.

http://imgur.com/LLdzTUB


Made a Channel to discuss issues with Exoplanets and trade Items obtained from project discovery.

Join -> New Eden Planetary Society
Photrius Pyrelius
University of Caille
Gallente Federation
#5 - 2017-07-14 15:08:50 UTC
Ghazbaran wrote:
Algorithm is off for recognizing data points.

I believe we need data tables to be able to better understand changes because we cannot always recognize something by how it looks visually and the vague approximations we get with the graph. Seeing the specific number on the data point would be highly beneficial.

http://imgur.com/LLdzTUB


Made a Channel to discuss issues with Exoplanets and trade Items obtained from project discovery.

Join -> New Eden Planetary Society



The reason it's marked wrong is almost certainly due to it being a two-planet system. I'm not sure how we're supposed to make such a distinction, though. You can't even tell by looking at the depth of every other trough in that picture. Funny story, I found one where every even dip was much shallower than every odd dip, so I marked it as a two-body transit only to find out I was wrong.
Ghazbaran
Sulima
Literally Triggered
#6 - 2017-07-14 15:11:29 UTC
Photrius Pyrelius wrote:
Ghazbaran wrote:
Algorithm is off for recognizing data points.

I believe we need data tables to be able to better understand changes because we cannot always recognize something by how it looks visually and the vague approximations we get with the graph. Seeing the specific number on the data point would be highly beneficial.

http://imgur.com/LLdzTUB


Made a Channel to discuss issues with Exoplanets and trade Items obtained from project discovery.

Join -> New Eden Planetary Society



The reason it's marked wrong is almost certainly due to it being a two-planet system. I'm not sure how we're supposed to make such a distinction, though. You can't even tell by looking at the depth of every other trough in that picture. Funny story, I found one where every even dip was much shallower than every odd dip, so I marked it as a two-body transit only to find out I was wrong.


Two system or not, they want transits. Also, look at the data points. almost identical and in the tutorial it clearly states that similar data is probably the same planetary orbit which was my reasoning behind it being 1. When folded, it folds perfectly into each other because their data points are almost identical
Photrius Pyrelius
University of Caille
Gallente Federation
#7 - 2017-07-14 15:14:18 UTC
Ghazbaran wrote:

Two system or not, they want transits. Also, look at the data points. almost identical and in the tutorial it clearly states that similar data is probably the same planetary orbit which was my reasoning behind it being 1. When folded, it folds perfectly into each other because their data points are almost identical



You'll get no defensive argument from me. Wouldn't being perfectly lined up like that mean that bodies are in synchronous opposing orbits around the star? What are the odds on that?!
Cypherous
Liberty Rogues
Aprilon Dynasty
#8 - 2017-07-14 15:37:54 UTC
Ghazbaran
Sulima
Literally Triggered
#9 - 2017-07-14 15:43:11 UTC
So whats your take on this?

http://imgur.com/a/Ssezo
Ghazbaran
Sulima
Literally Triggered
#10 - 2017-07-14 15:51:14 UTC
Cypherous wrote:
https://cdn.discordapp.com/attachments/319140823344087042/335444488392671232/unknown.png

https://cdn.discordapp.com/attachments/124557015736385536/335443190620356608/unknown.png

System is borked


YUP
Photrius Pyrelius
University of Caille
Gallente Federation
#11 - 2017-07-14 15:51:56 UTC
Ghazbaran wrote:
So whats your take on this?

http://imgur.com/a/Ssezo



Looks as indecipherable as many of the ones I got wrong yesterday. =-\ I really don't know what's wrong with the system, but honestly we should all know better than to think something's going to be perfect (playable?) on launch-day. And that's no dig against CCP; the whole industry launches stuff prematurely just to get it out the door, patch it up later.
Blade Darth
Room for Improvement
Good Sax
#12 - 2017-07-14 19:15:49 UTC
Baboo Yagu wrote:
I'll see your screwed up analysis and raise you this...

https://i.imgur.com/pAHYS3m.png

I had this evaluation sample, what you marked is a sudden jump in brightness, that, after de-something-cation looks like a "Z" instead of a "V".
Ghazbaran
Sulima
Literally Triggered
#13 - 2017-07-14 21:33:48 UTC
Loame en Welle
Federal Navy Academy
Gallente Federation
#14 - 2017-07-15 01:16:00 UTC  |  Edited by: Loame en Welle
I bought into the hype... unfortunately this is not analysis.

EDIT: A more constructive comment would be that I also have noticed significant discrepancies in the analysis and that without an understanding of modeling for each solar activity classification and its respective 'dip rates', it will be very difficult to visually assess significance.
Bones Prefect
The Scope
Gallente Federation
#15 - 2017-07-15 02:36:44 UTC
Ghazbaran wrote:



Dont ask me... https://imgur.com/a/Eqfox
Mike Denard
Stay Frosty.
A Band Apart.
#16 - 2017-07-15 03:30:25 UTC
Bones Prefect wrote:
Ghazbaran wrote:



Dont ask me... https://imgur.com/a/Eqfox


Same thing again here. The analysis is wrong a lot of the time. Not us.
http://imgur.com/a/9v2zt
Cypherous
Liberty Rogues
Aprilon Dynasty
#17 - 2017-07-15 20:59:26 UTC
https://cdn.discordapp.com/attachments/264781482348773376/335887428885938198/unknown.png

So are we going to get a blue post about this broke ass system? :P
Joseph Bramah
In Dock We Trust
#18 - 2017-07-15 23:33:06 UTC
wrong example sample:

http://i.imgur.com/CRJ3xSO.png
Blade Darth
Room for Improvement
Good Sax
#19 - 2017-07-16 00:44:40 UTC
Joseph Bramah wrote:
wrong example sample:

http://i.imgur.com/CRJ3xSO.png
that one is quite obvious, look at the timeline, the little dips look all the same
Galaxxis
Unicorn Rampage
#20 - 2017-07-16 00:52:33 UTC
http://imgur.com/a/7EDpd

This is getting irritating.
12Next page