These forums have been archived and are now read-only.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

Errors in Project Discovery
Author
Previous Topic Next Topic
Ice Astrid
Republic Military School
Minmatar Republic
#1 - 2017-07-23 17:54:45 UTC  |  Edited by: Ice Astrid
I've been, for most of this month, religiously going through steaming stacks of the transits from Project Discovery. And I mean, I've managed to incorporate Discovery in most of everything else I do to a whole new degree of multitasking that has never been seen before.

Currently, I'm at level 46 analyst, which means I've roughly completed 1600 transits.
In all that, I've found some very... let's say, "unbelievable" transits. This I say, knowing that there are almost-impossible-to-see variations, anomalies in the luminosity of all sorts, and occasional oddities.
When I say unbelievable, I mean it flat BS - and since we are apparently living in 2017 and instant-screen capture is a thing, I have a few examples saved to show.

WTF
Why?!
There is no God
Is that so?
Why would you do this to me?
Most of these transits are recurrent, which means I've seen far more than once (and I still can't spot immediately).


The point of all this?
To ask: Why? WHY does this happen?! Is it meant, MEANT to be this way? If it's a bug, it has to be squashed, if it's not, then I need to know why it keeps happening.
Altalicious
Burning Space
#2 - 2017-07-24 15:02:15 UTC
On a couple of those you may be finding multiple planets. Hence the fail analysis. I have had the same problem with some of the analyses too and am scratching my head on why I am getting a failure message.
James Zimmer
Republic University
Minmatar Republic
#3 - 2017-07-24 22:06:58 UTC
Altalicious wrote:
On a couple of those you may be finding multiple planets. Hence the fail analysis. I have had the same problem with some of the analyses too and am scratching my head on why I am getting a failure message.


No, those are at perfectly-spaced intervals, indicating that the planet is a certain, single distance from the star. The only other possibility is that the planets are at the same distance and aligned exactly opposite from each other in their orbits. From a physics standpoint that seems extremely unlikely, maybe impossible.

Another issue I've run into is the phantom transits. I swear, there is not an appreciable change in luminosity with quite a few of these transits. I suspect the scientists that found those had some different tools they used to find them, which makes me question the value of the information that we're giving.
Ice Astrid
Republic Military School
Minmatar Republic
#4 - 2017-07-25 01:42:11 UTC  |  Edited by: Ice Astrid
I'm all too used to the "phantom transits", though I've attributed it to my unability to correctly understand completely the nuance of reading the luminosity scale.

The strange part of this, I believe, is that these are set as "approved transits".
With this I mean, at certain point you stop getting the binary correct-or-failed result, and start participating in a consensus to determine whether there's actually transits in certain scales. These aren't those "open" consensus, but their result is already defined.

This means that at some point, somebody, or some program, looked at the numbers and said "yeah, this is okay", when it doesn't seem to be the case.

Maybe I'm just thinking too hard about it.

EDIT:

Alternatively, the idea above also occurred to me. What if we are getting different captures of the same scans? - Or different scans of the same rotations. Different points of view of the same thing. It would explain a few things.
Fish Hunter
Center for Advanced Studies
Gallente Federation
#5 - 2017-08-01 22:28:26 UTC
James Zimmer wrote:
Altalicious wrote:
On a couple of those you may be finding multiple planets. Hence the fail analysis. I have had the same problem with some of the analyses too and am scratching my head on why I am getting a failure message.


No, those are at perfectly-spaced intervals, indicating that the planet is a certain, single distance from the star. The only other possibility is that the planets are at the same distance and aligned exactly opposite from each other in their orbits. From a physics standpoint that seems extremely unlikely, maybe impossible.

Another issue I've run into is the phantom transits. I swear, there is not an appreciable change in luminosity with quite a few of these transits. I suspect the scientists that found those had some different tools they used to find them, which makes me question the value of the information that we're giving.


These are the ones i hate. Mark what could be a transit and it comes back with no this white noise section here is one. I stare at it for a while and there's no way we could tell that with the data we're given.
Fish Hunter
Center for Advanced Studies
Gallente Federation
#6 - 2017-08-01 22:30:43 UTC
Ice Astrid wrote:
I'm all too used to the "phantom transits", though I've attributed it to my unability to correctly understand completely the nuance of reading the luminosity scale.

The strange part of this, I believe, is that these are set as "approved transits".
With this I mean, at certain point you stop getting the binary correct-or-failed result, and start participating in a consensus to determine whether there's actually transits in certain scales. These aren't those "open" consensus, but their result is already defined.

This means that at some point, somebody, or some program, looked at the numbers and said "yeah, this is okay", when it doesn't seem to be the case.

Maybe I'm just thinking too hard about it.

EDIT:

Alternatively, the idea above also occurred to me. What if we are getting different captures of the same scans? - Or different scans of the same rotations. Different points of view of the same thing. It would explain a few things.


We're seeing a mix of already analyzed data and new data. The analyzed data is for determining your accuracy.