These forums have been archived and are now read-only.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

 
5 Pages123Next pageLast page
 

Tips for Project Discovery exoplanets?

Author
Caldari State
#1 - 2017-07-11 15:05:26 UTC
Hi everyone!

Has anyone some tricks and tips to locate the correct answer in Project Discovery? I really suck on it

Thanks
#2 - 2017-07-11 15:15:31 UTC  |  Edited by: Vanessa Celtis
Naril Mikjail wrote:
Hi everyone!

Has anyone some tricks and tips to locate the correct answer in Project Discovery? I really suck on it

Thanks


Same for me, sorry no tip, I tried for several hours and I found zero transition.

The samples provided do not have enough granularity to find any transition, it's un-playable and the user-interface sucks. There should be a way to zoom-in but having more granularity so you can actually see any existing transition pattern.

Fail fit!
#3 - 2017-07-11 15:33:01 UTC
Vanessa Celtis wrote:
Naril Mikjail wrote:
Hi everyone!

Has anyone some tricks and tips to locate the correct answer in Project Discovery? I really suck on it

Thanks


Same for me, sorry no tip, I tried for several hours and I found zero transition.

The samples provided do not have enough granularity to find any transition, it's un-playable and the user-interface sucks. There should be a way to zoom-in but having more granularity so you can actually see any existing transition pattern.

Fail fit!


You can zoom in. Maybe you should do the tutorial again.
Caldari State
#4 - 2017-07-11 15:34:40 UTC
All samples have too much noise and even zooming I can't see anything
#5 - 2017-07-11 15:42:47 UTC  |  Edited by: Vanessa Celtis
Sp3ktr3 wrote:
Vanessa Celtis wrote:
Naril Mikjail wrote:
Hi everyone!

Has anyone some tricks and tips to locate the correct answer in Project Discovery? I really suck on it

Thanks


Same for me, sorry no tip, I tried for several hours and I found zero transition.

The samples provided do not have enough granularity to find any transition, it's un-playable and the user-interface sucks. There should be a way to zoom-in but having more granularity so you can actually see any existing transition pattern.

Fail fit!


You can zoom in. Maybe you should do the tutorial again.



You can zoom in yes but when you zoom-in, the displayed details you see do not provide enough granularity to actually "with a human eye" identify any planet transits. It's a loss of time to run it with the currently provided samples at this stage. In fact, in a real case scenario, labs have special analysis software to actually identify the transits, this here is a joke.

BTW, the tutorial presents very obvious samples, and once you start with the real data the provided samples are way too noisy.

I do not understand why CCP is using such an important project like planets discovery as a spring board for their game with a big splash pretending its for science, then throws this junk on us and hope we take it.
Gallente Federation
#6 - 2017-07-11 16:16:04 UTC
The type of data analysis that this requires seems a lot more suitable for a computer algorithm than for a human.
Gallente Federation
#7 - 2017-07-11 17:27:02 UTC  |  Edited by: Edik Edik
This is very hard, example of one of my analysis: http://imgur.com/a/iSZBJ
Even after they gave me the result I still cannot get itOops
#8 - 2017-07-11 17:37:15 UTC  |  Edited by: Vanessa Celtis
Edik Edik wrote:
This is very hard, example of one of my analysis: http://imgur.com/a/iSZBJ
Even after they gave me the result I still cannot get itOops


I think CCP has reversed the hard mode with the easy mode by mistake; all of a sudden easier samples come in on my side, I don't know what happened, but I suppose CCP is looking at the logs and tuning things in real-time on the servers and fixes initial issues.
#9 - 2017-07-11 17:51:50 UTC
https://imgur.com/a/uh4hP
https://imgur.com/a/AZoR8

The images are the same correction viewed under folded and not folded to a failed attempt I made....

How are we supposed to pick out what appears to be completely random squiggles as a transit? So far nothing makes apparent sense in this system... Pick out the dips and it tells you that youre wrong, only to show you squiggles that dont even slightly resemble eachother or even resemble dips at all...
Phalanx Federation
#10 - 2017-07-11 18:19:37 UTC
The way this system is set up they make you feel all good about yourself with the tutorial and you are rocking out and getting them all right... And then you get to the actual part where you look at real data... And they make you feel like you were born yesterday... Get your confidence up and then smash you with data that looks completely unreadable. I have not gotten a single one right and less than 1 hour after I started the project discovery I am not even getting rewards anymore.
#11 - 2017-07-11 18:39:30 UTC
Its more easy use the KEPLER instead
Haighare Pirates
#12 - 2017-07-11 18:50:20 UTC
Citoplasm is the powerhouse of the exoplanet!

Works every time for me.
Goonswarm Federation
#13 - 2017-07-11 20:07:00 UTC  |  Edited by: M4cD0g
Sp3ktr3 wrote:
Vanessa Celtis wrote:
Naril Mikjail wrote:
Hi everyone!

Has anyone some tricks and tips to locate the correct answer in Project Discovery? I really suck on it

Thanks


Same for me, sorry no tip, I tried for several hours and I found zero transition.

The samples provided do not have enough granularity to find any transition, it's un-playable and the user-interface sucks. There should be a way to zoom-in but having more granularity so you can actually see any existing transition pattern.

Fail fit!


You can zoom in. Maybe you should do the tutorial again.


You mean like this:
Detrend 1h: https://imgur.com/gXoXz5P
Detrend 10h: https://i.imgur.com/aVz5yJH.png
No Detrend: https://i.imgur.com/bRjfydS.png

The picture i linked is supposed to be a transition ZOOMED IN! Sure, buddy, zoom in works great!

The system is complete bollocks. There is no pattern that we can see. Either they screwed up the data and solutions or they do not show the real data. It's broken.
Synergy of Steel
#14 - 2017-07-11 20:24:15 UTC
#15 - 2017-07-11 20:44:40 UTC  |  Edited by: NanoSpirit
This mini-game is unfun, frustrating and you dont learn anything between different sample... so difficult to improve...

WHen I look at the answer provided when you fail, and you zoom in, you are like.. wtf is that...

- Unclear
- Inconsistent between sample
- UI and presentation make it worse...

There is a limit between getting real life and video game gameplay.. I am not here to become a scientist
#16 - 2017-07-11 21:18:20 UTC
It seems like there are some false positives as well as completely missed transits in some of the evaluation samples. I have to wonder if they had grad students working on these, and how drunk they were at the time.
#17 - 2017-07-11 21:27:57 UTC
This Project Discovery has to be a joke. CCP are trolling us.

How the hell do the expect somebody to notice this, apparently the correct selections was just random selections that don't stand out from the background in any shape or form.

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

If it was the odd one I'd understand, but it seems to be 99% of samples that do this. No matter how you analyse any obvious blips in signal it always fails you and shows you something that nobody would ever notice. I've had more luck just hitting the 'No Transition' button on every sample.

Hello darkness, my old friend.

#18 - 2017-07-11 21:54:35 UTC  |  Edited by: DrysonBennington
Project Tutorial

The reason why the Project is difficult is because there was a mix up in the plates that you were given. They put the harder plates at the beginning.


I was at about 4% accuracy when I started getting easier plates to study. So keep at it, it does get easier.

Here are a few screenshots of successful transit discoveries, Consensus Plates and Missed Transits.


In the first image you will see the typical "noise" of the surface of the sun that is detected by the space telescopes.

On the left and right are the transits of a planetary body that look like fangs or ice sickles.

When you zoom in you can see the depth of the transit, or how much light is being blocked by the object, which in this the object is blocking 2.599 % of the suns light.

You subtract the amount of the dip from 100 where 100 is the surface of the sun. 100 is the output of light from the sun would be normal.

Normal means nothing blocking the light from the sun or solar flares causing an increase in the light curve of the sun.

Think of it this way. You have $100 and you add $5 now you have $105 or 105% the normal light curve of the sun. Once again you have $100 and purchase a Beefy Big Boy from the lunch room fun machine that costs $2.59....Hey come here a little closer, just to let you know I like to put my Beefy Big Boy in between the sandwich buns and then...Beep...Beep....Anuirison!..Sorry Bryson just showing them the Fun Machine, I mean vending machin...Back to the tutorial. Now you have $97.41 left which equals 97.41% of the sun light that is showing.

For those more in tune with planet hunting the readout would be .0259/.9741.

Click in the center of either transit and colored line will appear. Because there is only one transit on this plate simply drag your line to the second transit and then click Fold.

You will be able to tell that your transits match up as the further you drag your transit line to the second transit the other transit markers disappear.

So you probably guessed it by now that the more lines on your screen means that all of the lines have to match up with an ice sickle or fang in order to be considered a transit.


Along the bottom is your transit length and total transit time. In this case the transit on the left starts out at Epoch 4.235(day of the week when transit started) and takes 17.049 days to complete its orbit at Epoch 21.25.

An orbital period is the amount of time it takes an object to orbit another object. In this case the orbital period for Earth would be ~365 days.

Intro 1

https://www.mediafire.com/convkey/c121/acznjbcs73lj63d6g.jpg

Intro 2

https://www.mediafire.com/convkey/711e/89c1ps64zub6zgm6g.jpg

I will do more of the tutorial tomorrow.
Circle-Of-Two
#19 - 2017-07-11 21:57:22 UTC
I have no idea why this failed ...

https://imgur.com/gallery/DPaOb

#20 - 2017-07-11 22:05:37 UTC  |  Edited by: Vanessa Celtis
Valera Schwert wrote:
I have no idea why this failed ...

https://imgur.com/gallery/DPaOb




Because the red was a SECOND pattern / planet; you need to double click when validating to have in this case *two* patterns selected and then only submit.
5 Pages123Next pageLast page
Forum Jump