Feasibility Studies Mechanics

Research and Development in SotS2.
qjachym
Posts: 34
Joined: Thu Jul 26, 2012 1:47 pm

Re: Feasibility Studies Mechanics

Post by qjachym » Sat Sep 07, 2013 6:41 pm

Ok.
I have done some tests.
If you change TechFeasibilityDeviation to 0.0 the game crashes at first feasibility, if you change it to 0.1 the feasibilities show percentage below 5%, but if you change it to 1.0 the feasibilities show various values. I even on purpose failed (to 200%) Heavy Plasma Cannons (feasibility 28%).
It appears that this variable shows exact feasibility if you set it to 1.0.

I take back everything I wrote in posts before... unless all the feasibilities above 50% will be successful :)

Azrael Ultima
Posts: 3051
Joined: Thu Jan 27, 2011 3:20 am

Re: Feasibility Studies Mechanics

Post by Azrael Ultima » Sat Sep 07, 2013 8:44 pm

No, setting it to 1 just gives you random values in a larger range, but they are still random.

If any setting gave you "exact" feasibilities, everything would show up as either 1% or 99%, depending on what the game rolled for that session.
I've got a lovely bunch of coconuts.
The zeppelin of bluster Feldman excoriated Freddy with suddenly popped into a cloud of humility. (David Grand, The Disappearing Body, 2002)

User avatar
BlueTemplar
Posts: 3131
Joined: Wed Aug 29, 2007 9:15 am

Re: Feasibility Studies Mechanics

Post by BlueTemplar » Sat Sep 07, 2013 8:55 pm

Do these various values correspond to the tech chances here : ?
viewtopic.php?f=42&t=49879

qjachym
Posts: 34
Joined: Thu Jul 26, 2012 1:47 pm

Re: Feasibility Studies Mechanics

Post by qjachym » Sun Sep 08, 2013 5:10 am

I tested it for Suulka:
Heavy Plasma cannon - 28% (basic 65%, failed at 200%)
Deflector Shields - 73% (basic 45%)
Disruptor Shields - 1% (basic 20%)
Expert Systems - 48% ( basic 65%, failed at 200%) - as expected
Telekinesis - 1% (basic 65%)
Mecha Empathy - 13% (basic 15%)
UV beamer - 51% (basic 45%, done at 100%) - as expected
X-Ray beamer - 1% (basic 45%)

I will make furter tests, but It appears that techs above 50% are succesfull and below 50% are not.

User avatar
Karu
Posts: 420
Joined: Sun Apr 21, 2013 11:49 am

Re: Feasibility Studies Mechanics

Post by Karu » Sun Sep 08, 2013 11:24 am

qjachym wrote:I will make furter tests, but It appears that techs above 50% are succesfull and below 50% are not.


Untrue.
A Friend of mine which I play MP with loves hiver. The only thing he loves more then Hiver are Hiver with shields.
He allways have a go with them if they are over 10% feasability.

He had them failed at 60% feasibility, and also got them at 20%.
The only "easy" way to see if you got a tech is look at the savegame. Feasibility is only an indicator - not a true fact generator.
If you reach out more you will find - in rare cases - people that got techs from 10% feasibility and also failed techs from feasibility over 90% (95% chance of MK I shields as liir failed for me ... and yes feasibility was 95%).


Don't construct game mechanics. Accept them as what they are.


Sincerly

qjachym
Posts: 34
Joined: Thu Jul 26, 2012 1:47 pm

Re: Feasibility Studies Mechanics

Post by qjachym » Sun Sep 08, 2013 1:03 pm

Karu:

Are you reading all of the conversation or only last post?
The data I shoed are AFTER changing TechFeasibilityDeviation to 1.0.
It is not the Vanilla version of the game.

ScoSteSal118
Posts: 685
Joined: Fri Mar 09, 2012 7:58 pm

Re: Feasibility Studies Mechanics

Post by ScoSteSal118 » Sun Sep 08, 2013 1:15 pm

That test incorporates so little data and in general is constructed so..."something"-ishly (eg not testing the same tech in different games with the modded TechFeasibilityDeviation value) that its results are imo pretty much anecdotal and provide little support for general conclusions.

EDIT: Oh, and I should have noted that it contradicted clear previous moderator posts on the topic.
Last edited by ScoSteSal118 on Sun Sep 08, 2013 7:00 pm, edited 1 time in total.

qjachym
Posts: 34
Joined: Thu Jul 26, 2012 1:47 pm

Re: Feasibility Studies Mechanics

Post by qjachym » Sun Sep 08, 2013 3:38 pm

Do not misunderstand me. I have just found a hint that changing this value may show exact feasibility without any modifications. I encoureage you to test is by yourself, because I may be wrong. In my opinion now if you roll below 50% you will not get tge tech , and if you roll above 50% you will get it. This information is enough for me.

User avatar
Mecron
Kerberos
Posts: 38604
Joined: Fri Aug 05, 2005 7:26 pm

Re: Feasibility Studies Mechanics

Post by Mecron » Sun Sep 08, 2013 5:37 pm

And it is incorrect but if it makes you feel better, then it's all good ;)

qjachym
Posts: 34
Joined: Thu Jul 26, 2012 1:47 pm

Re: Feasibility Studies Mechanics

Post by qjachym » Mon Sep 09, 2013 6:24 am

Mecron wrote:And it is incorrect but if it makes you feel better, then it's all good ;)


Yes, I think I have reached tranquility at last :)

What you said made me thinking. You didn't say that I am entirely wrong.
The fact is that feasibility deviation has standard value 0.5, if you change it to 0.0 the game crashes so there should be a factor that gives exact feasibility value (my guess is 1.0)
I think that I am incorrect with the "below 50% is fail, aboce 50% is success" (And I think you meant only this that is incorrect)

If I take under consideration the facts:
- game rolls at the begining if you have the tech or not
- there are basic values for techs
- standard deviation factor is 50%

These formulas come to my mind:
Sol force has 90% for VRF Systems so we have 10% that the tech is not available +-50% (deviation) = 60% - If feasibility roll is above 60% that means that the tech is 100% available.
Suulka has 65% for Telekinesis so we have 35% that the tech is not available +-50% (deviation) = 85% - If feasibility roll is above 85% that means that the tech is 100% available.
On the other hand there can be a roll 51%-50% = 1% that can still come out successful (because the game rolls that the tech is available but feasibility roll was very low)

If we find the right factor to reduce the deviation to 0% the examples above would be interpreted easly:
Feasibility for VRF Systems for Sol force must be above 10%.
Feasibility for Telekinesis for Suulka must be above 35%.
Feasibility for a Tech with 20% basic chanse must roll above 80%.

For me 50% deviation is to much. I personally would like to reduce it to 10-20%.

shadowclasper
Posts: 20
Joined: Fri Oct 25, 2013 12:43 am

Re: Feasibility Studies Mechanics

Post by shadowclasper » Tue Oct 29, 2013 8:44 pm

Oookay... I'm still confused reading this...

As I understand it, there are TWO random rolls happening per tech?

The first roll is at the beginning of the game, the chances of it rolling are what we PUBLICALLY know, like Morrigi having a 95% feasibility for drone tech for example. That's the MAXIMUM feasibility I'm ever going to get with that tech. But at the beginning of the game it could be set to as low as 1%, or as high as 95%, or anywhere between those two ranges correct?

Then when I actually get to research that tech, I do a feasibility study, which finds out in a ball park range just how likely it is from the REAL feasibility value rolled for my civilization at the beginning of the game. Thus, I can find out, within a certain range, how likely it is for me to research the technology successfully.

Next, when I actually begin the research, the number of turns to research it is set by the REAL feasibility of the technology. Low 'real' feasibility means more turns to research it.

Finally, once the research comes to the end of the set number of turns, it rolls the dice AGAIN. A d% as it were. If I roll at or under the 'real' feasibility percentage set by the system at the beginning of the game, I get the tech, otherwise I fail to get it. Thus my technology could be at 95% feasibility, but my d% at this stage rolled a 96. No go.

So, in reality, I'm rolling TWICE for each technology?

User avatar
Karu
Posts: 420
Joined: Sun Apr 21, 2013 11:49 am

Re: Feasibility Studies Mechanics

Post by Karu » Tue Oct 29, 2013 9:32 pm

@shadowclasper: no

The "race tech %" is an average of how often you get the tech. 95% means 95 of 100 games.
It is rolled per "techlink" - if a tech has 2 links each chance get rolled seperatly.
And the techroll is a "Yes" or "no" - nothing in beetween. And yes it get's rolled for Turn 1.

The "shown" feasability if you tech is an estimate of your researchers. Or a wild guess. The higher the better, but there were techs that failed at 99% feasability and techs that were 1% got through. (Shown in techtree, not to confuse with the wiki %)

Your research speed has absolutly nothing to do with with feasability.
Every research costs a different amount of research points - generatated though your tech boost or through taxes.
Modified only by you % techspeed modifier. which means more research points per credit.


Sincerly

Azrael Ultima
Posts: 3051
Joined: Thu Jan 27, 2011 3:20 am

Re: Feasibility Studies Mechanics

Post by Azrael Ultima » Tue Oct 29, 2013 9:46 pm

You only roll twice for a technology when it has two links, as each link is rolled independently.

Feasibility does not play into research success and is only determined when success or failure are already set in stone.
I've got a lovely bunch of coconuts.
The zeppelin of bluster Feldman excoriated Freddy with suddenly popped into a cloud of humility. (David Grand, The Disappearing Body, 2002)

qjachym
Posts: 34
Joined: Thu Jul 26, 2012 1:47 pm

Re: Feasibility Studies Mechanics

Post by qjachym » Wed Oct 30, 2013 9:12 am

shadowclasper:

(Keep in mind that it is not confirmed)

You roll for a tech at 1st turn:
If Morrgi have 95% for drones the game rolls 1-100 and if the roll 1-5 you don't have this the but if the game rolls 6-100 you have it - the roll is hidden and you will never know what was the value.
Let's say the game rolled 36 - you will have it but if you make the feasibility roll (vanilla feasibility deviation is 50%) the range is 36-50 to 36+50 so the feasibility rolls can be 1-86%.

If the game roll (at 1st turn) is unsuccesful but the tech has 2 links you can reroll the tech and have a second chance to obtain it - I think that the game rolls for each link - not tech itself.

If you reduce the feasibility deviation to 10% (from 50%) the feasibility would be 36-10 to 36+10 = 26%-46% that would give you much precise information if you have it or not.

User avatar
Mecron
Kerberos
Posts: 38604
Joined: Fri Aug 05, 2005 7:26 pm

Re: Feasibility Studies Mechanics

Post by Mecron » Wed Oct 30, 2013 12:18 pm

It's not +\- 50%

Post Reply

Return to “The Tech”

Who is online

Users browsing this forum: No registered users and 5 guests