Research Cost Bug?

I am seeing this now, in 0.41.2.  I micromanage by 0.1 rp every turn.  Within a tech specialization (3-choice bubble), all 3 choices have the same cost, right?  But as I finish each one, my turns-to-complete for the other 2 goes up from 1 turn with 0.1 rp waste to 2 turns (which really means about 1 turn + 10 rp).  It doesn't annoy me because I'll micromanage it down to 1.0 turn anyways.

Conjecture: Every completed tech level adds a penalty to all research costs.  Which would actually punish breadth and reward beelining.  (Ideology trees work the same way in 0.41.2, so maybe they tweaked the wrong XML file.)

This tech cost increase must be slightly greater than growth, since both of us are presumably growing pop at the normal rate, yet we're still falling behind.  OTOH, my growth rate is tiny in % terms: techworld 15.0 bp +0.18/turn ~= 1.2%.  So a tech cost increase of 2% would suffice.

In one of the streams, Paul mentioned that the specialization techs have cost increases so they are always like researching the most advanced tech in that branch. I think that the idea is to encourage players to pick one of the three and to not go back at the end for cheap bonuses.

GalCiv 2 also had tech-cost inflation, so I doubt, that this is a bug.

Okay I can see if you having research inflation inside of one of the tech specializations as peregrine mentioned, that kind of makes sense.  But the regular techs (non specialized ones) still don't make sense even if I count for inflation.  Which if inflation is in there discourages me from wanting to ever do a research type victory.

Example:

• Xeno Entertainment costs over time:
• 29 RPT (research per turn) x 8 Turns to Research (TTR) = 232
• 34 RPT x 7 TTR = 238
• 39 RPT x 7 TTR = 273
• 51 RPT x 6 TTR = 306
• 52 RPT X 5 TTR = 260 ???? huge outlier on the inflation theory
• 64 RPT x 5 TTR = 320
• 69 RPT x 5 TTR = 345

That's over 38 Turns so from the min to the mix we are talking 2.97RPT inflation per turn for just this tech.  Seems a little high to me but the 260 one tells me there is something wrong somewhere or the techs are cross related to other tech trees that's going to be hell of hard to figure out.

how much micro are you doing to achieve these numbers it could be that your wasting most of a turn of research and then when you hit 52rpt it reaches very little waste

It is working according to design, the techs get more expensive as time goes on. Please keep in mind that the game is in early development and those values are subject to change. Thank you for the feedback.

I dislike this mechanic immensley and feel it does not make any sense whatsover. 

I appreciate the feedback from the devs, just seemed like it was super quick for the inflation on the research which is why I though I should point it out in case it was a bug.  I understand it's a beta, hence why I'll continue to mess around with stuff and try and see if I can break things.

Here's research cost creep in action.  Before surveying an Artifact, I always switch my current tech to a brand-new tech at the highest level I'm eligible to begin (and then switch it back).  Twice, I won the little lotto of "25% of your current tech project".  I haven't touched either tech since.

The grant amount stayed the same, it's the bars that got wider

Neat mechanism.  It's another way to punish (discourage) beelining down one path, since at some point you must go back and finish several low-tier techs in other areas.  If the cost of all of those low techs has doubled or more by the time you get to them, that could be less efficient compared to a breadth-first approach.

We don't know the parameters of the tech cost increase mechanism.

• Trigger: per tech level completed?  per rp spent?
• Amount: fixed constant?  percentage of cost?

We can rule out "per turn" as a trigger, since you can nigh-reliably predict tech completion time for your current tech level.

actually i think its quite the opposite this mechanism encourages beelining for only the core techs you absolutely need to win

Well, open competition vs. players just as skilled as you will already encourage that.

But within your subset of core techs ... take them breadth-first! 

e.g. I've taken 39 techs without W3 Weapons Specialization, and its cost is now >= 6*46 + 1 >= 277(!).  I'm going to get that cost up to 400+ before I do weapons.  This strategy is viable against the 0.41.2 placeholder AIs (even on Godlike), but it surely won't work in multiplayer (or Beta 2).  So ... maybe that's a core tech, and we should all take it sooner.

G2 Diplomacy is >= 27*8 >= 217(!).  That's off my beeline!

I agree with androshalforc.  The current way inflation is applied makes it seems stupid to research anything that is not core to winning as quickly as possible.  If I research 1878 research per turn in late game why in the world would I spend two turns researching missile augmentation, espcially as the game currently stands you don't get any left over research back after a turn.  They need to figure out a way to better balance the research tree.  Examples would be to make the future techs a lot more expensive, which causes you to either expand more to get more base research from colonies and or research research related tech bonuses.  In order to do either above you have to research other techs in the base of the game.  I don't know if anybody played Endless Space on here but the research flowed pretty smartly and was balanced nicely (had four tech trees as well).  Yeah you could go back and research previous techs pretty easily and quickly in that game, but due to the strategic resources unlocked by certain techs and a closer tie between techs in a tree you usually didn't skip many techs on your way to winning the game.  Well at least I didn't, anybody else play that game?