Tech tree: flattening tech era costs.

To avoid opening new threads, if you want some analysis of data you can ask and I'll try to post here. Actually I'll start for this one, as some people are suggesting to flatten tech ea costs.

I made some numbers regarding research costs and turns. I look how much turns took to research techs.

This first graph compares the time to research a tech of different eras simply dividing its science costs by science production. Also I added a line to compare with similar developing in ES1 (assuming you were picking one tech of 3 different branches of same tier).

In x axis are total number of unlocked techs (not stacked with eras, as past eras are simulated), and in y axis the turns to research next tech.

Here you can see the research times going up in era I, and near to starting era II they drop. If you continue with era II techs they go up again and fall down. Something similar happens with era III.

The line simulating ES1 tree increases it research times progressively, so I suspect the actual progression of science production won't feed well with fixed values from Es1. If you want to go back to this system, you'll need tweaking it from the start (an option I'm not against, but just to compare).

Now: there's people suggesting flattening the costs. I made same simulation. In first graph they're flattened using a mean of techs unlocked to reach next era, and in second the mean using the sum of all available techs per era in my game till turn 125.

As you can see, this means a starting with a research time of 15 turns for Era I techs, 20 for Era II and 10 for Era III, with a continuous drop after that. I felt it game pace breaking. So I'm not vary fond of this approach, and to resolve actual system it will be better to tweak again science inflation, with higher discounts on past era research (to say in Era III era I tech should be always 1-turn unlock, and era II 2-3 turn unlock at maximum).

Also there are good suggestion in this post by Sotnik.

That early spike is my problem, especially as Sophons, with the way outposts and early population growth works you are stuck at your turn 1 science production for a long time while the cost of science goes up quickly. And there is diminishing returns on technological advancement so rushing a tech/improvement to smooth out that spike hurts you in the long run on empire growth.

Kareal wrote:

That early spike is my problem, especially as Sophons, with the way outposts and early population growth works you are stuck at your turn 1 science production for a long time while the cost of science goes up quickly. And there is diminishing returns on technological advancement so rushing a tech/improvement to smooth out that spike hurts you in the long run on empire growth.

I would still argue the largest problem is the escalating cost. Want that other Era I tech? Just know grabbing it will screw you for Era II!

Kareal wrote:

That early spike is my problem, especially as Sophons, with the way outposts and early population growth works you are stuck at your turn 1 science production for a long time while the cost of science goes up quickly. And there is diminishing returns on technological advancement so rushing a tech/improvement to smooth out that spike hurts you in the long run on empire growth.

Yes. Some happening to me.

Romeo wrote:

Kareal wrote:

That early spike is my problem, especially as Sophons, with the way outposts and early population growth works you are stuck at your turn 1 science production for a long time while the cost of science goes up quickly. And there is diminishing returns on technological advancement so rushing a tech/improvement to smooth out that spike hurts you in the long run on empire growth.

I would still argue the largest problem is the escalating cost. Want that other Era I tech? Just know grabbing it will screw you for Era II!

I think it can be tweaked. Put a high bonus to research for past eras, and doesn't make it count into more advanced eras tech inflation.

After seeing the graphics, I'm not against tech inflation, but consider each tech should contribute only to it's era inflation.

My idea (need to have game with 5 eras to provide enough numbers) is to give past eras a 20% discount for era unlocked. This way, if you're in Era II you'll have 20% bonus to Era I, and if you're in ERa V, you'll have 20% for Era IV, 40% for Era III, 60% for Era II and 80% for Era I bonus.

lo_fabre wrote:

Romeo wrote:

Kareal wrote:

That early spike is my problem, especially as Sophons, with the way outposts and early population growth works you are stuck at your turn 1 science production for a long time while the cost of science goes up quickly. And there is diminishing returns on technological advancement so rushing a tech/improvement to smooth out that spike hurts you in the long run on empire growth.

I would still argue the largest problem is the escalating cost. Want that other Era I tech? Just know grabbing it will screw you for Era II!

I think it can be tweaked. Put a high bonus to research for past eras, and doesn't make it count into more advanced eras tech inflation.

After seeing the graphics, I'm not against tech inflation, but consider each tech should contribute only to it's era inflation.

My idea (need to have game with 5 eras to provide enough numbers) is to give past eras a 20% discount for era unlocked. This way, if you're in Era II you'll have 20% bonus to Era I, and if you're in ERa V, you'll have 20% for Era IV, 40% for Era III, 60% for Era II and 80% for Era I bonus.

Actually, by looking at your data, the game already offers a discount (though not cumulative) on techs from a past era.

As you can see, the science cost grows according to a specific curve if you follow the 'optimal' path of 10 techs per Era. When researching techs from previous Eras, it seems that the costs are defined by a another curve with slower growth. Noting that, I was enticed to make a simple prediction of how much science is required to reach Era 3. I simply calculated the total cost in science points required to research a total of 20 technologies according to the number of Era 1 techs researched. I assumed you researched all those Era 1 techs before any techs from Era 2; it probably isn't optimal but it was easier and is sufficient to make my point.

Now I know reaching eras faster doesn't necessarily help, but I still find it awkward how the costs progress and how you can't 'cut' a third of the science cost to reach Era 3 by taking only the basic ones. Overall, I just think the Era system, in regards to how it is currently integrated into the tech 'tree', is inconsistent in many aspects with (what I believe to be) the intended gameplay experience. I'm not saying we need to throw away the concept of Eras altogether, but I really think we'd be better off with a real, conventional tech tree. Eras could progress simply according to the total number of science points created during the game, according to the number researched techs in that fixed-cost tree or whatever else. As long as we lose the feeling of constrained progression and can specialize again. I understand how the Era-tree makes some players less concerned about researched by not having dependencies, but it drives players obsessed with path optimization (like me) to madness, since we cannot 'unsee' the limitations imposed by growing costs or researching in previous Eras.

Trentius wrote:

lo_fabre wrote:

Romeo wrote:

Kareal wrote:

That early spike is my problem, especially as Sophons, with the way outposts and early population growth works you are stuck at your turn 1 science production for a long time while the cost of science goes up quickly. And there is diminishing returns on technological advancement so rushing a tech/improvement to smooth out that spike hurts you in the long run on empire growth.

I would still argue the largest problem is the escalating cost. Want that other Era I tech? Just know grabbing it will screw you for Era II!

I think it can be tweaked. Put a high bonus to research for past eras, and doesn't make it count into more advanced eras tech inflation.

After seeing the graphics, I'm not against tech inflation, but consider each tech should contribute only to it's era inflation.

My idea (need to have game with 5 eras to provide enough numbers) is to give past eras a 20% discount for era unlocked. This way, if you're in Era II you'll have 20% bonus to Era I, and if you're in ERa V, you'll have 20% for Era IV, 40% for Era III, 60% for Era II and 80% for Era I bonus.

Actually, by looking at your data, the game already offers a discount (though not cumulative) on techs from a past era.

 

As you can see, the science cost grows according to a specific curve if you follow the 'optimal' path of 10 techs per Era. When researching techs from previous Eras, it seems that the costs are defined by a another curve with slower growth. Noting that, I was enticed to make a simple prediction of how much science is required to reach Era 3. I simply calculated the total cost in science points required to research a total of 20 technologies according to the number of Era 1 techs researched. I assumed you researched all those Era 1 techs before any techs from Era 2; it probably isn't optimal but it was easier and is sufficient to make my point.

Now I know reaching eras faster doesn't necessarily help, but I still find it awkward how the costs progress and how you can't 'cut' a third of the science cost to reach Era 3 by taking only the basic ones. Overall, I just think the Era system, in regards to how it is currently integrated into the tech 'tree', is inconsistent in many aspects with (what I believe to be) the intended gameplay experience. I'm not saying we need to throw away the concept of Eras altogether, but I really think we'd be better off with a real, conventional tech tree. Eras could progress simply according to the total number of science points created during the game, according to the number researched techs in that fixed-cost tree or whatever else. As long as we lose the feeling of constrained progression and can specialize again. I understand how the Era-tree makes some players less concerned about researched by not having dependencies, but it drives players obsessed with path optimization (like me) to madness, since we cannot 'unsee' the limitations imposed by growing costs or researching in previous Eras.

Yep, good to see some actual figures on it, though it some seem to justify what many have been saying: The system basically punishes you for taking anything besides "your 10" each Era.

And I still feel that the Era system is more constraining than the old tree ever was. Knowing that I have to follow the same growth curve as everyone else, knowing I can't focus on a particular tech, knowing that I can't specialize in a particular field... It's all very limiting.

Trentius wrote:

Actually, by looking at your data, the game already offers a discount (though not cumulative) on techs from a past era.

 

As you can see, the science cost grows according to a specific curve if you follow the 'optimal' path of 10 techs per Era. When researching techs from previous Eras, it seems that the costs are defined by a another curve with slower growth.

Yes, as I know there's a discount. Din't try to calculate how much is it. Bu it feels not enough, as going back to past era techs cumulates with actual tech and cripples your research.

 Noting that, I was enticed to make a simple prediction of how much science is required to reach Era 3. I simply calculated the total cost in science points required to research a total of 20 technologies according to the number of Era 1 techs researched. I assumed you researched all those Era 1 techs before any techs from Era 2; it probably isn't optimal but it was easier and is sufficient to make my point.

Looking at your graph it seems that taking always Era I tech lowers the total cost requiered for Era III (not sure if feaseable, I think Era I has 19 techs).

Not sure what exactly meansor if you can use it in your favour, but if not considering anything except numbers, one will say (of cousre will be mistaken, there's always the benefits from techs) that the better way to get to Era III is skipping Era II techs.

Also not sure if this have any sense in any way...

Now I know reaching eras faster doesn't necessarily help, but I still find it awkward how the costs progress and how you can't 'cut' a third of the science cost to reach Era 3 by taking only the basic ones. Overall, I just think the Era system, in regards to how it is currently integrated into the tech 'tree', is inconsistent in many aspects with (what I believe to be) the intended gameplay experience. I'm not saying we need to throw away the concept of Eras altogether, but I really think we'd be better off with a real, conventional tech tree. Eras could progress simply according to the total number of science points created during the game, according to the number researched techs in that fixed-cost tree or whatever else. As long as we lose the feeling of constrained progression and can specialize again. I understand how the Era-tree makes some players less concerned about researched by not having dependencies, but it drives players obsessed with path optimization (like me) to madness, since we cannot 'unsee' the limitations imposed by growing costs or researching in previous Eras.

 May be a good solution making Eras related to total tech researched. I'm more of the mixed system, or the ES1 tech tree, but if Amplitude wants to mantain the Eras, they can improve it a lot changing the way costs are calculated. As said many times, there are lots of good suggestions in this forums.

lo_fabre wrote:

 May be a good solution making Eras related to total tech researched. I'm more of the mixed system, or the ES1 tech tree, but if Amplitude wants to mantain the Eras, they can improve it a lot changing the way costs are calculated. As said many times, there are lots of good suggestions in this forums.

Yeah, the more I think about it, the more this seems like the only alternative that will save both the Eras and grant real freedom in the tech tree. Personnally, I still believe we could have a game without Eras, but many people seem to like the sense of pacing they bring and I respect that.