"likelihood of being available"

The techs are not guaranteed to show up when you make a breakthrough. Each tech has a chance to show up or not. Green techs are very common and will show up most of the time. Yellow techs are more uncommon and will show up sometimes. Red techs are rare and won't show up a lot of the time. So for example, if you're researching Civilization you always have a good chance that "Farming" will show up on the list of breakthroughs. But once you get to Advanced Farming (a red tech), you can go several breakthroughs without it showing up.

Hope this clarifies a bit

How likely that, when you have spent enough time researching a technology area to achieve a "breakthrough", that you will be able to choose THAT particular breakthrough as your reward. I hope that addresses your question.

When you get a breakthrough in the given area, the likely and unlikely stuff won't necessarily appear for you to select. They are rolled for separately in each breakthrough.

thanks all!

Barrett

i wonder if the ultra rare grey techs are still in?

Care to give an example?
I haven't seen any gray techs...

Nor have I fill me in.

The posters above me explained it very well.

I wouldn't mind having the probability mechanic in the tech tree go away. In addition to the balance issues created by having a tech tree designed this way, I think these random based mechanics are standing in the way of a well-defined, structured and meaningful tech tree.

Yeah, I'm not really sure what this aspect of research adds to the game.  It doesn't really *randomize* tech development to any degree. You still tech up the same every time. You still want what you want, when you want it... only now you just have to suffer through the frustration of maybe not getting it, despite sinking the time into researching it.  At best, it's a speed bump. At worst it is downright annoying.

I generally just keep reloading a save til I get what I want anyways.  If they really want that random aspect to function, it would be better for the game to decide what options will show up at the beginning of the research cycle.

hate when you only have Unlikely skills to unlock, then at the end of your research, you get nothing but a reset research timer

I thought it was kind of a cool change from the very specific stuff like 'I'm going to research aerial ninja combat tactics' you see in other games  Research is always so ridiculously specific, in this case you're just researching a field and when you have a breakthrough, you can choose what your people discover.  I've never had one turn up empty though, that owuld make me pretty angry. 

I like it as well.

And this is a very tame version of uncertain research, take master of orion for instance, there if the game didn't give you scatter pack-V missiles then you were out of luck and never able to research them. In elemental, all you have to do is to try again

The random element adds a nice quirk of unpredictability.  You can't just race to one tech, you have to adopt a wider strategy.  I love it.  For the many things that need work in this game research is not one of them.  I feel like the techs are one of the few things they nailed.

How do you "adopt a strategy" around a random event? A tech tree based on random breakthroughs discourages the player from making new strategies. Researching A, B and C will no longer yield D on a consistent basis. Instead, a player is at the mercy of a random number generator.

That just didn't make sense. If your strategy relies on having technology X, and you don't get it, it doesn't discourage you from making a new strategy, it forces you to make one!

I think he is saying that you will always take the path which offers the most likely return on investment.

No, it does make sense. I will rephrase my point by asking a question: What incentive do I have for creating a strategy which only works when I am lucky?

The difference is entirely superficial, though. When you research a general category, it is because there is a specific tech you have decided you want.  You're still targeting a specific tech.  The difference is, all those turn you invested into the research cycle might turn out to be a waste.  So you keep on researching that general category, until you get what you want. If you have a need for housing tech, you're researching that category to get housing tech.  Getting mining tech instead when you have a boatload of iron, is not going to make you happy. I still need that housing tech.

It adds delay and possible frustration, but no real randomization. It serves no purpose that I can see, other than a time sink. I'd just prefer they let me research exactly what I want at a given time, and just add in the extra time the system adds in as an average.  I'm going to research till I get the tech anyways.

I can understand the point, however I think its a difference between sim gamer and power gamer.  Some people want to dominate the game, so they form A to B strategies and pursue that each time.  Others like myself don't do that, and appreciate the less silly idea of researching 'weapons' instead of specific research targets.  So I understand why you might not like it, but I do for this one. 

I think it's an interesting system, a welcome change from Civ's produce-as-much-science-as-possible-till-ICBM "strategies".

I'm not talking about a concrete, inflexible optimized build order.  I'm talking about responding to the needs of your kingdom at any given time.  Certainly people pursuing a specific strategy will pursue the same techs, even if they get them in slightly different order each time. But you don't have to want to "dominate the game" to simply run your kingdom well.

If your cities are at population cap, you need to expand your cap. So you need housing tech.  If you get useless +mining tech instead, thats not going to meet your needs, and you are still going to simply research that category until you get housing tech. 

If you're being invaded by powerful massed armies, you need massed army tech to compete.  Naval tech doesn't meet your needs.  Or maybe you decide you can counter them with a new trainable monster unit, so you research that tree instead. You are still going to research until you get what you need for your current situation.  Your needs don't change, just because you don't get the tech you want at a given time.

There is no real randomization in any meaningful way. Only potentially frustrating delays in the deliberate pursuit of your tech.   I would be in favor of  a random element if it actually added anything to the game. I see no compelling argument for anything this mechanic adds other than a time sink.  So I say fine.  If you want techs to take longer to research, just average those extra turns into the research time as a representation of the "trial and error" involved, and let us pursue the techs we are going to  research anyway, no matter how long it takes, but without the pointless "randomization" elements.

If you really wanted to add a randomization element that wasn't quite as alienating, that maybe served to limit accessibility to some of the more powerful techs in the game, it would be better to add the randomization at the *beginning* of the research selection, rather than at the end.  Whatever techs appear at a given time to be researched (based on probability) are the techs you can pursue.  It won't make you any happier that there is no +housing tech available to research if you need +housing tech, but at least you wont feel like you are wasting your time getting useless tech, and you can make an informed decision about researching something else that may be of use to you until +housing becomes available.

regarding rare techs:

Initially, there were supposed to be super-duper "purple" or "grey" techs of awesome. Either they were removed, or they are hidden in data.zip. I'm inclined to say they were removed.