First, to answer a question in the "Individuals" thread, here's what YR7 proposed quite a while ago:
I'd also been thinking of giving Insight a greater role, but what I liked most about YR7's suggestion was comparing a total sensitive to Int, Inventive Genius, and MT against a rather high Ease Factor roll. I thought that extremely useful, mechanically.
We're still toying with the numbers, but using a high Ease Factor roll we've come up with a system that's sensitive to Int, IG, MT, and the magnitude of the experimental effects - we wanted to encourage high-level effects rather than low-level ones. There's also a failure state with a push-your-luck-mechanic.
Basically, when you want to do OR, you generate a "Research Total": Int + MT + 3 (if you've got IG) + the magnitude of the effect you're creating + roll.
For Minor Research your "Difficulty" (Ease Factor) is 18. For Major it's 21, and I think we're going with "Greater" (24) and then Hermetic (28). You generate points toward finishing the researched equal to however much you exceed the Difficulty. The Difficulty also increases the botch dice. (Up to 8, for Hermetic.) Points needed: 30, 45, 60, 75
You're no longer fishing for Discoveries. The law of large(ish) numbers still applies, but is no longer nearly as "streaky." I think this makes fine tuning everything much easier.
Minor is for research that doesn't so much increase a mage's power, but rather makes their magic more interesting or unique. It's designed to be relatively quick and easy. Greater and Hermetic involve breaking limits, and a mage needs to have both high Int+MT and high Art scores.
Then you spend a season to Stabilize (though we've fallen into the use of "Confirm") what you've learned - this is when you are actually awarded the points of progress. This gives you 1 warping point.
For any OR you're attempting, you can do this a number of times equal to your MT. This is the failure mode - You can guarantee enough progress with a high enough total, but the numbers can get pretty high: You might try the research without a certainty of success. This can make not knowing the exact number of points needed quite troubling. You can break this MT restriction, but you have to start getting somewhat reckless with your magic, and it generates more warping points. (We're still discussing how many: might go with magnitude - roll.)
We altered the Extraordinary Results chart: 0-4 is Side effect, and 8 is "modified effect" unless it's the last season of work. (This is due to one player who rolled "8" twice in a row early in the saga. We've taken mercy.)
After fiddling a lot with the numbers, we're happy with the results. Minor research will often be easier, which we like, while Hermetic Research is tough*. We're likely to adjust the Difficulty for specific projects. The player currently involved with some minor research was quite happy to switch to making two or three high level spells, that will mostly likely have several quirks, rather than a bunch of mid-level spells. He'll almost certainly get done in two thirds the time, too.
Initiation:
Replacing the Initiation system, the saga's Criamon character will be using OR, with quests and other activities in place of most - but not necessarily all - of the experimentation. Some of these activities will be cult rituals, but reliant on a modified Research Total for success.
*Using 28 as the threshold, a mage with 4 int and 12 MT, 16/20 Art scores with a Focus, and an excellent lab, and inventing 45th level spells in 1 season, will, with average rolls, succeed at Hermetic research after a mere 7.5 years. That strikes me as not long enough. OTOH, for our saga this is a mage of very rare power, and almost certainly rather old, and the last 3 projects will accumulate (using the what-we're-thinking-this-afternoon figures) an average of an additional 12 warping points - 4 at a time - on top of the 12 already accumulated via OR.
So maybe 7.5 years is OK. A PC would have to work hard to get to the point it's possible, and worry about death or Final Twilight. NPC mages can quite plausibly be made to either succeed, or die, vanish, or accidentally turn into a bedwarmer.