One thing that bothered me about HOI3 was that you knew exactly when your technology research would be complete. In real life I doubt the USA knew when they would have their first successful A-bomb test (they just knew they wanted to focus on it and beat the Germans to it).
What I think would be a nice change is that you can decide how much to focus on certain areas and this would reduce the MTTH. Naturally, there should be huge negatives to researching certain chains ahead of time (need to prevent nukes in 1939). What this would do is it would mean you could not always be sure you would be able to start producing the next unit at a certain time. Additionally there would need to be some diminishing returns so that assigning 10 times as many scientists does mean something is discovered 10 times as fast).
Additionally, if you are researching old tech the MTTH time would be the other way. As an example if its 1940 and you are finally researching 1938 tanks it would be faster then researching the tech in 1938. Its always easier to copy then break new ground. This could of course be further complicated by spies / allied discovery etc...but that me be too much for a grand strategy game.
So basically my suggestion comes down to not knowing when exactly you will discover a tech, but still allowing the player to focus on certain techs as more important then others.
What I think would be a nice change is that you can decide how much to focus on certain areas and this would reduce the MTTH. Naturally, there should be huge negatives to researching certain chains ahead of time (need to prevent nukes in 1939). What this would do is it would mean you could not always be sure you would be able to start producing the next unit at a certain time. Additionally there would need to be some diminishing returns so that assigning 10 times as many scientists does mean something is discovered 10 times as fast).
Additionally, if you are researching old tech the MTTH time would be the other way. As an example if its 1940 and you are finally researching 1938 tanks it would be faster then researching the tech in 1938. Its always easier to copy then break new ground. This could of course be further complicated by spies / allied discovery etc...but that me be too much for a grand strategy game.
So basically my suggestion comes down to not knowing when exactly you will discover a tech, but still allowing the player to focus on certain techs as more important then others.