From: St.Petersburg, Russia
With all the talk about ships, I forgot to make a comment about the flak performance...
I certainly don't find it ahistorically low in general, although I can easily see how this opinion can form in actual game, where both CAP and flak hit attacking planes and it is hard to recognize from which operational losses result. There are some discrepancies - if the game code considers extra vulnerability of torpedo bombers, particularly large 2E ones, at all, it is completely trumped by greater durability and greater ability to suppress flak by taking out enemy ships - but overall 1/5th - 1/3rd of attacking planes lost from flak alone, even against a relatively unimpressive concentration of ship-based AAA, seem to be right, maybe even a bit too good. I also feel that this level of flak is sufficient to produce a game balanced roughly to historical level of outcomes with PDU OFF.
I haven't done an equally detailed test with the same TF using DaBabes flak upgrade, because I didn't have time or patience (at least yet) so I can just say that in a few turns where I tried to use the device list from there, Japanese losses jumped to roughly 40-50% of the attacking planes from flak alone.
Overall, while DaBabes flak seem to produce much more historical results in the beginning of the war (judging by the PH strike mostly), I think that it is too good by the time of Downfall. The stock flak might use a little bit of a buff maybe, but not the almost doubling of results. From the narrow historical viewpoint I'd prefer the stock flak...
Now, in the wider viewpoint, acknowledging that generally Japanese air attacks will be several times more powerful than historical, particularly with PDU ON (and I prefer to play PDU ON), while the Allied fleet strength is fixed, DaBabes flak might produce results that, in the big picture, are closer to the historical outcomes of campaigns. I can't say for sure. I'll certainly continue testing, and hopefully our ongoing games will provide more data as well. At the moment I won't recommend adopting DaBabes flak model for Scen 70.
Regarding ongoing games, by the way... the test has shown that:
1)Torpedo attacks are vastly more effective than bomb attacks, at least until the appearance of late-war planes with 800-kg bombs. Sure, they miss alot, but hits count for much more. Doing massive damage to the enemy early on also greatly helps to reduce losses - B7As suffered much more as divebombers.
2)Ki-49 and Ki-67 are the best survivors by far. Once again, small differences in stats seem to produce large differences in results.
3)Ki-48 fails in the divebomber role as well. It has no worth and should not be built, much less researched.
4)G4M3a actually outperformed P1Y2... which surprised me. The difference is not too big, though, and, with relatively small number of tests, might be explained by luck.