I think the costs of BV are overall reasonable (weight and PV), I like the idea of Scatter losing penetration at medium and long range, definitely like the idea of reduced damage leading to longer fights, agree that speed bonuses are important for areofighters and lighter mecha, think the situational bonuses are a good idea for more tactical combat in or out of tactical mode (getting a bit tired of 'Fly towards each other, turn around, do another pass, repeat'), and I sort of agree with Frumple, at relatively even odds armor isn't very critical currently, although I do like having it when I'm outnumbered 3-4 to 1 or more. However, there's a limit, anything that actually reaches the slow/heavy range gets blown to pieces. The biggest I'll go is a Savin stripped down with slightly lighter armor to reach at worst -2/-2.
Basically, with damage as high as it is, and accuracy where it is currently, anything above -2 or -3 maneuverability seems like a death sentence, and I'd like for actual honest to goodness heavies to at least have a place. Although in part, I think that might require de-coupling TR/MV, since in the end anything that reaches -5 MV also has pretty crippling TR issues.
Part of it is, I believe, technology inflation since GH1/system changes. Currently, starting mecha often manage -2/-2 compared to the -4/-4 of most GH1 starters, this extends through most of the range with "decent" mecha coming in around -2 and "heavies" being -3, unlike GH1 getting -1 or even 0 rated "super" mecha is harder (due to the number of components needed to manage good speed, defenses, computers, etc), but unlike GH1 there isn't room for -4 or -5 heavies, which I think added a lot of needed space for the curve. part of it is, I think that in GH1 the TR/MV penalties, while coupled in terms of how they accumulate were de-coupled in terms of off-setting them. When you could get a Tarcomp 8 to cover targeting, letting MV drop to -4 with just your built in Gyro 4 wasn't as much of an issue. I do *like* being able to modify MV, though, and think the computer system works well, but maybe change the scaling between TR_boost and MV_boost? I dunno.
Also, I think the cost of Hardening and Antibeam are excessive. While I don't think we want them to dominate the playing field, I think they should be an option. Like you can have a decent beam weapon mecha in the 500k and 1 million range, despite the high cost of beam weapons. There's almost no way to make a mech with even partial hardened modules/some armor without topping 1 million. So, yeah, you'd never see a starting value mech with any of that, but I'd like to see it priced so you could have a 500-800k bracket semi-decent mecha whose "Thing" is hardened armor. (actually, having just made a few, it's difficult to make armored mecha anywhere below the 500k mark period. I did make a decent starter mech (312k) with class 4 armor, that tops out at -4/-4 though, which I think is reasonable.)
One of the mecha I made, the Strikebreaker, is a class 5 battroid, all modules hardened and hardened class 5 external armor with some mass reduction (plus a hardened shield), it's got less than 300k worth of weapons on it, no particularly advanced electronics or movement, is -3/-3 and comes out to 2.5 million. Now, it's fairly high end for an armored behemoth, -3/-3 isn't bad for heavy armor, but at class 5 it just doesn't "feel" like it should be running neck and neck with a Savin. Now, it's "Thing" is that it's armored, so it might very well withstand a few shots from a Savin, but it's really "More difficult to get rid of Popcorn" rather than a real high-end unit. I get that Class 5 armor is pretty high-end overall (With only the really big boys getting better), and Hardening makes it just that much more so, but from the "feel" of it, I'd be happier if the Strikebreaker came in around 1.5 mill instead of 2.5.