1
This first PTS session was supposed to be all about Infotech, but it also removed all weapon quirks, changed many mechs' mobility profiles, and added more structure buffs than I could possibly keep track of. What I mean by this, is I think it would be easier to provide feedback if we could focus that feedback toward very specific changes. Maybe on Friday we do JUST structure bonuses, on Saturday we do JUST mobility changes, on Sunday we do JUST infotech. Then maybe the following week we do a day or two of global weapon balance, a day of per-mech weapon quirks, and a day to check Clan/IS balance. If you throw a lot of changes at us at once, and they're not finished and just an initial pass? You going to get a lot of "ERMAGERDZ, THIS IS TERRIBLE EVERYTHING IS S MESS" and that's exactly what you got.
#2: consolidate and organise quirks.
Almost every mech in the game has this atrocious list of quirks. Here's the Banshee 3M for instance:
ACCELERATION RATE -15%
DECELERATION RATE -15%
ADDITIONAL STRUCTURE (CT) +11
ADDITIONAL STRUCTURE (LT) +10
ADDITIONAL STRUCTURE (RT) +10
ADDITIONAL STRUCTURE (LA) +10
ADDITIONAL STRUCTURE (RA) +10
ADDITIONAL STRUCTURE (LL) +10
ADDITIONAL STRUCTURE (RL) +10
TARGET RETENTION TIME -2
TARGET SCAN TIME SHORT RANGE +400%
TARGET SCAN TIME MEDIUM RANGE +82.5%
TARGET SCAN TIME LONG RANGE +20%
TARGET ACQUISITION DELAY +4
Aside from the fact that I disagree with the premise that almost all mechs in the game have been blessed with between 25% and 200% structure bonuses except for an unlucky few, this needs to be simplified. I much prefer the way this is done on the live build: only structure quirks for mechs that are in dire need of it or in cases where it just plain makes sense. Such as the hunch on a Hunchback, or the overgrown monstrosity of a right arm that the Dragon has along with its very unfortunately overexposed CT. If you want to increase TTK, I would much rather you increase the base structure of all mechs in the game, not via quirks.
Now, if you have to add structure to all seven components of a mech, just do it like this:
ADDITIONAL STRUCTURE (all) +10
No need to worry about the CT having +11... it doesn't really matter. Just consolidate everything into one quirk. Same goes for acceleration quirks. Why two quirks when you could have one?
ACCEL/DECEL -15%
For Target Retention Time, it says (-2). But what's the unit? It's not %, it's not seconds, it's ... what is it? Tell us what it is. Better than telling us whether it's "plus" 2 or "minus" 2, just tell us what the physical delay is with the modifier applied, i.e., what does it add up to in reality? Same goes for Target Acquisition Delay.
Similar with Target Scan Time. This is something that Phil said in his recent discussion video. Instead of massive scary numbers like 400%, just give us a hard values measured in seconds. Then you don't need to break it down into all three range categories. Also, keep it separate from the quirks. Create a new section of the Mech Stats UI that just says "Sensors". Then put in all the hard numbers with minimal need for modifiers.
With all these changes in, we'd see a quirk list that looks more like this:
Quirks:
ACCEL/DECEL----------------------- -15%
ADDITIONAL STRUCTURE (all)-- +10
Sensors details:
TARGET RETENTION TIME ----- = 0 seconds
TARGET SCAN TIME (all ranges) +3 seconds
TARGET ACQUISITION DELAY - = 5 seconds
#3: just give us the data. I know you want to put emphasis on actually testing all the changes, but there's a lot of changes that could be addressed with zero testing at all. For instance, the Locust changes. I can look at the quirks list for the Locust and tell you immediately without playing it that it makes the Locust terrible. Same for the Spider, same for the Commando, same for the Mist Lynx... these are all underperforming mechs as long as they are without weapon quirks. I could say the same for the BJ-1X, the DRG-1N, and a lot of other mechs that are only good because of weapon quirks. They're good for a reason, and if you take that reason out from under them, you need to make some incredibly substantial changes to make up for it. Personally, I don't think anything short of weapon quirks will do, because mechs need to be able to compete in the damage department or else they won't be worth taking. For more information, see Rak's video.
"The Locust should be a high-risk high-reward mech. And ... when you take out its weapon quirks, it's just a ... high-risk mech."
#4: Omnimechs are a disastrous mess.
The base CT quirks for most of them seem to be massively in the negatives, and then the individual omnipods, when assembled in various configurations, make up for it and some stats wind up in the positive while others are still in the negative.
This is grossly unintuitive. Pleasseeee don't do this.
Why does such a benign and relatively useless mech as the Mist Lynx... have to start in the negatives? This doesn't make people want to figure out how to play omnipod Tetris with their Mist Lynx in order to maximise the best quirks... it just makes them want to quit the game in frustration or go find an online guide where somebody else has already figured it out. It's not user-friendly and it needs to be a LOT simpler.
#5: some variants are extraordinarily different from others.
Why? Why does the AS7-D have the structure values of a 110-ton mech, while the other variants do not? It's the same chassis. Why does the ADR-PRIME get +80% acceleration when the other Adders get +5%? It's the same chassis. Why does the BJ-1X get +30 yaw angle, and the BJ-1 gets -20? It's the same bloody chassis. Don't mess with the qualities that define a chassis. If a chassis needs buffed in regards to acceleration, structure, torso twist, etc... buff the chassis, not just certain variants. Where some variants fall behind the others in the same chassis is probably because they can't compete in firepower. The solution to this is to add weapon quirks.
Edited by Tarogato, 13 September 2015 - 11:27 PM.
