DeaconW, on 28 June 2013 - 02:59 PM, said:
No, Kunae is illustrating the logical fallacy of reductio ad absurdum.
*cough* Reductio ad absurdum is not a logical fallacy... It's a form of argument. That was covered in my Logic 102 class, but here's a wikipedia link for you if you don't believe me:
https://en.wikipedia...tio_ad_absurdum
DeaconW, on 28 June 2013 - 02:59 PM, said:
Either way poptarting is minimized. If they can change it without bringing the old broken meta back, I could support that.
Yes, and PGI could also minimize or eliminate poptarting by just removing JumpJets entirely. That doesn't make it a good idea.
Screen shake doesn't QUITE remove JumpJets entirely, but it DOES have the effect of reducing ALL JJ activity by 75% across the board. That is not a good way of countering poptarts.
Now, I have mentioned several times a number of changes that could be implemented to counter-balance poptarts, reduce the effectiveness and overpoweringness of ALL direct-fire snipers, and increase the variety and diversity of mechs and mech loadouts. What do you think of these?
1. Nerf direct fire support weapons, and PPCs in particular, by reducing the travel speeds of all non-missile projectile weapons (possibly excluding the AC/20), by increasing the cycle time of both PPCs and the Gauss Rifle, and by increasing the heat of the PPC back to the table top stock value of 10 (up from 7), and the ERPPC back to the TT stock value of 15 (up from 11). This would fix the current meta of PPCWarrior Online, and make all direct fire weapons less effective. Convergence would remain the same, but snipers would have to be much more careful with their shots in order to hit, and would pay much more for those hits in terms of heat and ability to fire again, and pay dearly for misses (this would affect poptarts, as well; the tonnage taken up by JJs is usually pulled from heat sinks rather than weapons, and most poptart builds run close to the heat limits of the mech already).
2. Adjust the model scaling of all mechs to make them sized appropriately for their tonnage, especially Lights and Mediums (as noted in
this thread, the mech scaling is WAY off). This would make sniping the lighter mechs harder, as they would be appropriately smaller targets, and make Lights and Mediums in general, the true bane of poptarts and all fire support mechs, much more viable.
3. Increase the damage and reduce the spread of all SRMs (Streaks can have the ammo per ton and/or damage per missile reduced from the new figure to help balance them), and reduce the weight of the AC/10 by one ton, and the weight of the LBX-AC/10 by two tons. This would make SRMs, the bread and butter of brawling Lights and Mediums, viable weapons again, and the reduction in weight of the AC/10 and LBX-AC/10 would make both more worthwhile for the weight, and make them more viable for Medium mechs to carry. Both would make Lights and especially Mediums much more viable mechs in general, and increase the number of fast harasser mechs (the bane of the poptart) in the game.
4. Change weapons so that the firing mech's velocity is added to the weapon's velocity, requiring mech pilots to lead their targets when moving, even if the target is stationary, as they would have to IRL. This would make all targeting more difficult, but the snipers would be most effected by it, and the poptarts most among the snipers, because lead distance for movement is a function of relative speed and distance to target (Lights move really fast, but most of their engagements occur at close range, where lead distance would be limited), and poptarts would have to account for their own vertical movement as well as any horizontal movement.
5. Now that we have HSR and a net code that PGI seems confident enough in its ability to handle 12-mans that they're adding it to the game in the next patch, IIRC, lift the arbitrary speed cap, so that fast Lights actually take full advantage of the maximum speed they can get out of their chassis. This would allow for players to create super-speedy Light mech builds, which would be especially hard for snipers and poptarts to hit, allowing them to quickly close the distance on all fire support mechs and start harassing them, either presenting a serious direct threat to the fire support mech, or at the very least forcing them to turn away from providing their fire support (if properly balanced, a 3-ML Spider should not be an immediate threat to a Heavy or Assault fire support mech of any type if it turns to engage, but it is not something that can be ignored, either, because it will park behind the fire support mech and rip through his back armor in no time).
What do you think of these suggestions? 4 of them would be trivial changes for PGI to implement, and suggestion #4 should not be all that difficult for them to add. None of them significantly impact any other build type or play style than the targeted builds/playstyles, all of which are recognized as being over-powered, under-powered, or lacking a balance.
DeaconW, on 28 June 2013 - 02:59 PM, said:
You are entitled to your opinion. It is wrong, but you are entitled to it. And I am entitled to mine.
*sigh* That is not my opinion, I have no opinion on your motives for hating poptarts, I don't know them, and, honestly, I don't really care. I was using an example in an attempt to demonstrate a point: Your assumption that most people who oppose screen shake are ******* poptarts wanting their OP meta back is unfair, and your insistence on painting anyone who opposes screen shake in that light only makes you come across as an ******.
You just talked about how other people were coming across to you a post or two ago, please give some consideration to how YOU are coming across to others.