Stat squish confirmed: How will unsynced old content be handled?

[TheMightyMollusk]

Lower levels aren't really the problem. The problem comes at the end of expansions, where gear score explodes. We don't notice it much during the patches when they're current because we're just keeping up, but just look at the jump from, say, level 70 job gear (item level 290) to Augmented Scaevan (400). And then the next expansion has to keep up with that power creep.

[Payadopa]

I hope this will mean lower-level content like the Crystal Tower will be less of a broken faceroll. The enormous leaps in stat growth are a huge problem, and I'm glad it's getting addressed.

I think that's the reason why everyone and their pet moogle want to be done with it as quickly. There is no engagement. It should require a modicum of effort. Would still be pretty easy, but, y'know, engaging. At least force us to do mechanics again. lol

[Ammokkx]

Lower levels aren't really the problem. The problem comes at the end of expansions, where gear score explodes. We don't notice it much during the patches when they're current because we're just keeping up, but just look at the jump from, say, level 70 job gear (item level 290) to Augmented Scaevan (400). And then the next expansion has to keep up with that power creep.

You can look at the stats you have in Bozja and compare it to what you have equipped outside it

[Rongway]

But can someone explain to me like I'm 7, why stat squish will affect unsyncs? Like didn't he say relatively, if players and mobs are squished by 25%, meaning everyone gets weaker by 25%? Why does it affect unsync?

Not quite ELI7, but....

The growth from 50 to 80 isn't linear. The curve gets steeper the higher you go. That is, the amount of growth between 50 and 60 (10 levels) is not the same as the growth from 60 to 70 (10 levels), and neither is 1/3 the growth between 50 and 80 (30 levels). The difference is much greater.

Here are some hypothetical graphs:



The top row is an example of the current power growth curve. The heights of the pink bars show the power at a lower level, 50 or 60, and the heights of the blue bars show the power at a higher level, 60 or 80. You can see that the difference between overpowering Lv50 content at Lv80 is several times more than overpowering Lv50 content at Lv60.

The bottom row shows a hypothetical stat squish. In this particular example, the post-squish Lv80 is about 1/3 the power of pre-squish Lv80. Notice that pre-squish, Lv60 was about 2.5x as powerful as Lv50 but post-squish Lv60 is only about 1.5x as powerful as Lv50. Likewise, pre-squish Lv80 is about 16x stronger than Lv50 and about 6x as strong as Lv60, but post-squish Lv80 is only about 6x as strong as Lv50 and 3x as powerful as Lv60.

The new power curve would mean that a Lv80 doing Lv80 content is still relatively the same pre- and post-squish, and a Lv50 player doing Lv50 content is still relatively the same pre- and post-squish. But compared to unsyncking content pre-squish, the post-squish Lv60 player doing Lv50 content is about 4/5 as powerful as they were before, while the post-squish Lv80 player doing Lv50 content is about 1/3 as powerful as they used to be.

Of course we have no idea how steep the new curve will be in reality. This is a fairly exaggerated example to make the concept easier to visualize, but hopefully this explains why there would be a difference at all just for unsyncked players. This could very well mean that things some players are used to soloing might minimally need to be duoed, or maybe a duty you can currently duo might minimally need 1 each of tank, healer, and DD after the squish.

[Deceptus]

A stat squish. A concept WoW has attempted multiple times and gotten wrong EVERY.SINGLE.TIME. Something always goes wrong.

If this is true and well developed:

Damage reduction example: 530 BLM using Fire 4 with no crit/direct hit deals about 50k damage, this will become about 10k damage. Ratio of damage to enemy health will stay the same

Then this should NOT occur.

Undersized parties might be more difficult

The FFXIV Developers have two choices in front of them. Make enemies stronger and old, outdated content harder or make players stronger and old, outdated content easier.

It's a tough balancing act and I'm glad I don't have to be the one pulling the trigger on the decision. I am concerned with the FFXIV development team because their track record on system overhauls isn't exactly stellar, or even good at this point.

Lower levels aren't really the problem. The problem comes at the end of expansions, where gear score explodes. We don't notice it much during the patches when they're current because we're just keeping up, but just look at the jump from, say, level 70 job gear (item level 290) to Augmented Scaevan (400). And then the next expansion has to keep up with that power creep.

The Devs have pretty much standardized ilvls over the last few expanions. It's usually 120-130 ilvls from old expansion tome gear to new expansion tome gear.

[Hycinthus]

Thank you, the graph helps and makes it very clear.

[TheMightyMollusk]

The Devs have pretty much standardized ilvls over the last few expanions. It's usually 120-130 ilvls from old expansion tome gear to new expansion tome gear.

That was my point, yes. How much do stats escalate over those item levels? We jump that every time we go to a new expansion, and we're going onto the fourth expansion soon. That means three big end-of-expac gear jumps, with corresponding stat jumps. That's a big chunk of what's being smoothed out.

[MsQi]

"440,000,000HP!"

oh, are we still using windows 98?

[Feronar]

The problem is pretty much the exact same as WoW had at the end of Mists of Pandaria. Blizzard had to design around the problem in Siege of Orgrimmar, the final raid. The second-to-last boss, Paragons of the Klaxxi, was actually multiple bosses daisy-chained together, and the final boss, Garrosh Hellscream, self-healed twice over the course of the fight (Three times if playing on the highest difficulty level). Garrosh's effective total HP across all phases was about 3.77 billion (37.7億) on 25 player heroic difficulty.

[Daeriion_Aeradiir]

"440,000,000HP!"

oh, are we still using windows 98?

No, but they're likely approaching the 32-bit integer limit which with their current scaling, enemy HP would 100% hit that overflow point in 6.0.
And no, simply re-assigning certain values to higher datatypes isn't as simple as it looks. Since packets would have to encapsulate that data, any change in the variable size for certain values would increase packet size. Which magnified by -x- amount of players doing the boss across -y- amount of instances (which at the beginning of new raid patches, can be thousands of instances) can increase server I/O load substantially.

Given that increasing our inventory by even 1 slot seems to be a big task that they have to meticulously plan out due to the increase in packet size it brings, it's no surprise they'd rather stat squish than re-program many of the int values in the game into higher encapsulating datatypes.