Achieving fixed Speed wrt to Charge Time? [partial solution]

[R999]

I wanted a way to make it so that every character's speed to be fixed at roughly equal to 1 in-game speed stat, with no speed growths at all. And then have their speeds adjusted by their speed multiplier. In a way to make sure that the charge time abilities stay consistently fast or slow throughout levels 1-99 in the game, and more direct control over enemy characters' speeds (so a battle is not any easier or harder on speed scaling, independent on the player's level).

I like Philsov's approach to halving the speed growths in the game. That will hopefully help alleviate the problem with casters requiring Short-Charge by endgame. It's just that that's not an ideal solution for perfect consistency, and I do want a higher level of control over the speeds of specific battles; so the player can have a consistent level of difficulty independent of the player's speed levels.

In your opinion, what is the best way to achieve something like this? Or is it even remotely possible? Yes I am aware that the status durations and ability charge times would need to be adjusted accordingly.


edit: See replies below for a partial solution.

[formerdeathcorps]

This is easy.  All you need to do is tweak the speed multipliers and C values and all speed boosting equipment.  There is also an ASM hack in the Wiki that you can implement with orgASM that changes how CT is calculated as one levels up.  Another ASM hack exists to cap maximum level.

I personally don't believe either is necessary, since I prefer reducing the charge time of all spells (without any changes to effectiveness except for the really broken spells), giving some physical unit classes MP costs or charge times (so they can't skill spam as well), and reducing the availability (as in obtainable in Muround Death City) of high power gear (so they are less likely to 1HKO charging mages).

[R999]

By the Wiki, you mean here? http://www.ffhacktics.com/wiki/ASM_Hacking ?

I couldn't find anything related to Speed there. And I actually do not want levelling to have anything at all to do with Speed. I really want to have complete control over exactly how much faster or slower a unit is relative to another, and while also keeping every charging ability consistent throughout the game. I don't believe that orgASM hack can give me this.

As for tweaking the multipliers and growth values, I can do that I guess. Will play with it now. But I never really fully understood how the speed growths are calculated. I know the SPC is a number for growth division, but since I can't make this SPC=0 (because you can't divide by zero), what are my alternatives? I thought about using SPC of 1, but that speed would still scale with levels... at least it appears to me that way.

[formerdeathcorps]

I mean this.  http://www.ffhacktics.com/wiki/CT_Hack

We've seen that FFT remains fairly balanced at level 35 between magic and physical units from the AI tournaments.  If you set that as your maximum level (or even up to 50), you'll be fine.  There are ASM hacks in that section that let you change JP and EXP gained as well.  Combine that with speed C value growth at 255 and adjusted multipliers, and it'll work (provided you remove most speed adding gear).

EDIT: Units with 255 speed C value will gain a speed point at around level 30-35 if their speed multiplier is above average (level 1 base speed = 6 or 7), so this should suit your purposes if you cap levels and downgrade EXP/JP gain, though that's hardly necessary since until speed starts exceeding 10-11.

[R999]

Well there are so much restrictions for the AI tournaments (1.3) that it's making it somewhat balanced. But when you have access to the full arsenal of skills and gear (including speed gear, which I intend to keep), it's going to be tougher and less predictable. So that is not a solution although it's pretty close.  Even if it is mostly balanced, it's hard to say for non 1. 3 abilities. I have a new long range 100% Charm ability, Mind Control, that has a very long charge time, for example. I don't want this ability to end up like another "Meteor" in later portions of the game. Likewise, I wouldn't want this skill to be overpowerd in the early sections of the game either. It's just not the right balance I am looking for.

Thanks for the level cap trick, I do want the level cap to be at most 60 as far as PA and MA growth goes. I do want to get rid of short-charge altogether as well. But if restricting to level 35 can give the right balance, it is worth changing the cap.

[formerdeathcorps]

Well, do you really need speed to be exactly fixed at one value?  Most spellcasters tend to lose power over physical fighters when WP/PA > 12-15, SP > 10, and move > 5 (assuming you don't increase the power of spells scaled for charge time from 1.3).  Keep within those limits, and even if the speed for a ninja grows from 7-10 (while a mage in the same period of time grows from 5-7/8), you're still within balance (unless you apply 1.3's dagger speed bonus, that may be trouble).  Thus, a maximum level cap of 50 probably would still work if C values were 255.

EDIT: If you use a maximum level cap of 60 and a C value of 255, all you guarantee is that every unit class will gain precisely one speed point (except for maybe the original version of calculator) by level 60.

[R999]

If everyone remained with a low base speed, gaining 1 point of speed is such a big difference in relative terms, is it not?  

edit: hmm... I am thinking it might not be a big issue if I scale the multipliers large enough. I'll have to test it out ingame  and see how it would look like. This might turn out to be more trouble than its worth at the moment.

[formerdeathcorps]

Yes, but by level 60, everyone with a class multiplier higher than 70 (standard is 100) would have gained 1 speed point.  Most people with normal speed growth (around 100) gain it in their 30s.  It shouldn't be too bad since you can have enemy bosses with higher class speed multipliers (near 200), effectively guaranteeing that they will always have more base speed than you (and thus be more likely to reach the super-broken range where mages start to fail).

EDIT: Remember, at least in the original and in 1.3, mages are much deadlier in such low speed situations so increasing the speed by 1 really doesn't hamper them much.  A unit with a maximum speed of 8, even with 1.3's additions, can really only get +5 speed and haste, which is 19 speed (the maximum in your game).  That's slower than 5 CT magic.  If you scale your magic around this, you shouldn't have any problems.

[philsov]

I wanted a way to make it so that every character's speed to be fixed at roughly equal to 1 in-game speed stat, with no speed growths at all. And then have their speeds adjusted by their speed multiplier.

To be literally 1?  Modify the pre-raw stats of generic males, females, monsters, and ramza to be 1; (effective raw value of 16384).

At 255 growth and 100 SPM, speed is 1 from levels 1 to 99.  No speed growth.
At 255 growth and 200 SPM, speed is 2 from levels 1 to 99.  No speed growth.

At SPMs between 150 and 190, a speed point is gained somewhere in the level scheme.  (1-2)
At SPMs between 220 and 255, a speed point is gained somewhere in the level scheme.  (2-3)

But the gap between a 2 speed and 1 speed unit is HUGE.

But I never really fully understood how the speed growths are calculated.

There's a long article in the BMG, but really I like 454's spreadsheets.  Use the search tool in the 1.3 forum.

Edit:
Pre-raw value of 2 with 255 growth also shows 0 effective growth at 100 SPM (stays 2 throughout), but at higher SPMs a speed point always crops up.  You could go 50 SPM / 100 SPM with these scheme and get identical results, not that its any different.  At higher raw values speed increases always happen because of the formula that FFT uses.  

Hm.

With a pre-raw value of 3 (and 255 growth), at 96 SPM you see 2 speed until level 12, then 3 all the way through to 99.  68 SPM needs 2 from 1 to 99, and 34 SPM is 1 all the way through.  These gaps are HUGE, once again, but it does provide 3 different speed tiers if you can somehow balance the rest (not likely). At 97+ SPM 4 speed crops up at the end of the level curve.

edit2:

Editing pre-raw stats is pretty easy.  You need a hex editor and a CD extraction tool.  Locations found here:
viewtopic.php?f=8&t=1785

[R999]

philsov, I read your response at least 10 times in order to absorb it properly. Thank you for cranking out those numbers for me, as I had imagined it'll run into problems. 3 speed tiers is not manageable if their relative difference is so large. I'll get back to this after more reading. Once again thanks a lot for taking the time to analyzing this.

[formerdeathcorps]

Philsov, I think he meant to have speed in-game to be fixed at one number, not be 1 (which would be ridiculous and allow all sorts of double turns).

Back to the original issue, R999, I really don't see why you are being so obsessive about keeping the equilibrium between magic and speed.  1 speed growth between units with 5-7 base speed isn't going to ruin the game.  A rough equilibrium is perfectly fine; the magic/physical ratio don't have to be absolutely unchanged from level 1-99 (nor should any gamer expect it to be), as long as it stays within a reasonable range (meaning that a mage team and a physical team both will be needed in battles in every phase of the game).

[philsov]

I'll get back to this after more reading.

Lemme speak broader and simpler then.

When a unit levels up, their previous stat values get combined with the class growth constant in a funky inverse formula.  Basically, it's a seed, and growth that occurs at a very low level carries on through the higher levels.  Even at the lowest possible level of growth, 255, speed gains still occur unless the multiplier is so low that the gain is muted out.  However, this is not an ideal solution as the base speed at level 1 (defaults as 6) is still a rather large seed.  By changing the default values ("pre-raw stats", as this what gives generics 6 speed and 44 HP and X/Y PA/MA/etc) you can make that initial seed smaller and cause zero effective speed change over the course of all levels, which, by itself, is fine and has great balance potential.  

But it is simply mathematically impossible to have different classes with different speeds, while still maintaining the same speed over the course of the game.  IE, Ninja is always 8 speed, squires are always 6, Knights are 5, etcetcetc without some very drastic steps like making everyone level 99 (remove leveling completely) with modified initial stats and then disabling the soldier office.

[R999]

Thanks for the explanations. It is a shame that that isn't possible. I wonder if there is a way to overflow or underflow some of these calculations to fool the game having a growth even higher than 255 in effect. Or even changing the speed growth formula itself via assembly hacking.

Oh, I couldn't find 454's spreadsheets in the 1.3 forum, hm..


edit: So I have tested with FFTPatcher and Philsov's guesses were about right. I tested a generic male with levels 1, 10, 15, 20, 25, 30, 40, 50, 60 using my Test Job with Speed Growth 255. 2 points of speed is gained from level 1-60 with SPM at 150. When SPM at 100, 0 speed is gained for those 60 levels. Just one more thing that is puzzling me... I'll get back on this after more testing is done.

I have also tested with a Growth of 1, have everyone's speed set at 50, and then use the SPM to lower its value manually (say, SPM of 10). It turns out that while the game does cap speed at 50, however, SPM is not applied on this capped number.

[R999]

Update: I think I have found a desirable speed tier that I am comfortable now, here's the data sample I gathered via in-game testing for levels 1-40 (lv1, 5, 10, 15, 20, 25, 30, 35, 40), below are actual in-game speeds with respect to Speed Growth 255 + Default Raw stats. You can see at which level speeds are gained with respect to different SPMs. I have decided to set back the level cap to 40.

The data below is 100% from in-game, not calculated via the formulas to avoid possible roundoff/underflow/overflow calculations.

SPM 100
lv 1 speed 6 -> to -> lv40 speed 7

SPM 140
lv1 speed 8 -> to -> lv20 speed 9 up to 40

SPM 145
lv1 speed 8 -> to -> lv10 speed 9 up to 40

SPM 150
lv1 speed 9 -> to -> lv30 speed 10 up to 40

SPM 155
lv1 speed 9 -> to -> lv25 speed 10 up to 40

SPM 160
lv1 speed 9 -> to -> lv15 speed 10 -> to -> lv40 speed 11

SPM 165
lv1 speed 9 -> to -> lv10 speed 10 -> to -> lv30 speed 11 up to 40

SPM 170
lv1 speed 10 -> to -> lv25 speed 11 up to 40

SPM 175
lv1 speed 10 -> to -> lv15 speed 11 -> to -> level 40 speed 12

SPM 180
lv1 speed 10 -> to -> lv11 speed 11 -> to -> lv30 speed 12

SPM 185
lv1 speed 11 -> to -> lv25 speed 12 up to 40

...

SPM 200
lv1 speed 12 -> to -> lv25 speed 13 up to 40

The lines highlighted in bold are the ones where a total of 1 SP is gained throughout levels 1-40, and it happens exactly at level 25. The first 3 SPMs 155,170,185 gives me 9-10, 10-11, and 11-12 speeds respectively, with the speed gain happen at level 25 (assuming all jobs share 255 Speed Growth). Now I'll just have to test out the right values for the status effects and cast times to get a good balance.

Also something else to note is that if you would to base on this pattern, you'd guess that at SPM-145 the speed gain should also happen at level 25, but that is actually not the case (happens rather at level 10). So be cautious if you are thinking about going similar speed tiers to balance out your patch.

[philsov]

viewtopic.php?f=18&t=1737&hilit=spreadsheet