Let's assume fixed 3e's equal spell progressions: NG human (custom Major BL) Trickster [DragMag353p85] ST 1 with Precocious Apprentice (CArc181), Negotiator, & Spell Focus (Conjuration) & Skill Focus (Perform) (later us Hewards Hall [CS] 5k magical location). Make a faustian pacts for pop psychology feats: Apprenticeships (entertainer & philosopher) [DMGII 176]. Not that by the time the "time limit is up on tithing, we gain the ECL to not care
1: ST CL1=1base
Find an elf who is Heartfire Fanner [DragMag314p23] to "inspire you" (Very fluff compatible). Help optimize his build (psy ref is good) with Song of the Heart (ECS60) and words of creation or Focused Performer & Focused Performance (DragMag338 89). Become a Heartfire Fanner 1.
2: ST CL2=1base+1HfF
Borrow an Aquamarine of Spell Extending (PGtF123) and use it to find and get a Metamagic Storm (CM151) magical location. Now you can use your BL feat of Santum Spell. Saving some XP allows an early Nceropolitan so we can take over inspiring ourself. Don't forget to pay your apprentise dues for third level a little early. Get the Otyugh Hole (CS) for Iron Will and Dragonsblood pool magical locations. Using words of creation and inspire greatness ourselves (alone), we can enter NDb 1.
3: ST CL3=1base+1HfF+1BL, NDb CL5=1base+(ST's3)+1BL
We get mild depravity and take Corrupt Arcana [HoH120] so we can prepare from a spellbook to enter UM1. Without Practiced Spellcaster we would have to advance spellthief and we are fine with that since we will max out NDb anyway.
4: ST CL4=1base+1HfF+1BL+1UM, NDb CL6=1base+(ST'4)+1BL if it recalculates (it will by the below regardless)
SC1 requires nothing. A strict interpretation would require not letting a year go by since getting the metamagic storm at 2nd level
5: SC CL8=1base+(NDb's6)+1BL, NDb CL9=1base+(ST's5)+1BL+2SC, ST CL6=1base+1HfF+1BL+1UM+2SC. We do not recalculate NDb CL because if we did, it could bump SC's CL the same logic. If we snuck MST in via a BL feat, it would then bump twice and therefore NDb's CL would bump again,etc so that the CL would grow exponentially. I'll be using the conservative approach to what order to run the calculations in so that it renders lowest CL and is therefore most likely to keep your DM happy.
Unfortunately which class is boosted by UM depends on their CL and since we can only boost 2 of the three, we have to step through each level. I have purposely not taken MST yet, although we could easily now. @UM2 & 3 we boost NDb & SC:
6: SC CL12=1base+(NDb's9)+1UM+2BL, NDb CL14=1base+(ST's6)+4SC+1UM+2BL, ST CL8=1base+1HfF+1UM+3SC+2BL
7: SC CL19=1base+(NDb's14)+2BL+2UM, NDb CL18=1base+(ST's8)+5SC+2BL+2UM, ST CL10=1base+1HfF+1UM+2BL+5SC
ST takes the boost at UM4 and will clearly do so again at UM7:
8: SC CL23=1base+(NDb's18)+2BL+2UM, NDb CL20=1base+(ST's10)+5SC+2BL+2UM, ST CL11=1base+1HfF+2UM+2BL+5SC
9: SC CL26=1base+(NDb's20)+2BL+3UM, NDb CL23=1base+(ST's11)+2BL+6SC+3UM, ST CL12=1base+1HfF+2UM+2BL+6SC
10: SC CL30=1base+(NDb's23)+2BL+4UM, NDb CL26=1base+(ST's12)+2BL+7SC+4UM, ST CL13=1base+1HfF+2UM+2BL+7SC
11: SC CL33=1base+(NDb's26)+2BL+4UM, NDb CL27=1base+(ST's13)+2BL+7SC+4UM, ST CL14=1base+1HfF+3UM+2BL+7SC
12: SC CL36=1base+(NDb's27)+3BL+5UM, NDb CL32=1base+(ST's14)+9SC+3BL+5UM, ST CL17=1base+1HfF+3UM+3BL+9SC
13: SC CL42=1base+(NDb's32)+3BL+6UM, NDb CL37=1base+(ST's17)+3BL+10SC+6UM, ST CL18=1base+1HfF+3UM+3BL+10SC
At this point NDb is maxed out so we'll advance ST:
14: SC CL46=1base+(NDb's37)+3BL+7UM, NDb CL39=1base+(ST's18)+3BL+6UM+11SC, ST CL20=1base+1HfF+4UM+3BL+11SC
We are out of UM, but the text says that if there were to be levels above 10, they not drop spellcasting progression anymore. Legacy Champion time! The first level doesn't advance anything but might force a recalculation:
15: SC CL50=1base+(NDb's39)+3BL+7UM+0LC, NDb CL41=1base+(ST's20)+3BL+6UM+11SC, ST CL20=1base+1HfF+4UM+3BL+11SC+0LC
16: SC CL53=1base+(41NDb's)+3BL+7UM+1LC, NDb CL42=1base+(20ST's)+3BL+6UM+12SC, ST CL22=1base+1HfF+4UM+3BL+12SC+1LC
17: SC CL55=1base+(42NDb's)+3BL+7UM+2LC, NDb CL45=1base+(22ST's)+3BL+6UM+13SC, ST CL24=1base+1HfF+4UM+3BL+13SC+2LC
You'll notice that we are out of SC levels. But we can still gain more for purposes of calculations, just like BL.
18: SC CL59=1base+(45NDb's)+3BL+7UM+3LC, NDb CL48=1base+(24ST's)+3BL+6UM+14SC, ST CL26=1base+1HfF+4UM+3BL+14SC+3LC
19: SC CL63=1base+(48NDb's)+3BL+7UM+4LC, NDb CL51=1base+(26ST's)+3BL+6UM+15SC, ST CL28=1base+1HfF+4UM+3BL+15SC+4LC
20: SC CL67=1base+(51NDb's)+3BL+7UM+5LC, NDb CL54=1base+(28ST's)+3BL+6UM+16SC, ST CL30=1base+1HfF+4UM+3BL+16SC+5LC
Phew. Looking above, it's really now surpise that a PrC CL can get high (especially with BL). Theoretically a PrC could manage almost CL40 that way and this is a little above 60. The progression we see is that SC is what we want to advance (+1). NDb dries up quickly and so we often also boost ST (+1), which also get's the SC's+1 back (total of +2 each level). The NDb get's the SC's +2 and the SC's +1 for a total of +3 each level. The SC get's the NDb's +3 (which is +1SC*2 and the other side boost of +1 to ST) and then it's own original +1 that we mentioned at first for a total of +4 each level. Put in the 16SC boots and 3 BL and you've got that 67 above.
So what about MST? All it really does is take any normally separate arcane CL pools and instead make them include your ST CL. It also makes your ST CL include all your other arcane CLs. Obviously it can't do this recursively as that would give you infinite CL at level 2 (ST CL=1+1 and other CL =1+1 but with recursion it would then be ST CL=2 and other CL = 1+2 so then ST CL = 1+2 and other CL would be 1+3, etc). First let's do the other CL(s) and then we'll do the ST CL, just like above. Note that since the PrC CL's depend on the ST CL, we can't do the ST CL first just like we mentioned originally because it could still lead to infinite CL, similar to the MST recursion example:
20: SC CL97=1base+(51NDb's)+3BL+7UM+5LC+30MST, NDb CL84=1base+(28ST's)+3BL+6UM+16SC+30MST, ST CL151=1base+1HfF+4UM+3BL+16SC+5LC+(97SC's+84NDb's) or ST CL151=1base+1HfF+4UM+3BL+16SC+5LC+(67SC's+54NDb's) if treated as all being simultaneous such that no entry has time to look at how another as changed.
_
In fact, we could have done that for the above and solved: the recursion problem, the "does it recalculate problem" and arrive at more balanced numbers. Let's redo the above if the simultaneous nature prevents referrencing any changes to numbers from the same level:
Level 1and 2 are still CL1,2. Level 3 actually still stays CL3 because BL levels come just before 3, 6 & 12. It was condensed to save space.
We assume UM to also be simultaneous in its advancements 4: ST CL4=1base+1HfF+1BL+1UM, NDb CL5=1base+(ST'3)+1BL
5: SC CL7=1base+(NDb's5)+1BL, NDb CL6=1base+(ST's4)+1BL+0SC, ST CL4=1base+1HfF+1BL+1UM+0SC
6: SC CL10=1base+(NDb's6)+1UM+2BL, NDb CL11=1base+(ST's4)+3SC+1UM+2BL, ST CL8=1base+1HfF+1UM+3SC+2BL
7: SC CL13=1base+(NDb's11)+2BL+2UM, NDb CL17=1base+(ST's8)+4SC+2BL+2UM, ST CL9=1base+1HfF+1UM+2BL+4SC
8: SC CL22=1base+(NDb's17)+2BL+2UM, NDb CL19=1base+(ST's9)+5SC+2BL+2UM, ST CL11=1base+1HfF+2UM+2BL+5SC
9: SC CL25=1base+(NDb's19)+2BL+3UM, NDb CL22=1base+(ST's11)+2BL+5SC+3UM, ST CL11=1base+1HfF+2UM+2BL+5SC
10: SC CL29=1base+(NDb's22)+2BL+4UM, NDb CL24=1base+(ST's11)+2BL+6SC+4UM, ST CL12=1base+1HfF+2UM+2BL+6SC
11: SC CL31=1base+(NDb's24)+2BL+4UM, NDb CL26=1base+(ST's12)+2BL+7SC+4UM, ST CL14=1base+1HfF+3UM+2BL+7SC
12: SC CL35=1base+(NDb's26)+3BL+5UM, NDb CL31=1base+(ST's14)+8SC+3BL+5UM, ST CL16=1base+1HfF+3UM+3BL+8SC
13: SC CL41=1base+(NDb's31)+3BL+6UM, NDb CL35=1base+(ST's16)+3BL+9SC+6UM, ST CL17=1base+1HfF+3UM+3BL+9SC
14: SC CL46=1base+(NDb's35)+3BL+7UM, NDb CL37=1base+(ST's17)+3BL+6UM+10SC, ST CL19=1base+1HfF+4UM+3BL+10SC
15: SC CL48=1base+(NDb's37)+3BL+7UM+0LC, NDb CL40=1base+(ST's19)+3BL+6UM+11SC, ST CL20=1base+1HfF+4UM+3BL+11SC+0LC
You'll notice that the "dead" LC level allows the CL calculations to catch up to almost what it was before and that most of the time it was only off by one to three.
16: SC CL52=1base+(40NDb's)+3BL+7UM+1LC, NDb CL41=1base+(20ST's)+3BL+6UM+11SC, ST CL21=1base+1HfF+4UM+3BL+11SC+1LC
17: SC CL55=1base+(41NDb's)+3BL+7UM+2LC, NDb CL43=1base+(21ST's)+3BL+6UM+12SC, ST CL23=1base+1HfF+4UM+3BL+12SC+2LC
18: SC CL57=1base+(43NDb's)+3BL+7UM+3LC, NDb CL46=1base+(23ST's)+3BL+6UM+13SC, ST CL25=1base+1HfF+4UM+3BL+13SC+3LC
19: SC CL61=1base+(46NDb's)+3BL+7UM+4LC, NDb CL49=1base+(25ST's)+3BL+6UM+14SC, ST CL27=1base+1HfF+4UM+3BL+14SC+4LC
20: SC CL65=1base+(49NDb's)+3BL+7UM+5LC, NDb CL52=1base+(27ST's)+3BL+6UM+15SC, ST CL29=1base+1HfF+4UM+3BL+15SC+5LC
Then MST comes along and can reference the changes for this level, but not during it's own changes: SC CL98=1base+(52NDb's)+3BL+7UM+5LC+(30ST's), NDb CL86=[1base+(30ST's)+3BL+6UM+16SC]+(30ST's), ST CL147=1base+1HfF+4UM+3BL+16SC+5LC+(SC's65+NDb's52). Note that once MST comes along, it probably forces a recalculation to allow that last SC boost to kick in.
You'll also notice that NDb appears to double dip because just like SC, the first ST CL is hidden within the main PrC's CL for its spells. I could be wrong about this, though because it would seem by adding a CL boost, you could "proc" the calculations as many times as you wanted, leap-frogging your way to higher CLs. It then seems we have an arbitrary limit (one proc per bonus) limitation.
This makes me want to remove all multiple instances of ST's CL and redo the calculations. While it would be more conservative to only remove the like terms that show up directly, one could simply alternate CL boosters on the lowest CL classes whilst the total was still below 20. To illustrate:
6a: SC CL10=1base+(NDb's6)+1UM+2BL, NDb CL11=1base+(ST's4)+3SC+1UM+2BL, ST CL8=1base+1HfF+1UM+3SC+2BL
6b: SC CL10=1base+(NDb's11)+1UM+2BL, NDb CL16=1base+(ST's8)+4SC+1UM+2BL, ST CL10=1base+1HfF+1UM+4SC+2BL+1x
6c: SC CL20=1base+(NDb's16)+1UM+2BL, NDb CL19=1base+(ST's10)+4SC+1UM+2BL+1y, ST CL10=1base+1HfF+1UM+4SC+2BL+1x
6d: SC CL23=1base+(NDb's19)+1UM+2BL, NDb CL19=1base+(ST's10)+4SC+1UM+2BL+1y, ST CL11=1base+1HfF+1UM+4SC+2BL+1x+1z
6e: SC CL23=1base+(NDb's19)+1UM+2BL, NDb CL21=1base+(ST's11)+4SC+1UM+2BL+1y+1w, ST CL11=1base+1HfF+1UM+4SC+2BL+1x+1z
As you can see the same sort of CL progressions are happening, which is no surprise given that the ECL progression is just a special case of CL boosting. It's basically a system of equations that settles down to a change close to the input values after a few iterations. This really makes me want to unravel the "other class" values so that we can remove multiple ST or SC stacking. Ug
_
The only thing that changes is NDb which doesn't get the SC boost from inside ST: "equal to his NDb level + his levels in one other spellcasting class." The SC boosts aren't a level in that other class. Oops. Likewise the SC CL boosts to other classes doesn't reapply to the SC itself. Again only other class levels (of 1 class) add back into SC. On yet ANOTHER look, I edited the below (as I did to the above once too) because the SC and NDb CL is NOT dependent on the CL of ST until MST. We want to wait to take MST until after UM7 (ECL11 here). Levels 1-6 don't change:
6: SC CL9=1base+5ST(1base,1HfF,1UM,2BL)+1UM+2BL, NDb CL12=1base+5ST(1base,1HfF,1UM,2BL)+3SC+1UM+2BL, ST CL8=1base+1HfF+1UM+3SC+2BL
7: SC CL10=1base+5ST(1base,1HfF,1UM,2BL)+2BL+2UM, NDb CL14=1base+5ST(1base,1HfF,1UM,2BL)+4SC+2BL+2UM, ST CL9=1base+1HfF+1UM+2BL+4SC
8: SC CL10=1base+5ST(1base,1HfF,1UM,2BL)+2BL+2UM, NDb CL15=1base+5ST(1base,1HfF,1UM,2BL)+5SC+2BL+2UM, ST CL11=1base+1HfF+2UM+2BL+5SC (note a temporary CL booster before the UM level rep-procs ST and SC to be tied so we can choose)
9: SC CL12=1base+6ST(1base,1HfF,2UM,2BL)+2BL+3UM, NDb CL17=1base+6ST(1base,1HfF,2UM,2BL)+2BL+5SC+3UM, ST CL11=1base+1HfF+2UM+2BL+5SC
10: SC CL13=1base+6NDb(1base,2BL,3UM)+2BL+4UM, NDb CL19=1base+6ST(1base,1HfF,2UM,2BL)+2BL+6SC+4UM, ST CL12=1base+1HfF+2UM+2BL+6SC
11: SC CL14=1base+7NDb(1base,2BL,4UM)+2BL+4UM, NDb CL20=1base+6ST(1base,1HfF,2UM,2BL)+2BL+7SC+4UM, ST CL14=1base+1HfF+3UM+2BL+7SC
Now add in MST. We don't care if the ST CL skyrockets (the whole point of the thread) because all UM and over-cap UM levels are full double casting boosts. Do note that I'm taking a very conservative assumption that since MST procs immediately after the CL change from ECL, it doesn't re-proc ECL since it's already proc'd once from that one change. If your DM is nicer about this, MST simply puts the ECL CL calculations back on track rather than looking one level behind. Lastly note that I had to stop myself writing the progression (again) to make sure SC's levels were higher than ST and use it when appropriate for NDb. Looking even hard (is that possible?) I saw another one of my earliest objections: stacking. Calculating NDb off of SC is a bad idea because SC already adds to NDb. They both use generic "plus"/"add" "to"/"in" the casting "class". This sounds too much like a general bonus to your CL which can be used in a CL check. Remember that this kind of "... check ... do(es) not stack if they have the same type or come from the same source." It's kind of hard to argue that the number which is solely determined by your SC level isn't the same source as that other SC level you just added in. In contrast, MST is a bit more of a grey area: "Your ST and arcane spellcaster levels also stack when determining your CL." It seems pretty like specific trumping general to me: these things stack even if they were already adding. I'm aware that normal separate CL pools don't add (this isn't 5e) and that the feat might have just been trying to say this, but it is pretty direct about the stacking. One could argue that perhaps if MST is added before the first SC level, then the addition from SC might not stack. But when MST came afterward (and logically has to proc after the ECL change), it's hard to ignore MST. If your DM has a problem with this middle way interpretation, just play a red wizard and save yourself the agony of calculating all this. That's why we will stick to ST for the chosen NDb class, even though it appears we could get some slightly higher numbers by using SC instead. Lastly, as you'll in the next session NDb is going to quickly become unimportant anyway.
12: SC CL25=1base+8NDb(1base,3BL,4UM)+8MST+3BL+5UM, NDb CL33=1base+8ST(1base,1HfF,3UM,3BL)+8SC+3BL+5UM+8MST, ST CL34=1base+1HfF+3UM+3BL+8SC+9NDb(1base,3BL,5UM)+9SC(1base,3BL,5UM)
13: SC CL27=1base+9NDb(1base,3BL,5UM)+8MST+3BL+6UM, NDb CL35=1base+8ST(1base,1HfF,3UM,3BL)+3BL+9SC+6UM+8MST, ST CL37=1base+1HfF+3UM+3BL+9SC
+10NDb(1base,3BL,6UM)+10SC(1base,3BL,6UM)+10SC(1base,3BL,6UM)
14: SC CL29=1base+10NDb(1base,3BL,6UM)+8MST+3BL+7UM, NDb CL36=1base+8ST(1base,1HfF,3UM,3BL)+3BL+6UM+10SC+8MST, ST CL41=1base+1HfF+4UM+3BL+10SC+11NDb(1base,3BL,7UM)+11SC(1base,3BL,7UM)
15: SC CL30=1base+10NDb(1base,3BL,6UM)+3BL+7UM+9MST+0LC, NDb CL39=1base+9ST(1base,1HfF,4UM,3BL)+3BL+6UM+11SC+9MST, ST CL42=1base+1HfF+4UM+3BL+11SC+0LC+11NDb(1base,3BL,7UM)+11SC(1base,3BL,7UM)
Remember that MST procs after ST's CL due to ECL change has been calculated:
16: SC CL32=1base+10NDb(1base,3BL,6UM)+3BL+7UM+1LC+10MST, NDb CL40=1base+9ST(1base,1HfF,4UM,3BL)+3BL+6UM+11SC+10MST, ST CL44=1base+1HfF+4UM+3BL+11SC+1LC+11NDb(1base,3BL,7UM)+12SC(1base,3BL,7UM,1LC)
17: SC CL34=1base+10ST(1base,1HfF,4UM,3BL,1LC)+3BL+7UM+2LC+11MST, NDb CL43=1base+10ST(1base,1HfF,4UM,3BL,1LC)+3BL+6UM+12SC+11MST, ST CL47=1base+1HfF+4UM+3BL+12SC+2LC+11NDb(1base,3BL,7UM)+13SC(1base,3BL,7UM,2LC)
18: SC CL37=1base+11ST(1base,1HfF,4UM,3BL,2LC)+3BL+7UM+3LC+12MST, NDb CL34=1base+11ST(1base,1HfF,4UM,3BL,2LC)+3BL+6UM+13SC+12MST, ST CL50=1base+1HfF+4UM+3BL+13SC+3LC+11NDb(1base,3BL,7UM)+14SC(1base,3BL,7UM,3LC)
19: SC CL40=1base+12ST(1base,1HfF,4UM,3BL,3LC)+3BL+7UM+4LC+13MST, NDb CL36=1base+12ST(1base,1HfF,4UM,3BL,3LC)+3BL+6UM+14SC+13MST, ST CL53=1base+1HfF+4UM+3BL+14SC+4LC+11NDb(1base,3BL,7UM)+15SC(1base,3BL,7UM,4LC)
20: SC CL43=1base+13ST(1base,1HfF,4UM,3BL,4LC)+3BL+7UM+5LC+14MST, NDb CL38=1base+13ST(1base,1HfF,4UM,3BL,4LC)+3BL+6UM+15SC+14MST, ST CL56=1base+1HfF+4UM+3BL+15SC+5LC+11NDb(1base,3BL,7UM)+16SC(1base,3BL,7UM,5LC)
That's a big difference: It reminds me of simple vs compound interest. You'll notice that another ioun stone or whatever will reproc the ECL calculations to adjust that +4LC to +5LC for +1 to NDb and SC. These are finally about the numbers I was expecting. Note that I removed another set of calculations for MST at the end. It turns out it wasn't needed. Concerning balance, I don't find an -2 to +16 CL off from CL40 in a campaign of circle magic users that powerful, even if it breaks the normal cap RAW or your DM might to expect. Note that the above is the fastest progression or "worst case scenario" for fast caster stacking without going full 999 limberger. It's still not the most powerful in total, though.
