Steve Jackson Games - Site Navigation
Home General Info Follow Us Search Illuminator Store Forums What's New Other Games Ogre GURPS Munchkin Our Games: Home

Go Back   Steve Jackson Games Forums > Roleplaying > The Fantasy Trip > The Fantasy Trip: House Rules

Reply
 
Thread Tools Display Modes
Old 03-22-2022, 01:10 PM   #61
Bill_in_IN
 
Join Date: Dec 2021
Location: Indiana
Default Re: Alternate XP progression schedule

Quote:
Originally Posted by Steve Plambeck View Post
Untested -- just one of the models I came up with after reading this thread the first time. Figured something slower than original ITL, but faster than Legacy ITL, might be an appealing compromise for some

Personally I don't like the slower start this formula would impose on new characters, but once over 36 points it will move much faster than Legacy's rules (which strike me as abysmally slow).
Thanks. I'm running a data analysis on EXP per Attribute Point Cost between Classic TFT, Legacy TFT, and three different House Rules mentioned in this thread. I will summarize later.
Bill_in_IN is offline   Reply With Quote
Old 03-22-2022, 02:50 PM   #62
hcobb
 
hcobb's Avatar
 
Join Date: Aug 2004
Location: Pacheco, California
Default Re: Alternate XP progression schedule

The new attribute points are much more useful than the old ones, except for ST which has been reduced in value by at least a point.
__________________
-HJC
hcobb is offline   Reply With Quote
Old 03-28-2022, 07:00 PM   #63
David Bofinger
 
Join Date: Jan 2018
Location: Sydney, Australia
Default Re: Alternate XP progression schedule

The great unexamined idea in experience is exponential increase. Every now and then, whether it's every attribute point or every five, the cost to go up doubles.

It has all sorts of consequences I think are undesirable. Characters rush through the early levels, hardly getting to experience 32 points before it's gone for ever. But the higher levels they sit at for ages, their character development stalled, dreaming of new abilities they won't see for a long time. This can't, to my thinking, be desirable.

Yet basically every time I hear people say, "I have a new XP system," it turns out to be exponential. There's generally no argument why it should be, it's just assumed this is how it should be, that all experience systems are exponential. Legacy sort of has one, that character development is less important later because people have come to love the campaign, but I think enjoying a campaign (during adventures) and enjoying character development (between adventuring) are basically separable activities. People enjoy both, and should get both, rather than being offered first one, then the other.

Why do people use exponential increase? As far as I can tell, just because everyone else does. (In particular because D&D's original edition did, which is basically what everyone is copying. But original D&D assumed the XPs earnt would also increase rapidly, which according to Legacy guidelines-as-written isn't the case. Even in D&D, when they use a "milestone" experience system, progress is more or less at a constant rate.)

My way of thinking about this:
  • Take the number of sessions you want a PC to sit at a level. (Never mind NPCs, they'll take what they're given and like it.) I think this should, over the development of a character, be more or less constant. How long the campaign is supposed to be will determine what the constant is.
  • Multiply by the number of XPs per session. GAW suggests constant. You can vary that if you like but I don't see the point.
  • There'll probably be a maximum level you want it to be hard to exceed, so as you approach this point you can make it much more expensive to progress (e.g. doubling every point).
The net effect is that experience cost to progress should be more or less constant for a long time, then go up fast as it approaches the maximum desirable character level.

I appeal to game designers to consider alternatives to exponentiation, and to ask themselves what they want to happen before they touch pen to paper.
David Bofinger is offline   Reply With Quote
Old 03-28-2022, 07:41 PM   #64
phiwum
 
Join Date: Jun 2008
Location: Boston area
Default Re: Alternate XP progression schedule

I think the tendency to exponential growth, at least at higher levels, is easily explained.

I want my player's characters to advance, but I don't want a bunch of supermen running around. The campaign is open-ended, so a planned advancement as you suggest is impractical. If they advance quickly for a while and then the XP costs go to exponential (except for new talents in TFT, of course), then older characters will be a lot better than newer characters, but they won't be invincible.

Now, it might work out poorly once they reach 40 attribute points and thereafter gain literally every talent available at their IQ, but I suppose we'll see just how likely that is. At 500XP per point, talents aren't exactly cheap.
phiwum is offline   Reply With Quote
Old 03-28-2022, 08:26 PM   #65
hcobb
 
hcobb's Avatar
 
Join Date: Aug 2004
Location: Pacheco, California
Default Re: Alternate XP progression schedule

Replace exponential with quadratic.
The first added attribute costs 1k, the second 2k, the third 3k, and so on.
__________________
-HJC
hcobb is offline   Reply With Quote
Old 03-29-2022, 07:55 AM   #66
phiwum
 
Join Date: Jun 2008
Location: Boston area
Default Re: Alternate XP progression schedule

That's linear, not quadratic. Perhaps you meant 4k for the third point? If so, the next point would cost 7000 pts.
phiwum is offline   Reply With Quote
Old 03-29-2022, 08:00 AM   #67
tomc
 
tomc's Avatar
 
Join Date: Aug 2004
Location: Carrboro, NC
Default Re: Alternate XP progression schedule

Fibonacci for the win!
__________________
OgreMap2

Freedom of Speech is not Freedom of Podium
tomc is offline   Reply With Quote
Old 03-29-2022, 10:47 AM   #68
hcobb
 
hcobb's Avatar
 
Join Date: Aug 2004
Location: Pacheco, California
Default Re: Alternate XP progression schedule

Quote:
Originally Posted by phiwum View Post
That's linear, not quadratic. Perhaps you meant 4k for the third point? If so, the next point would cost 7000 pts.
Differences between quadratic values is linear.
1+3 = 4, 4+5 = 9, 9 + 7 = 16, etc.


So total number of attribute points scales as the sqrt of total XP.


Current system scales as log total XP.
__________________
-HJC
hcobb is offline   Reply With Quote
Old 03-29-2022, 02:31 PM   #69
phiwum
 
Join Date: Jun 2008
Location: Boston area
Default Re: Alternate XP progression schedule

Quote:
Originally Posted by hcobb View Post
Differences between quadratic values is linear.
1+3 = 4, 4+5 = 9, 9 + 7 = 16, etc.


So total number of attribute points scales as the sqrt of total XP.


Current system scales as log total XP.
I don't understand at all.

Let's take the current system, which doubles for each new point after 40. Thus, the progression is

4000, 8000, 16000, 32000,...

or, in other words,

4000 * 2^0, 4000 * 2^1, 4000 * 2^2, 4000 * 2^3,....

The total number of XP spent on attribute points is thus

4000 * (2^1 - 1), 4000 * (2^2 - 1), 4000 * (2^3 - 1), 4000 * (2^4 - 1),...

This is what is meant by exponential growth, whether we mean the amount required for the next attribute point or the sum total to get to the next attribute point.

Accordingly, using RAW and assuming that all XP is spent on attributes to simplify matters,

AttrTotal = ceiling(39 + log(XPtotal - 4300))

where 4300 is the sum needed to reach 39 attribute points.

Quadratic growth means that the growth is like a quadratic. Clearly, 1000, 2000, 3000, ... is not quadratic growth. There is no ellipse containing the points (40,1000), (41,2000), (42,3000). Those points are collinear.

Per this rule, the sums required would be

1000, 3000, 6000, 10000, ...

or 1000 * sum_{1 \leq i \leq n} i

I don't see how to relate attribute total to XP in terms of sqrts here. As I wrote below, this was just a momentary memory loss. 1 + ... + n = n (n + 1) / 2. This is quite a famous little theorem, one with which I am familiar, but I totally blanked today.

So, let's suppose that we were talking about actual quadratic growth. In that case, our sequence might be something like (returning to 4000 as the 40th point)
4000, 8000, 16000, 28000, ....
where the nth point is given by 4000 * (n^2/2 + n/2 + 1).

The total experience (after reaching 39) needed to reach a point is
4000, 12000, 28000, 56000, ....
or, dividing by 4000, we have the sequence
1, 3, 7, 14, 25, 41, ....
I don't know much about this sequence. It is not quadratic (that is, there are no quadratic functions which match these values). Given this, I doubt you're right that attribute value scales like the sqrt function as XP increases.

It's totally possible, mind you, that I just am missing something. It's clear that the sequence you first put up can't be called quadratic -- not in any sense I know, anyway. But maybe there is some growth related to sqrt. I'm just not seeing it anywhere.

Anyway, let me know if I am indeed missing something.

Last edited by phiwum; 03-29-2022 at 08:06 PM.
phiwum is offline   Reply With Quote
Old 03-29-2022, 04:23 PM   #70
hcobb
 
hcobb's Avatar
 
Join Date: Aug 2004
Location: Pacheco, California
Default Re: Alternate XP progression schedule

Under my scheme:
32 point character: zero XP
36 point character: 10k XP
40 point character: 36k XP
48 point character: 136k XP

So doubling the number of added points increases the total number of XPs by about four times. Hence quadratic. Under the current system adding two attribute points increases the total cost by about four times, hence exponential.
__________________
-HJC
hcobb is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Fnords are Off
[IMG] code is Off
HTML code is Off

Forum Jump


All times are GMT -6. The time now is 04:33 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2024, vBulletin Solutions, Inc.