Whether from the desire to create a new spell or from dissatisfaction with a lack of power, it is certain that magic-users will eventually get around to inquiring about the procedures necessary to research and create a spell. Whether the spell is new or a listed spell which the character has been unable to locate during the course of his or her adventuring, the process is nearly the same.
First, you must determine two salient facts: Is the character of sufficient Intelligence to “know” another spell of that level? Is the desired spell of a level which the character’s ability allows? Success demands positive replies to both questions.
Determination Of Spell level. When desire to research a new spell is stated, the player must carefully draft the details of the spell, i.e., in the format found in section on Spell Descriptions. The DM will use this to compare and contrast it with and to existing spells in order to determine its level and any modifications and additions necessary in order to have it conform to “known” magic principles.
Any super-powerful, absolute killer, or like spell will be thrown out immediately. The DM should be certain to consider all of the ramifications of a spell and all of the descriptive wording, to weed out spells which give the user some extraordinary, undeserved, and/or unwarranted power. Anything better than the 9th level spells (7th level with respect to clerics) is in the category of impossible. Research will be allowed, but it will always be useless.
The DM will check the range, duration, area of effect, and saving throw in particular to see that the spell has limits and restrictions. The DM may choose to place more rigid restrictions than are shown. Similarly components may be added and/or casting time lengthened. After analysis of the spell and adjustment of its parameters, it will be possible to assign it a level. If the spell is a variation of an existing spell, with only minor differences, improvements, or extended effects, it will probably be only one level higher. If it is superior in two categories, this will place it two levels higher, and so forth. If it is a spell which is markedly different from all others, the DM will find the spell or spells which most closely resemble it, and then set its level according to the relative power of those covered in the book compared to the new spell. Spells with combination functions are at least equal in level to the sum of the spell functions plus a level, i.e., a spell which combined audible glamer with phantasmal force would be 6th level (a 2nd level plus a 3rd level plus 1 level for combining spells).
Cost Of Research. The basic cost for spell research is only 200 gold pieces per spell level per week. Note, however, that this assumes that the researcher has a laboratory (or shrine in the case of a cleric) and library at hand (built up in gaining levels of experience and whatever other reasons are applicable). To the base cost must be added a weekly variable of 100 to 400 gold pieces per level of the spell, the variable accounting for additional materials needed. If no library of materials is owned by the researcher, the base cost increases by a factor of 10 (2,000 gold pieces per spell level per week), the researcher being assumed to be acquiring arcane texts and scrolls, thus creating a library. The player must establish the amount to be spent, based on his or her appraisal of the level, without assistance from the DM (who will then confirm after time, whether this was enough)!
Length Of Research. Initial preparations and research will consume a number of weeks equal to the level of the spell being researched. After this initial period, there exists a chance to achieve success, the chance being given weekly. Research therefore will always extend to a number of weeks equal to the level of the spell being researched plus 1, i.e., a 1st level spell will take a minimum of 2 weeks research. Research expense accrues each and every week. There is no limit to the extension of research, although practically, even minimal expenditure will result in successful research after the basic period plus 6 or so weeks on the average.
Conditions Under Which Spell Research Is Possible. It is absolutely mandatory for the researcher to be of sound mind and body and to have privacy and seclusion free from interruption during the course of his or her spell study. This necessity precludes any adventuring or general interaction during the period of research. It requires about 8 hours per day of work, and only an hour or two per day can be spent doing other things - instructing underlings, disbursing funds, etc. Any interruption of research will be a setback. Each day of such hiatus will cause 1 full week of lost time. Also, as noted above, a library must be at hand, and this means that either the researcher must have his or her own stronghold or highly private quarters, or obtain them prior to beginning the project. If a personal library is not possessed, the research must be carried on in or within one day’s journey of a town or city in which the research materials can be obtained.
Research Materials. Acquisition of materials includes not only texts and scrolls, but also various components for the material needs of the spell. Written works cover the whole spectrum of knowledge, as the researcher must be aware of any and all aspects of the magic he or she wishes to use. Thus, works on history, geography, astrology, alchemy, etc. must be obtained.
Chance Of Success. The base chance of success is 10%, plus the Intelligence score of the researcher (Wisdom with respect to clerics), plus the level of the character researching the spell, minus twice the level of the spell being researched:
10 + Int/Wis Score + Level - (2 x spell level) = Base %
This base chance is modified upwards in 10% increments by doubling of maximum base cost per level of the spell (2,000 gold pieces). Thus, an expenditure of base cost + variable + 2,000 g.p. on a per spell level basis increases base chance from 10% to 20%, and 4,000 g.p. additional per spell level increases base chance from 10% to 30%, and so on, to a maximum base chance of 50% (expenditure of 8,000 gold pieces additional per spell level). Insufficient expenditure means a 0% success chance.
Determination Of Success. After the initial research period (equal to the level of the spell in weeks), the DM will determine the base chance, adjust by Intelligence, character level, and spell level, and then secretly roll the percentile dice. Any score greater than the percentage of success chance indicates a negative result. The DM will continue to check for success each successive game week as long as research is uninterrupted and proper funding for the work continues. Remember, if a spell that is impossible for the character is being researched, this procedure remains constant, but there will NEVER be success, and the dice roll is simply frosting. (Eventually, even the most determined and least wise will begin to understand their aim is fruitless.)
Confidentiality Of New Spells. Once a player succeeds in researching a new spell, it is his or her character’s alone. Only that character has the proper formula, and no mention of the spell should be made by the DM to other players.
Additional Notes. There is no way that costs in money or time can be reduced. Likewise, conditions must be as set forth. Research materials must be obtained. It is possible to use another’s facilities, but this will most certainly not reduce costs.
Gathering a library assumes that the character is in a metropolitan area with libraries, booksellers, etc. from which copies can be obtained. If such is necessary, this must be done prior to the beginning of actual research, as it will take about one week per spell level to gather sufficient materials - the greater the level of the spell, the more esoteric and rare the tomes needed.