Thank you everyone, I really do appreciate your answers. Making a game is neither a simple, nor an easy process, and the last thing I want to do is go through all the time and effort only to have alienated my entire market.

So far, here's what I have been getting (on average), let's see if my head is straight

On the topic of dice, most people don't, precisely, mind a whole handful of dice, assuming it is the same kind of dice, or you don't have to roll them often.

Die size for a given action is decided at character creation, and will almost never vary from there. Granted, different archetypes have different die sizes for different actions, but then you'd expect a fighter to be able to swing a sword better than a wizard. Also, you will NEVER 'mix' dice in any situation. So if you are rolling multiple dice, they will all be the same kind of dice.

The 'default' setting is Steampunk, with both Magic and Technology. Gunpowder weapons exist, but are both expensive and chancy things (poorly crafted firearms have a chance to explode, bad powder has a significant chance to fizzle). They are, however, very powerful. Steam technology also exists, although without the Bessemer Process for making molded steel, steam engines are rare. Consider the early Napoleonic era for a rough translation. Magic also definitely exists as well. While it may not necessarily be as unbalancingly powerful as magic in 3.5 (where full caster > everything else), it is also quite potent in skilled hands. Anyone who can find a willing teacher can learn some, but it takes dedication to truly master.

However, the masters of each tend to greatly distrust one another, therefore it is extremely difficult to learn, much less master, both.

This also lets GMs decide "Okay, no tech in this game, we're in an area of the world that is big on 'pro-magic' and 'anti-tech', and owning any tech will get you lynched, because I want to run a more fantasy game this time. In fact, you've never heard of 'technology', you just know that some things are evil". By that same token, they can also say "Okay, this game is going to be similar to a Shadowrun game, with very little magic, which is generally seen as either superstitious or evil" and run a more tech-heavy steampunk game.

On the subject of Multiple Stat Dependency, allow me to clarify a few things:

* There are six stats total. Each class will absolutely require one or two, then have a significant benefit from one or two, then have one or two that doesn't come up often, but might be a weakness that could be exploited if it was too low. An 'average' character is perfectly playable. Being a 'big, dumb Thog' character is perfectly playable. His skills might go up a bit slower if his intelligence is so low that he has trouble learning how to use it, but he'll be dishing out some fairly obnoxious damage in melee.

* There will never be a way for a different stat to govern the same action. For example, in D&D 3.5, Wisdom governs Will saves. However, there are feats which allow Will saves to be based off of other stats. That won't happen here.

*There is a 'hard cap' on stats. The driving force in the game are skills. Characters are heroic compared to NPC's, however a Character's true power lies not in his physical and mental abilities, but in how he applies them in a given situation. There is a mechanic for improving stats, but it is very expensive in terms of character points. In short, we're showcasing the characters skills, not how 'bad arse' they are.

On the Class vs Classless System front, we're pretty much split down the middle, with a general consensus that if I'm going to do Class-Based, then for goodness sakes at least keep them flexible enough so a player can make the character they want to.

Any character can learn any skill in the game. However, if it is not a primary or secondary skill for the archetype you are playing,the die size is smaller on skill checks related to that skill. There are also 16 archetypes in 4 categories, so you have a lot of 'hybrid' archetypes to allow you the flexibility to do what you want. I tried real hard to make sure there was 'tank', 'skillmonkey', 'healbot', and 'utility' in several different flavors, each with their own focus.

It is also rather difficult to create a true 'one-trick pony' due to the cost of skills increasing in point costs as you advance in skill. While it is by no means impossible, it probably won't be as effective outside it's one trick than any given character. For example, a person who wanted to be the best swordsman in the world can do that, eventually. He wouldn't be able to use an axe, or a spear, or a staff... and probably doesn't have too many points in anything that isn't 'kill people with a sword' or maybe 'don't get hit while hitting them with a sword', but damn, with that sword, he's the best in the world. Good luck being effective in anything other than in combat with your sword, but hey... that's the trade-off, right?

On the topic of worldbook + Mechanics book, I think the concensus here is that I should make the mechanics, with some flavor behind it. Then, later on, release a worldbook that takes the flavor and turns it into a full five-course banquet. I'll be giving examples in the book in the 'flavor world', which is but thinly sketched out, and gives an idea of the 'default setting' for the game.

Enough people have asked about the publish-on-demand that I should address this.

I despise, with a passion, 'vanity' publishing, or 'publish on demand'. They are in it to make themselves money. If you don't publish a lot of books, then they make their money with a higher percentage per book. This comes in the way of fees to the author and significantly hiking up the price tag on the other end as well. The 40-50 dollar price was if I could find a decent game publisher (like, for example, Victorious Press) and get them interested. Print-On-Demand would increase these by AT LEAST 25%.

Also, I may well simply toss my hands up at trying to make any money off of this, and set it up as an SRD, much like the d20 SRD website.

As a happy medium, how about a .pdf which is formatted specifically to be printed out and put in a three ring binder? Perhaps have a smaller 'travel version' which only has the rules pertinent to actually playing (thus, would not have character creation rules or 'flavor') which can be printed out, stapled, and folded in half as a booklet?

Due to overwhelming majority response, there will NOT be any form of DRM. I was just calculating how many hours I've spent on this project already, and how many more are left before it is publishable... and didn't want all that completely bypassed by some jerk and some upload site.