Thank you for your spot on excellent comments!
Well I was going for an approach of changing as little as possible. Do you have some better suggestions as to what to do? Because I would like the '+3' weapons to be good but I am in favor of changing as little as possible. (Honestly, I think there is a lot that could get changed in the weapons but I will defer to the RAW as much as humanly possible here)
You could simply say something like, all +3 proficiency weapons get +1 to their damage rolls per [W]. That about fits with the existing system where +2 prof weapons tend to have about 1 point higher [W] than equivalent +3 prof weapons.
You are absolutely right, shields are not right. My first thought was that light shields can act as cover and heavy shields can act as superior cover. But then I thought, superior cover would be too good. So then I thought, I could make them both cover, but then what is the point of a heavy shield and then I thought, well perhaps total defense action could benefit. You are right though, why give total 'd' a benefit? It is nearly never used. What about my initial thought, was I right that -5 would be too powerful?
It seems like the problem you're having is that having a shield gives you a "passive" penalty (not having a 2-handed weapon, or having to spend a feat to use it) that is on all the time, while the benefit it gives you is "active" - you have to reduce your attack or take total defense in order to use its benefit. And taking a penalty to attack to get a bonus to defense is usually disadvantageous, because you can direct attacks to where they are most needed, while if you increase your defense enemies can (usually) attack someone else.
Essentially under your system the choices are (1) have a two-handed weapon, which gives you a better attack all the time with no downside, or (2) have a weapon and shield, which gives you an option to reduce your attack to increase defense - an option which is usually not advantageous, and even when it is it has a downside.
Maybe you should make it just like armor: say a light/heavy shield gives +1/+2 HP per tier. You could also say it gives +1/+2 AC when you take a total defense or second wind - that would make it more useful because second wind is used muh more often than total defense.
And I also was thinking about this, I want to use the 5 levels of armor they have developed and not go with just light and heavy (even though the more coarse gradation would be better). I was thinking if the first feat gets you 2 HP per tier and then all the others basically add +1 per tier that is a lame feat. I thought of increasing them to +2, so the five types would give 2/4/6/8/10 which could be over-kill. Plate armor at +10 per tier definitely kicks butt.
I think to implement this idea correctly armor has to be HP bonus or Resist "damage". Since hitting chance is static, the only thing that offsets your ability to defend is the ability to take more damage in combat- not how well you hit. So, it is the right idea I think.
Another issue to think about is that the way it currently works, classes with only light armor proficiency tend to be high in dex or int, which offsets their low AC. Since in your system dex and int no longer help AC, this nerfs light armor users vis-a-vis heavy armor users. Of course if that is what you want then so be it.
Resist all damage is probably not a good idea. If effectively everyone had resist all, that would affect lots of monster powers - for example ongoing damage zones, and other powers which do a little bit of damage over and over again, would be a lot less effective. Increasing HP would probably be better.
This is very true also, like I said spot on. I ask though was it supposed to be that way? When they sat down did they intend every attack to virtually be the same chances to hit? If that is true then *all* attacks should hit on a 10+ which would further streamline and make the system that much more elegant.
I only ask about the designer intent vs. the actual implementation because my gut hunch is that other defense attacks were meant to be easier to hit than AC. Most monsters have a low other defense and this assumes you are targeting a lower defense to take advantage of their weaknesses. So, the 8+ sort of takes the, "Hey, maybe they hit their low defense" and makes that way all the time. Perhaps this makes the attack other defense powers more powerful than initially conceived, I don't know. Opinion?
There are effectively three major types of attacks in the game:
(1) Weapon attacks that target AC.
(2) Implement attacks that target other defenses.
(3) Weapon attacks that target other defenses.
AFAIK, there are no implement attacks that target AC.
Now, weapon attacks tend to be 2-3 points higher than implement attacks because of the proficiency bonus, but AC tends to be 2-3 points higher than other defenses. Thus in the existing system, attacks of type (1) are balanced with attacks of type (2) because the two effects cancel out, but attacks of type (3) are better. So if you wanted to replicate that you could say that (1) and (2) type attacks hit on 10+, while (3) type attacks hit on 8+.
There's also a third type of attack: attacks that don't have a weapon or implement, but do have a tier-scaled bonus, such as a Dragonborn's dragon breath. In this case the tier scaled bonus substitutes for the implement enhancement bonus, so it's effectively like an implement attack, and can be treated as such.
Another thing is combat maneuvers like bull rush and grab. These are not weapons or implements, but have no tier-scaled bonus to replace the implement bonus, making them very hard to hit with at higher levels. This is probably not an intended feature of the game, so you can fix that by treating those like implement attacks in the system above, so they would hit on 10+. However that would provide an unusual situation where the 8 strength wizard would be just as effective at grabbing opponents as the 20 strength fighter.