Tried to post this before and the system went wild, so if this double posts (or triples) my apologies.
I had heard about Grim n Gritty before, and just now read Ken's stuff. I like it, I think, having not seen it in action, but I have a question:
One of the stated objectives of the system is to make dodging good, armor better, and cover best. However, I think that as the numbers with D&D 3.5 cover stand now, anything less than total cover is not very helpful. Mayhaps you meant total cover, but I would think that even 9/10 cover ought to be very helpful, don't you? The main reason that I think it is ineffective concerns Area of Effect spells or weapons. I realize that you've said the system was not designed with a specific magic/psionics system in mind, but I think that ANY AoE attack overpowers 3/4 or 9/10 cover a little too easily, be it a fireball, grenade, or a sack of acidic flour or something
(I don't have my books with me here, and I cannot seem to find cover on the systems reference document, so I'm working off of memory) If I've got 9/10 cover, it provides me with something like +7 to AC (my defense roll) and +4 to Reflex saves. All AoE attacks I have heard of do NOT care about AC, just reflex saves. In such cases, cover is about getting behind the cover quick enough to dodge the explosion. My suggestion is to give cover some sort of soak bonus. Whether this soak should be in addition to its defense roll or instead of the defense roll I don't know, but given that this system goes for realism, the way cover works is misrepresented a little, I think.
The big problem in my head is that this aspect of cover pretty much just includes AoE attacks. Against other attacks it still makes sense. The two options in my head, then, are to give cover a soak bonus to just AoE attacks, which seems a little clunky, or give a soak bonus to all attacks, which seems to stretch it a little.