#1 - Feb. 1, 2017, 12:42 p.m.
A good candidate for it to compare to is the Manald heal, as they are both proc-based elemental themed damage (and weapon slot and jewelry slot are overall about as precious as each other).
-------------------------------------------------
Wins for Manald heal:
1. Damage Number: 13000% ~ 14000% vs 10000%, slight win (~1.35x).
2. Proc of Status: Paralysis is always 15%, independent of spell proc coefficient or monster cc resistance (ignore Juggernauts as well), while frozen depends on the spell or effects, often depends on the proc coefficient, and always affected by monster cc resistance works very bad on Bosses and does not work on Juggernauts. Big win (estimate at least 2.0x).
3. Proc of damage: Manald heal proc 100% once paralysis is triggered, while Rimeheart still depends on spell proc coefficient and needs to multiply it by 10% (or 20% if using the new value). As most fast attacking or large AoE spells have minimal proc coefficient, I would call this a huge win (estimated at least 20x for the old value).
Wins for Rimeheart:
1. No class restriction and does not need specific passive. I would call this only a minimal win (~1.1x).
Can't think of any other good on Rimeheart.
-----------------------------------------------------------------
So according to the estimate Rimeheart would need 1.35 * 2.0 * 20 / 1.1 around 50 x buff to be comparable with Manald Heal, and this may even be an underestimate. Therefore, buffing the proc chance by 2.0 is... I wouldn't even bother to do so.
The big thing is really removing the monster cc resistance dependence and spell proc coefficient dependence, making the status proc whenever a spell with frozen hits.
Haven't really had time to test it in PTR, correct me if they changed the proc mechanisms without having in the patch notes.