EDIT: Upon further thought, I'm not sure that the curation rewards are affected (they are if curation rewards AREN'T given for comments; they are NOT affected if they are -- which, I believe is the case). If curation rewards aren't affected, it would have the effect of making the impact on the post author pool larger -- more like 16.5% (or 1 in 6)

If you look at the front page of SteemDB, it shows how MVests were distributed over the past 30 days. The current ratio (over 12 days of HF19 and 18 days of not) is

70.7% post author rewards
14.2% curation rewards
6.6% comment author rewards

8.5% interest and witnesses (this will remain constant)

Before HF19, comment author rewards were 1.5%. For the last 12 days, it has been 14%. That extra 12.5% percent comes out of post author rewards and curation rewards so likely
before HF19, the ratio was 75/15/1.5/8.5
AND
after 30 days, the ratio will be 64.5/13/14/8.5

WOW! I hadn't thought of it that way before . . . . BUT . . . .
==> The drain on both author & curation rewards is about 14%
which makes since the addition 12.5% needs to come out of 90% of the original pool.