Originally Posted by
zinnin
I don't agree that the point of view should be 'waiting on Shadow Bolt' to generate fury, it should be more focused on literally how much fury you can dump into a strong cooldown or proc. If I can dump a full fury bar into DS + a trinket (which AD will nearly always allow you to do), then it doesn't matter if another proc happens 20s later, because I just spent the majority of my fury in the highest possible scenario.
Spending Soul Fires over ToC is very much a risk / reward / planning thing. If you do not feel comfortable that those 10s of Meta up time are placed in the 100% correct position then sure, maybe it isn't a good idea to risk the fury there. But during a strong Proc, or when you have a lot of cooldowns up and you are comfortable that there isn't going to be a higher damage window to dump fury, then you should be trying to dump as much fury as possible and not worrying about meta uptime.
Lets say you every point of fury is 1 point of damage. Dark soul modifies it 1.5x, you might have a trinket that modifies it 1.4, and maybe a minor proc that modifies it 1.2. If you can dump 900 fury into DS, that is going to be 1350 damage, compared to maybe only dumping 450 fury into DS, and then 450 into a smaller proc giving you 1305.
Good fury usage shouldn't be measured by meta uptime, but by how the fury was modified when spent. I feel that if a player feels uneasy about spending their fury because they don't know for sure that the situation is optimal, then that is an error in planning.
I've been doing:
Prepot / Precast Soul Fire, Grimoire of Service, Corruption,Doomguard, HoG, life blood / zerking, SB x2 (3 if bloodlust), HoG, DS / Meta, Doom
This generally gets me in meta with DS at the 6th or 7th stack of Black Blood, so I can for sure get a doom with 10s Black blood, and a chance get a 2pc proc before that happens.
You aren't going to have stellar uptime on the opener, but the main thing is to make sure you leave with no fury left and that your doom is put up with every proc possible.