Ah, missed that, sorry. In that case:

1s5r is not sustainable with tri-spec hemo, even with 2/5 T4. A 1-pt SnD lasts 17.4 seconds and thus you regenerate 174 energy while it's up. To do a 1s5r cycle you need to generate 4.8 combo points during this time, plus 20 extra energy for finishers; 4.8 Hemos costs 168 energy, so the total energy cost of the cycle is 188, which is more than the 174 you regenerate. Even 2s5r isn't quite sustainable; it takes 218 energy to perform 5.8 Hemos and cast the two finishers, and there's only time for 217.5 energy to regenerate. So in practice, even with 2/5 T4, the tightest cycle maintainable with tri-spec Hemo is 3s5r.

Hence the 20 energy (free rupture (0 energy) + 20% chance for free SnD (25 * .8 = 20) for "finishers".

Also, given that Ruthlessness only has a 60% chance to proc, saying that you're able to maintain the cycle only if you get a Ruth proc means that you're not really able to maintain the cycle. Just sayin'. :P

That's why the finishers are listed as 20 energy rather than 50. So I still don't see how it's sustainable.

EDIT: Found what I needed.

The 10 expertise thing is a display bug, you only get 5.

Quick note for the OP. On the Self-Buffs section, might want to add [Scroll of Agility V] and [Scroll of Strength V] as they stack with all other buffs (to my knowledge). I can generally find them on the AH for 2-3g a piece.

Hmm...I might need to look at that, as somehow the spreadsheet recommended that cycle. Clearly I must be maintaining it suffering a slow energy bleed and refreshing energy on movement. Most fights seem to require one to either reposition or dodge AOEs, so that might account for it. Having been a dagger rogue running 1s/3r, I'm pretty used to adjusting cycles on the fly, looking at the timing and cutting to a 4-pt Rupture when necessary to maintain 100% SnD uptime even with Hemo with 2pc-T4.

Granted, I can't think of very many fights where I'm allowed to just stand there for any extended period of time, so that must be how it's been working for me. At a certain point you just play. You pick how many points you are dropping for SnD and adjust accordingly by dropping Rupture with as many points as you have in order to not drop SnD. Maybe the problem is the current lack of modeling for 1s/3r and 1s/4r cycles. Something I want to add to the DPS sheet at some point.

But then you get to situations where you are forced to move. For example, if you are forced to move in and out say 5 secs every minute. That's 50 energy extra one could have bleed the last minute of cycle time and still break even. With the example above, you are losing 14 energy roughly every 17.4 secs or bleeding from max 48.276 energy. With the recouped energy where you have been forced to move (or ground slammed or knocked back or whatever), you can actually sustain a lower cycle. High King and whirlwinds, Gruul and Ground Slams, VR and poundings (we started with the dps backing out method), Hydross and the water bubbles, Tidewalker and Water Tombs, Lurker and Geysers all give too many opportunities to refresh energy.

It also seems counterintuitive with the 2pc-T4 bonus which benefits more from shorter cycles given that it adds more overall SnD time, the shorter your cycles. What's funny is with a minor gear switch I did notice the recommended cycle had changed all the way back up to 3s/5r which seemed a big jump for what seemed a minor 12 dps gear upgrade. Either that or one of the fixes in the sheet corrected part of the problem.

I hadn't considered the napkin math, but you are correct that 1s/5r Hemo recommendations don't seem to make a whole lot of sense. Looks like I need to figure out what is happening in the cycle sheets when that is occurring.

--heavy edit--

I'd say that the reason 1s/3r and 1s/4r don't perform as well as might initially be thought is because of the AP scaling of Rupture based on the number of combo points you're using. As Aldriana says in the post below you only gain 4/10/18/21/24% of AP at 1/2/3/4/5 combo points.

I do have a question though - do they actally peform better than xs/5r cycles? If so then why aren't we simply suggesting that they be used instead? Shaker's post seems to say "1s/xr cycles do perform better than xs/5r cycles but not by a lot so we'll suggest using the lower dps cycles for some reason". I would have thought that if a shorter cycle is just as stable and does more damage, even if it's only 1 dps, then it must be better to use it, particularly on fights where you may not have 100% time on your target (Gruul, Hydross, Leo, Vashj, Solarian...)

With regard to 1s/5r Hemo cycles, my personal experience (and I apologise that it's nothing more than that but it may help to explain things) is that they become sustainable for quite a while after a few lucky Ruthlessness procs. You tend then to be sitting at slightly higher base energy levels so when you don't get the proc there is a reserve for you to dip into and sustain the cycle. Of course, the moment you get a bad streak of Ruthlessness not proccing then your cycle falls apart again.

2s/5r, 3s/5r or the 1s/xr cycles are far more stable for Hemo trispec builds.

AP bonus of Rupture is 4/10/18/21/24%, because it's 4/5/6/7/8 ticks, each of which gives 1/2/3/3/3% of your AP. This is why it can make sense to do Ruptures down to 3 points (as you're mostly losing uptime and not damage), but it doesn't make sense to do 2 and 1 point ruptures in most cases.

Yes, I completely agree and that was also my point. One way to absorb it is by running a tighter cycle. If you are running a cycle that bleeds 14 energy every 17.4 secs, you bleed a little over 48 energy a minute. If combat factors take you out of action for a 5 sec period once a minute, you have essentially recouped that energy. This is of course assuming you pop your rupture right before the disruption.

Anyway, we are somewhat off-topic. Although this discussion has given me a thought. It probably wouldn't be too hard to model such disruptions. Check boxes could be added that quantified the disruption in seconds per minute. You'd add 10 energy every second of the disruption and multiply total dps by 1-[disruption]/60 and have a reasonable DPS estimate of any fight with fixed disruptions. The sheet itself would really auto-calculate your new preferred cycle. It's an interesting concept. Now, I'll probably hold off on this for a bit. It would need a separate page marked "Advanced" and maybe instructions on how to model it.

But let's take Gruul for example. Ground Slams are predictable (happen in X amount of time). One could go into one's own combat logs, look for their last auto-attack prior to the slam, find the time to the next attack after. Take a number of samples, derive an average downtime. Check the combat log for the frequency of the Ground Slam. (Downtime)/(Ground Slam frequency) = disruption. Although this might be a bad example given that the Ground Slam/Shatter might have a duration greater than one can absorb. The same concept could be applied to fights where one runs from AOEs. Check the combat log, out of combat 3 secs every 30 seconds, enter 6 secs in the disruption box and enable it, presto, fairly accurate broken fight modeling.