In fact, I still think I quoted all the relevant part to answer the question (remember it was How much energy for jump?).
Jump (nominally) requires EP
AND FUEL 
... so when you cherry pick your way into asserting that "fuel is irrelevant" you open yourself up to skeptical responses from people who know better.
The question you should have been asking is ... "why is fuel necessary?" ... rather than taking the stance that fuel consumption was irrelevant.
The real bottleneck is the jump capacitors, since they have a clearly defined capacity limit (36 EP per ton of jump capacitors) and they require specific minimum of power pre-loaded into them in order to perform their role properly.
For convenience of analogy, I'm thinking of the EP (not fuel) requirement as being akin to "pre-heating the battery" on a Tesla groundcar prior to supercharging ... which is done to "condition" the battery for optimal power transfer rates that won't damage the engineering. I'm sure that at least a few of our forum posting regulars here have at least a passing familiarity with how supercharging works, even if you don't own a Tesla yourself (you probably know someone who owns one and has done it, etc.).
The numbers work out to being 18 EP of max capacity per 100 tons of starship per parsec ... but a starship's power plant only needs to provide 2 EP of output per 100 tons of starship per parsec over 2 combat turns in order to ENABLE a jump to be attempted. This then leaves 18-2=16 EP of "unused headroom" in capacity, which seems like an oddly LARGE MARGIN if it's just there for "safety" as opposed to being something that gets routinely used.
The obvious implication (is obvious) then becomes that the 16 EP of "unused headroom" per 100 tons of starship per parsec to be jumped IS NECESSARY SOMEHOW in a way that correctly "load balance matches" what is needed in order to jump. How does that 16 EP of "unused headroom" in capacity get filled? By means of a "controlled power spike overdrive" of the starship's power plant that can deliver the necessary power needed (well in excess of nominal output) for a very short duration (a few minutes) which while intensely fuel INEFFICIENT can get the job done. One of those "stupid but works, may be stupid, but it still works" kinds of deals.
My
personal interpretation 
is that although a fusion power plant can deliver the necessary power OUTPUT needed for the jump drive to "do its thing" successfully, the actual "yield" of a fusion power plant in this "wasteful overdrive" condition is basically
DIRTY POWER, as opposed to
CLEAN POWER.
Basically ... this type of thing, to make it more obvious what I'm talking about:
The 16 EP of "unused headroom" per 100 tons of starship per parsec then
ISN'T being used for a (pure) "fill then dump" all at once type of purpose or use case. It's not like the 10 tons of fuel per 100 tons of starship per parsec is being used to generate a (mere) 16 EP in a sudden overdrive of output power spike which then gets rapidly discharged in order to energize the jump drive in order to jump.
In my
personal interpretation 
... the 16 EP of "unused headroom" per 100 tons of starship per parsec is needed for
POWER CONDITIONING so as to smooth out the "noise" of the "dirty electricity" output (as demonstrated in the image above) into something much more finely controlled and precise. The ACTUAL QUANTITY OF POWER being delivered is going to be WELL IN EXCESS of a mere 16 EP within a few minutes (

) but the TOLERANCE around that output is going to be
within +/- 8 EP which it is then the responsibility of the jump capacitors to "smooth out" into a usable power curve (that doesn't reliably misjump!).
So rather than the jump capacitors being a kind of "baseload storage" type of setup for a jump drive, instead the jump capacitors are just there to "tame the noise" in the staggering amounts of power being delivered to the jump drive. The jump capacitors are there to "clean up" the power spike from the power plant into something more controlled and precise that can be used by the jump drive reliably.
In other words, 10 tons of fuel per 100 tons of starship per parsec to jump DOES NOT EQUATE TO ... 16 EP of output to fill the jump capacitors to maximum and then dump a mere 18 EP "in an instant" in order to jump.
If anything, it probably means that 10 tons of fuel pwer 100 tons of starship per parsec to jump EQUATES TO ...
hundreds of EP of output which then jump capacitors condition to be within
+/- 8 EP of the desired throughput into the jump drive.
Additionally, consider that if jump capacitors yielded a higher power density than fuel, we'd be using "jump capacitor only drives" instead of fusion power plants on overdrive to generate the necessary power spike.
In other words, 10 tons of fuel
yields more power for jumps than 10 tons of jump capacitors (360 EP).
If the ACTUAL amount of power that needs to be delivered to initiate a jump is actually (numbers pulled out of the air purely for illustration purposes) ...
1200 EP per 100 tons of starship per parsec ... in order to supply that quantity of power you can either:
- Consume 10 tons of fuel per 100 tons of starship per parsec (construction cost MCr0 for fuel tankage)
- Install 33.4 tons of jump capacitors having 1202.4 EP capacity per 100 tons of starship per parsec (construction cost MCr133.6)
In other words, it's entirely possible that the reason why we have "fuel consuming" jump drives, instead of "capacitor driven" jump drives, could simply be a matter of practicality in the engineering. Nuclear fuel is simply "more energy dense" than any kind of jump capacitor technology that's available ... in addition to nuclear fuel tanks being "cheap" to construct, while jump capacitor technology is really expensive.
Therefore, the "lots of fuel plus a few jump capacitors" winds up being the More Efficient Option™ from both an allocation of displacement AND construction costs consideration, relative to the alternative of "a little fuel plus lots (and lots) of jump capacitors" when it comes to deciding which path to pursue.
Note that this interpretation does not foreclose on the option of Collectors (such as the one used by
Annic Nova), but rather relegates the Collector "branch" of engineering to a different "focus" of the tech tech tree (so to speak). Furthermore, note that the "3x the tonnage needed" illustrated above when looking at fuel vs jump capacitors is (ironically?) also mirrored in the displacement of the
Annic Nova main computer ... which is (also) oddly 3x larger than it ought to be.
Coinkydink?
Magic 8 Ball says:
Answer hazy. Try again later.
Incidentally, for anyone playing the Home Game™ with all this theorycrafting ... if we assume that "starship combat rounds" last 20 minutes (LBB5.80) that means that there are 36*7=252 combat rounds worth of duration per week. So a Collector on a starship that is generating 1 EP per combat round would collect 252 EP per week ... and after 5 weeks would have collected 252*6=1260 EP ... sufficient for a 100 ton starship to jump 1 parsec using a Collector ... assuming my
1200 EP per 100 tons of starship per parsec supposition above is even approximately correct (pass the entire salt shaker for that one ...

).
The one thing that we do know for sure is that (per LBB5.80) jump capacitors are NEEDED in order to be able to jump, but with the CT RAW for starship design, mere EPs "in the capacitors" IS NOT ENOUGH ... there is also a requirement for jump fuel in order to jump.
Annic Nova was an example of a Collector powered jump drive which we were given no construction rules for in CT. So Collector driven jump drives CAN WORK ... but HOW to make them work under CT was never adequately explained ... because doing so would "peel back the curtain" on how jump drives operate, which was a can best left sealed back in the days when CT was being published.
