First post, here, but I'd like to weigh in on this.
I should say, first thing off, I'm sorry but I don't know the answer to the main question: D'deridex class maneuverability compared to Galaxy class at impulse. No clue, in fact.
But the question of this ships' Artificial Quantum Singularity (AQS) power supply came up and I'd like to address it, with y'all's indulgence.... See, I had a long conversation on the Birth-of-the-Federations-2 forums about this subject in 2005 and I've been thinking about it ever since. I've figured out some more cool stuff on the subject since then that I'd like to share....
"Timescape" demonstrated that an AQS is not like a Natural Black Hole (NBH) because if it were, the alien's offspring wouldn't have had any problems and the story would stops making sense. This is really the only clue we have for testing any Romulan AQS model for accuracy.
My first thoughts about getting energy from an AQS was from an accretion disk and the resultant polar jets. If your singularity spins, it would have enough gravitational influence to literally pull space-time around it. Anything that's traveling through that spinning space-time will also be pulled into orbit around singularity. This forces the material into a disk, an "accretion disk". As material orbits, it rubs against the material orbiting just a little closer and just a little further from the singularity. The friction from this rubbing heats things up and the friction increases as the orbits get closer in. So much so, that Hydrogen fuses into Helium, which then fuses into Carbon, etc, all the way up to Iron, then further into the transuranics --this, even though fusion of elements heavier than Iron takes more energy than it gives off. Indeed, I read that an atom may give up as much as half its mass in energy in this process of fusion and frictional heating.
Any atom that absorbs enough of this released energy to escape the accretion disk would then follow the singularity's magnetic field lines to the poles where it would have a high probability of bouncing off --like a magnetic mirror-- creating high-velocity jets out the poles. (It's even thought that over the course of a NBHs lifetime, this process might introduce a considerable amount of heavier-than-helium elements into the universe, enough to be comparable to the amount given off by supernovae.)
This is the process proposed by Rick Sternbach in his article on getting energy from an AQS. But there are some definite problems to this proposal.
First off, if you're only 50% of the mass you throw in comes out as energy, the other 50% is feeding the singularity's mass, making it harder to control and steadily reducing the maneuverability of you ship.
Second, increasing or decreasing energy output would take time and planning. The material would have to travel from the matter injectors and deeply to accretion disk before it made any difference in energy output. When you take into account relativistic effects, this might take quite some time indeed. Depending on the geometry of the reactor, this could mean unexpected emergency power would have to come from somewhere else, as the AQS would need time to ramp up power.
Thirdly, and most significantly, a NBH small enough to be mobile --less than a million tonnes-- would want to evaporate so quickly from hawkings radiation as to be absurd: a 1e9 kg NBH evaporates in 84 nanoseconds if left to itself. (Oddly enough, it starts with a temperature of only 835 kelvin... Assuming I calculated properly.) To keep it form exploding, you would have to exert control over that singularity on the order of picoseconds! (pico = 1e-12) I don't know enough to do the calculations but I'd guess --due to the second problem-- this might be close to impossible. (I could be very wrong on that account.)
Again, the one thing we know about an AQS is that it's not like a NBH. So what is it like?
My second guess for extracting power from an artificial singularity, and the one I prefer, is that an AQS would be more accurately called an Artificial Event Horizon. I see it working like this: You have a spherical array of powerful gravity emitters focused to a point. (These emitters would be something like deflectors but physically smaller and specially designed for the task.) When they're turned on, the array creates an artificial event horizon whose radius can be made smaller or larger regardless of how much mass you have inside. ("Regardless" is a relative term, of course; relative to the power of the emitters.) From there, you would throw in mass and it would be saved within the event horizon to be extracted later via hawkings radiation. If you wanted more energy from the AQS, you make the event horizon smaller, increasing hawkings radiation; for less power, make the horizon larger. And the beauty of it is that the mass-to-energy conversion ratio is 100%, just like M/AM! However much mass you put in, 100% becomes available as collectible energy via hawkings radiation...eventually.
Of course, this set up does beg a question: What would happen if such an AQS were run without any mass within? Could you extract energy directly from vacuum energy? Naturally, I don't know but I suspect the answer is, "no". See, the energy of hawkings radiation is stolen from the mass of the singularity. If there's no mass within, the singularity would gain negative mass and this negative mass would eventually overcome the gravity projectors. How quickly "eventually" would arrive is up for debate, at least until someone familiar with general relativity could do the calculations. Still, I suspect it would happen pretty quickly. None the less, it brings up the possibility of running the AQS on empty for a while until you can stop off somewhere and fill the tank.
(Hrm... Then again, maybe not. If the singularity has a negative mass then wouldn't the particles of the hawkings radiation also have negative mass? If so, this would surely badly interfere with the gravity emitters and muddle the works. Opens up interesting possibilities, though... Oh, and antimatter, as it's currently understood, does not have negative mass.)
Oh, yeah! And the other advantage of this Artificial Event Horizon model of the AQS is that it can take any kind of mass and turn it into energy! You could throw in asteroids, gas-giant atmosphere or humanoids --though probably not dark matter-- and out would come hawkings radiation, ready for use. This means an AQS powered vessel needs very little infrastructure to keep itself going --no antimatter barges to rendezvous with, for example. And if the vessel was stocked with industrial replicators, there'd be even less need for supporting infrastructure. Indeed, this is one explanation for the huge enclosed area within the D'deridex: a place for the industrial replication of a smaller, possibly fusion powered vessels! Or a small space station! Or a colonial township which could then be beamed into place! Or terraforming infrastructure, likewise with beamed installation! Such a vessel could be a jack-of-all trades, capable of so very, very much!
And, finally, to wrap it up, an artificial event horizon would probably be a completely different environment than a natural one, giving the Timescape alien babys developmental issues.
Ok. That's my long, $0.02 post on AQS power generation.
Tell me what y'all think.
And, really, the point of this game --again, for me-- is to learn about what we know. Like being corrected by publiusr just now. 
AWESOME!!
