From a post to Baen's Bar Honorverse dated April 14, 2004:

Battleship and SD(P) comparisons

    The old-style battleships and superdreadnoughts were completely different animals. Whereas it was fair to call a dreadnought a superdreadnought writ small, there were fundamental design differences between battleships and true ships of the wall.

    In most respects, the battleship as of the beginning of the Havenite Wars was a leftover from a period two centuries and more in the past. At one time, the basic format of the battleship, including its armoring scheme, had been the ideal for most navies' main combatants, what had evolved into the ship of the wall as of the time of Honor Harrington. But it was never intended or expected to withstand the weight of fire that a later SD was expected to survive. The introduction of first the dreadnought and then the superdreadnought represented a tremendous increase in the survivability of capital ships even more than it did an increase in their comparative lethality. In many respects, an SD was actually less dangerous to another superdreadnought than a BB was to another battleship. It was simply the most dangerous ship in space, and was incredibly lethal compared to anything but another superdreadnought, but it had actually lost ground in its individual ability to kill other ships of its own type.

    I've discussed in earlier posts (which Joe has archived) the nature of the armoring scheme of ships in the Honorverse's navies. One point which I may not have made sufficiently clear is that the sheer depth of a superdreadnought's passive defensive systems is far, far greater than that of anything else. It's not simply a matter of thicker armor on the skin of the ship. It's a matter of greater volume for electronic defensive systems, much greater numbers of point defense clusters, greater numbers of counter-missile launch tubes, more decoys, vastly "thicker" sidewalls, additional redundancy (for example, a superdreadnought routinely has at least twice the number of generators required to maintain its designed sidewall strength, whereas a battleship might have a 30 percent redundancy factor, a battlecruiser might have a 15 percent redundancy factor, and lighter ships -- in most navies -- would have zero redundancy), etc. In addition, most navies (and this is certainly a factor for the Royal Manticoran Navy) design their wallers (and who came up with that delightful term?) to use a lot of those additional armored systems the way that naval designers in the years of coal-fired warships used bunker space. That is, the hull volume required to provide physical space for the additional decoys, the additional counter-missiles, the additional laser clusters and their control runs, are wrapped around the core hull of the ship. These are important systems, but they are less important under existing design philosophies than the ship's offensive weapons, her power generation, and her critical control systems. A certain percentage of the ship's purely defensive systems are also located within the boundaries of the core hull, but the layers and layers of redundant or additional active defensive systems serve a dual function as a damage absorber.

    In addition, "proper" ships of the wall are simply much more heavily built. Framing members are bigger and stronger, designed to withstand far greater punishment at the cost of additional building expense and the sacrifice of some internal volume which might otherwise have been used for additional offensive weapons systems. This is a major reason the concern was raised about the structural integrity of a hollow-cored pod-laying design: building that big an open space at the very center of a ship of the wall was considered by many to be an unwarrantable sacrifice, given the then unproven workability and advantages of the pod concept, of the sheer brute strength which had always been a major aspect of superdreadnought design. Beyond that, however, even inside the "armored belt" proper of a SD, structural bulkheads are also armored bulkheads. In a battleship or a lighter unit, they are designed to meet the normal load bearing requirements of the hull; in a waller, they provide additional armor, creating confinement zones designed to limit and contain damage that manages to penetrate the outer defenses. And in addition to armoring bulkheads for probably between 25 and 33 percent of its total beam, a superdreadnought also has additional passages in that same space. And, finally, the core hull itself is totally surrounded by what is in effect a complete secondary "belt" of extremely heavy armor. So, in essence, a superdreadnought is a thick-skinned honeycomb of tightly integrated defensive barriers, all of its designed to keep the ship alive as long as a single one of its scores of offensive weapons remains operational.

    The reason a superdreadnought is so much more powerful offensively than an equivalent tonnage of battleships is that things like impeller rooms, hyper generators, inertial compensators, and other "core operating systems" don't scale proportionately. A battleship's impellers and other drive equipment probably consume something like seventy-five percent of the same volume and mass as a superdreadnought's. The SD will have greater generating capacity and much more power storage capacity than a battleship, so one might argue that the total engineering spaces come closer to scaling directly, but the primary purpose for all of that additional power is to feed the ship's offensive and defensive systems, and the energy mounts and missile tubes themselves are (relatively) modest consumers of tonnage compared to things like power supplies and missile magazines. This is not to say that the weapons themselves are small, by any stretch of the imagination, but they might be thought of as the end points of a complete delivery system (delivering laser or graser fire, in the case of energy mounts; missiles, in the case of missile tubes), and each of them is not very huge compared to the system as a whole.

    Traditionally, battleships have been considered missile platforms. Prior to the huge increases in the efficiency of point defense (which have already been noted in the Honorverse novels as one of the reasons for the reemergence of the missile pod in order to swamp those defenses), missile combat was far more decisive than it had been for some centuries before Honor's naval career began. The battleship was the capital ship of that era of combat, and traditional battleship designs emphasized magazine space and missile tubes over energy weapons. This was essentially the design philosophy of the People's Republic of Haven's battleships which were used in the opening years of the Havenite Wars. There isn't a lot of "stretch" in a battleship-sized hull to match the upgrades in SD offensive firepower, and there's even less room to build in the necessary defensive systems to create a workable parity in BB-vs-SD(P) tonnage and salvo density. The smaller ship will always be overwhelmed by the combination of the larger ship's weight of fire and sheer defensive toughness.

    I once said that Honor's SDs at Yeltsin would have wiped out all of Operation Stalking Horse's battleships, without taking significantly greater damage themselves, had Theisman's ships not already been detached. This may have been a slight exaggeration, but not a very great one. The briefness of the engagement window when their head-on vectors brought the two fleets into energy range favored the superdreadnoughts over the battleships because the SDs actually had an enormous amount of overkill capability which wasn't really necessary to kill the ships they engaged. The battleships were "killed" two or three times each, aside from the single ship which (if I recall correctly) got away virtually completely undamaged because it was able to interpose its wedge against Honor's fire. The superdreadnoughts which survived, on the other hand, did so without taking critical damage. The damage that they took was superficial, in the sense that the core hull remained mostly intact and the various damage-absorbing elements of the design had done what they were supposed to do. Had Theisman's ships been added to the Peep battleship force for the original engagement at point-blank range, they might have gotten through to the core hull of the survivors and inflicted critical (as in ship-kill vs. mission-kill) damage. It's unlikely, however. All of the other Peep battleships would have been destroyed, however, because Honor had overkill capacity to devote to the additional targets. Once Honor's offensive systems had been attrited in that original engagement, she had lost much of that overkill capability. It was restorable, in that the ships were repairable, but she wasn't going to be able to restore it out of shipboard resources before Theisman got into engagement range. The degradation of their targeting sensors and EW capability, degradation of control capability because of lost missile uplinks, ammunition depletion, and the incapacitating damage to energy weapons emitters and missile tubes had all greatly reduced her remaining ships' offensive punch. Against an officer of Theisman's ability, that degradation would have been sufficient for him to get in close after a missile engagement which would be heavily in his favor and destroy or complete the incapacitation of the superdreadnoughts. But even then -- as he himself recognized -- it would cost the effective destruction of his entire remaining battleship force.

    Similar disparities in the combined striking power and defensive capability of SD(P)s and BB(P)s would continue to apply. Indeed, I would suspect that the volume and mass consumed by the missile handling equipment which is part of any pod design would further exacerbate the situation.

    Now, all of this is not to say that the navies of the Honorverse won't eventually build something which might once have been classified as a "fast battleship." If they do so, however, it's going to be because of "bracket creep" and changes in the minimum platform size to continue to fill existing tactical niches. What is much more likely to happen than for anyone to design a "battleship" which anyone but a lunatic would take into action against true wallers is that someone (like, he said innocently, the RMN) will design a "battlecruiser" whose size will fall well into the tonnage range of a relatively small battleship as of the time of The Short Victorious War. This ship, however, will not be intended to engage, hold off, or deter "proper" ships of the wall. It will be intended to do precisely what battlecruisers have always done under Manticoran doctrine: kill cruisers, protect/raid commerce, carry out deep-penetration raids on deep-space industry, etc.. If they have absolutely no choice but to engage ships of the wall, they will be far more capable in that role than smaller, older battlecruisers might have been, but that doesn't mean that it will ever be a Good Idea. If the RMN builds such ships it will scale battlecruisers up rather than scaling superdreadnoughts down. Either way, the traditional "battleship" as a sort of intermediate step between cruisers and wallers will almost certainly disappear from the design philosophies of all "deep-space" navies. It's quite possible that something in that range will continue to be built by second-and third-rate navies as the equivalent of a "coastal defense battleship," but no one with pretensions to power projection or with large areas (and the resources to match) to protect will use such vessels. I can tell you right now that the RMN is already thinking in terms of extending its nodal defensive posture and relying on relatively cheap weapons to deal with relatively minor threats. A mix of LACs, missile pods, light combatants, and quick-response nodal forces is going to be far more cost-effective for them than trying to station sufficient combat power to stand off task force-level attacks in every critical system. They cannot be strong everywhere, so they will opt for overwhelming strength in the areas where they must be strong and a flexible, counter-attack-based posture in other areas.