A collection of posts by David Weber containing background information for his stories, collected and generously made available Joe Buckley.
The short answer is "Very." A somewhat longer answer follows.
In The Short Victorious War, I said that "Truly enormous ships, like superdreadnoughts, might mount broadside weapons on as many as four or five separate decks, and their length was as much as seven or even eight times their maximum beam, which meant that each gundeck offered twenty-five to thirty times the weapons volume available to their chasers." There are a couple of things about this statement. First, note that I said they might mount broadside weapons on as many as 4 or 5 decks; I did not say that all of them do. In fact, most do not, because building in that many decks armed with offensive weaponry cuts into the defensive capabilities and magazine space of the ship. An analogy might be to look at some of the pre-WW One BBs built for the navies of third-world powers, like HMS Agincourt (taken into Royal Navy service at the outbreak of the war, but originally designed for the Brazilian navy). Agincourt mounted no less than fourteen 12" guns in seven twin turrets, which was a most impressive broadside, but to build it into the ship, she sacrificed armor, internal armored bulkheads, and a wing passage inside the belt, and accepted a weaker main girder strength. That is, her designers cut so many holes into her main armored deck to fit in the barbettes for the turrets (which, by the way, were also very thinly armored to save weight) that, coupled with the sheer length of hull necessary to fit in that many weapons, her structural integrity was highly suspect. In the case of HH universe SDs, minor powers have a tendency to "over gun" their ships (remember Warnecke's designs in Honor Among Enemies?) at the expense of armor, ECM, and other passive and active defenses. Major powers, with large amounts of combat experience, tend to follow the design philosophy of people like the Manticorans and the Andermani, settling for three decks armed with the most powerful possible energy weapons plus missile launchers and using the volume/mass which might have gone into the additional weapons decks for better defense and more magazine space.
Second, the statement about the amount of volume available is misleading, which is entirely my fault. What I actually meant is that the gundecks can offer seven or eight times the volume inside the hammerheads on a per-deck basis. This is very important for the chase weapons in terms of missile tubes, but it is less so (in a way) for energy weapons. The reason for this is that the missile tubes, in their entirety, have to be fitted into the space forward (or aft, depending on which end we're talking about) of the impeller rings, but the energy weapons normally do not. The emitters for the energy weapons have to be crowded into that space, but the weapons themselves are almost always spinal mounts which extend through the central core of the ship with the impeller rooms and rings being built around the weapons. This allows the designers to put the bulk of the weapons deep inside the hull, allowing more room for missiles (and magazines) in the heavily armored hammerheads. It is also true that the hammerheads normally carry very heavy outfits of point defense laser clusters, designed to give the ships their best shot at destroying laser heads as they attempt to maneuver for "down the throat" or "up the kilt" attacks.
This design feature also leads to another very significant point about energy weapons in capital ships (or, for that matter, almost any warship): the chasers of a given ship are always considerably more powerful than the broadside mounts. There are several reasons for this, but the biggest is that the chasers (or, rather, the bow chasers) have the widest vertical arc of fire and likewise cover the vessel's most vulnerable aspect, since the throat and kilt cannot be closed with sidewalls (well, yeah. There is the Shrike, but you know what I mean). The object is to provide the captain of a ship with the most powerful, longest-ranged light-speed weapons possible in order to give him the best chance of killing an opponent before said opponent can kill his ship. The chasers thus use up a portion of the internal volume of the main body of the ship, which, in turn, steals at least a little volume from the gundecks.
The broadside energy weapons of a ship tend to be considerably lighter than her chase weapons. As explained above, chase weapons will be few in number whatever a designer does, so naval architects go in for the heaviest possible "guns." Broadside weapons, on the other hand, face some rather different constraints. One is the necessity to sometimes engage more targets simultaneously, which requires numbers of mounts, not just individual power. Another is redundancy; designers opt for more of somewhat lighter weapons so that single hits will take out a smaller percentage of the broadside's total power. Another is simple competition for internal volume, which, coupled with the need for numbers of weapons, helps hold their individual power down.
The chase energy weapons of a late-generation SD as of the time of Echoes of Honor are probably somewhere in the 5.5 to 6 meter range. Broadside weapons for the same ship would run in the 4.5 to 5 meter range. Needless to say, both are very destructive.
Another factor to bear in mind when thinking about numbers of energy weapons is the provisions made for "on-mount" power in the event that one or more of the ship's fusion plants goes off-line. Any energy weapon aboard a modern warship has a back-up super conductor power ring--capacitors, if you will--to allow the weapon to continue to fire for at least a short time even if external power to the mount is lost. (This could occur either through simple interruption of the power grid to the weapon, or the complete loss of one or more fusion plants.) For lighter vessels, the volume required for the back-up power storage puts a severe squeeze on total internal volume, which means that their energy mounts tend to be capable of no more than a very few shots from "reserve power." SDs have the mass tio provide much greater energy storage, and although their weapons are massive power hogs compared to those of lighter units, they can still supply the power for as much as a dozen shots in local control and local power. However, not even they have unlimited volume, and part of the price for sustainable offensive firepower is the acceptance of somewhat lighter energy weapons than the maximum which could possibly be fitted in.
Dreadnoughts of a given navy tend to mount the same weapons as that navy's superdreadnoughts, but to mount them in lesser numbers. Hence when HMS Bellerophon took out Admiral Pierre's BCs in The Short Victorious War, she was showing what a superdreadnought could have done to a considerably larger force of BBs, but the disparity in firepower on a ship-for-ship basis was very much the same.
SDs are also the most capable defensive platforms in space. They have the most volume for internal EW systems, carry massive (comparatively speaking) quantities of decoys, mount extremely large numbers of counter-missile launchers and point defense clusters, and have the toughest sidewalls and thickest armor. A BC at minimum range would find it almost impossible to critically damage an SD with energy fire through a sidewall; an SD, on the other hand, would effectively rip a BC to splinters. The disparity owes a great deal to the difference in sidewall strength, but it also reflects the sheer toughness of the SD's armor.
SDs also have the largest magazine capacity of any warship type. As a general rule, an SD missile tube uses up less surface area than an SD energy weapon and is individually less massive. By the time magazine space and ammunition handling requirements are factored in, however, an SD's missile armament ends up using considerably more total internal volume on a mount-for-mount basis than its energy weapons do. The relatively low number of braodside tubes aboard most RMN SDs reflects several design considerations. One is the need to limit one's self to a total broadside which can avoid impeller wedge fratricide on launch. Another consideration is that, like the gunsmoke of pre-space broadsides, each missile salvo temporarily causes a massive degradation in the broadside fire control sensors of the firing ship as the missile wedges block them, which means that there is little point in mounting larger numbers of tubes (or, for that matter, tubes with much shorter cycle times) simply because one must wait for "the range to clear" before launching the next salvo anyway.. There are other considerations, including ammunition supply concerns, but perhaps the overriding one is that prior to the introduction of missile pods, the defensive power of opposing SDs was such that in a one-on-one duel, no SD was likely to be able to get any significant number of hits through her opponent's active and passive defenses even with the maximum number of tubes which could be packed in (at least, without making totally unacceptable sacrifices in close in [i.e., energy weapons] firepower and/or defensive systems) in a single opening salvo. The tactical officer (and his computers) needed time to get a feel for the opposition's EW tactics and defensive fire plans and maneuvers before he could hope to get any significant number of hits, and that required sustained combat power and the best possible defenses of his own rather than sheer numbers of tubes. Of course, there are limits in all things, and virtually all RMN designs tended to be missile-light compared to their Peep counterparts simply because the Manticorans had more faith in their missiles' accuracy and penetration aids. The Andermani philosophy reflected in Derrflinger was somewhat different from either RMN or PN philosophy. Designed before the new Manticoran compensators came along, Derrflinger was optimized for the long-range missile role not only against her counterparts in someone else's wall of battle, but also against lighter opponents. Her job was to carry out the preliminary missile engagement to both (hopefully) damage the opposing ships of the wall and (almost certainly) strip away the opposing wall's screen at long range. Her lower mass should have given her the acceleration advantage to enable her to hold the range open and avoid an opponent's optimum engagement zone, and also would make her much more effective against the lighter "capital ships" likely to be encountered in the hands of Silesian privateers and liberation governments--a point of some importance, given the Andermani preoccupation with Silesia and the Empire's territorial ambitions therein. In this regard, she may be thought of as an Iowa-class from WW Two or perhaps a Queen Elizabeth class from World War One. She does not represent the "mainstream" design of the Andermani Navy. Rather she represents a special "fast wing" designed with a specific, specialized tactical function.
With the Manticoran reintroduction of the missile pod, a new approach to missile warfare became available. There was no particular point in designing ships with larger missile broadsides if the Harrington/Medusa design proved workable (which it did). A pod-equipped SD (I suppose we might call them SDPs) can pump out more shots in a single salvo than any conceivable conventional broadside could fire, so the RMN decided to retain the essentials of its pre-SDP broadside armaments (including the missile tubes) for engagements at shorter ranges. The magazine capacity for the "standard" tubes is far, far lower than it once was, but they probably would not be used at all while the supply of pods lasted. Indeed, it is even possible (though by no means certain) that the RMN may ultimately decide to scrap the standard broadside tubes entirely in favor of additional anti-missile defense capability, although at the moment the Manties are thinking more in terms of massive improvements in electronic warfare (including the Ghostrider systems).
It is important to remember why these ships mount energy weapons at all. One, as already noted, is that opposing SDs are very resistant to (conventional) missile fire. More importantly, perhaps, missile armaments are subject to ammunition limits. In the final analysis, however, the overriding consideration is that eventually a fleet will attack an objective the enemy must defend. When that time comes, the long-range sparring of a missile duel and/or the decision to break off and run will no longer be options for the defender, and action may be forced to close quarters, at which energy armaments rule supreme. And, of course, the provision of a very powerful energy broadside helps preclude the possibility of an SD finding herself "swarmed" by lighter units. Whether this will continue to hold true against targets like a Shrike-class LAC, which can attack in very great numbers and is a very hard target, remains to be seen, of course.