I'm seeing $50K for Excaliber, $40K for GMLRS. Spendy rascals. On the other hand, I think these munitions are going to reduce the need for CAS, since with that kind of precision there is less use in having a pilot fly a bomb in, except for hardened targets and such.

Form my own experince in fire planning for both weapons the cost of PGM ammo for MLRS must be considered as the primary ammo cost, if a complete replacement of cannon with a MLRS is the question.  Ordinary dumb HE rounds from cannon are more accurate than MLRS un guided projectiles.  Cannon HE are more effective than the rocket ammo with submunitions for attacking buildings & entrenchments as they have stronger projectile cases and fuzes with a delay time setting.  For a MLRS to replace cannon ammo capable of destroying enemy positions in masonry or concrete buildings and field entrenchments or bunkers must be provided in bulk.   

In Lebanon last year much of the Israli artillery was of marginal effect as it was not penetrating the enemy positions.  Nor was it hitting them with precision often enough.

I have long favored missle artillery for its range.  But not dumb rocket projectiles that only scatter bomblets.  Once rocket or missle artillery is provided with a full array of ammunition as cannon are then it may replace cannon.  And, I dont mean MLRS ammo types that are advertised in venues like this web site.  But ammo that is actually in use routinely on the battlefield.

The second is the IMI Trajectory Control System (TCS).  This is a simpler missile guidance section and radio datalink.  It uses ground based missile tracking to send guidance updates to the missile.  Rumor has it, it has a 40-50m CEP and is significantly less expensive than GMLRS.  This system was used by the Israelis in Lebanon.

Another step will be multiple size warheads in the launcher magazine.  If the desired effect calls for light precise rounds that is loaded on order.  Or if a building penetrator is needed... or a large capacity bomblet carrier....  All should be available for launch in seconds.

However, it would complicate maters, guaranteed to end up with part used pods, this becomes more important as lighter lnchrs only carry a single pod. Even more difficult in lower intensity ops where single lnchrs may be deployed widely seperated, this seems to be what UK are doing in Afg where they fired their first GMLRS earlier this week, quoted as having a '200 lb' warhead, it unclear if this is PR licence or true.

Incidentally I'd say that gun HE is more effective against some personnel type targets, bomblets are contact fuzed, gun HE typically comes fzed with at least PPD if not PPDD in W armies these days, airburst is more effective against prone or dug in men without OHP.  It's interesting that after GW1 Iraqi PW told UK questioners to the effect that bomblet wasn't too bad but airburst HE was seriously nasty (of course they weren't using HE M107!). 

When we worked the US Army MLRS of the 1980s & 90s into our fire plans its configuration as a one trick pony (well maybe two trick) was more than anoying at times.  Tac Air with PGM or rotor aircraft often had to resorted to where the long range MLRS did not have the appropriate ammo or precision for the targets.

You are thinking inside the box there.  Theres no law of nature requiring a single pod or magazine.    There is also no imutable reason the ammo carrier has to be seperated from the launcher for long periods.  In the cannon artillery, towed or SP, we positioned the ammo vehicals according to requirements.  A rocket or missle launcher system with the same degree of flexibility would be vastly more usefull than the current systems I am familar with. 

I don't actually understand this.  A pod containing mixed natures means heroic assumptions about the relative need for different types.  If you get this wrong you end up with a stak of part used pods.  I suspect this is the reason that to date pods have been single nature, you might still end up with a part pod (although with dumb munitions the fireplanning application can be designed to round off to the whole pod (been there, done that).  The pods with less required natures can be supplied to meet actual demand.  The only problem is the time taken to unload and reload a pod.

Gun ammo is different, you bcan have mixed nature ULC,and flatracks with ULCs containing differnt natures (either mixed or single nature ULCs).  The point is that with flatracks and ULCs the bty can swap ULCs around using their MHE or manually consolidate the contents of a ULC.

Think real hard about this part

Theres no law of nature requiring a single pod or magazine.    There is also no imutable reason the ammo carrier has to be seperated from the launcher for long periods.  In the cannon artillery, towed or SP, we positioned the ammo vehicals according to requirements.  A rocket or missle launcher system with the same degree of flexibility would be vastly more usefull than the current systems I am familar with. 

These are all statements of the obvious and irrelevant.  The point under discussion is whether or not a mixed nature MLRS pod is a good idea.  Mixed nature means the pod holds rockets with warheads of two or more different types.  If you are using a lnchr, eg M270 SPLL, which carries two pods then you can have a mixed load, one pod with one type of warhead the other pod with another type, no rocket science here.  However, newer light weight launchers are only holding one pod, which is an added complication. 

The problem with a mixed nature pod is that it's virtually impossible to predict the balance of natures that will be expended.  This will lead to part used pods, which either have to be abandoned or taken along for the ride.  Given the limited number of pods that a truck can carry this could become a logistic drag real soon.  With single nature pods you can always plan to shoot the entire pod, although with more expensive munitions this may not be very popular.  With single nature pods you can minimise the number of part used ones left over, and make adjustments in the logistic system to deliver pods with the more used warheads.  Topping up part used pods in the field has not been an option as far as I know.