Starship Construction Timeline

[Dukhat]

I'm writing up some starship-related essays in my spare time for fun, one of which concerns the conjectural Starfleet ship classes. In order to pin down attributes for each class, I needed to find out in what time-frame they were hypothetically built, by using the registry numbers of the known ships of each class, along with the numbers of some other official contemporaneous classes. So by using such research materials as the Chronology, Bernd Schneider's timeline from Ex-Astris-Scientia, the TNG Tech Manual, statements made about the ships in dialogue and other factors, I made a hypothetical timeline of ship production. Not only did it help me out, it also showed some interesting information, such as that production of new ship classes happened in roughly ten-year increments between the 2270's and 2360's, the timespan of the timeline.

Please understand that these are my own unofficial opinions based on my data, and that there has never been official proof that registries are chronological (and even in some cases, my timeline has no good explanation for some classes of ships, i.e. the Oberth production). But I've tried to do my best with the information I had. Please bring to my attention any mistakes or omissions, as I want this to be as complete and accurate as possible. Canon information is in black font, and conjectural info is in blue font.

http://www.box.net/shared/sp4ynx9pm5

 

[Unicron]

Are you attempting to use the registries as a chronological reference for purposes of the timeline? The reason I ask is that it's been my experience that it's probably not the best bet, as a lot of the registries for both official and unofficial classes don't make a lot of sense in that respect. It might be easier to treat them simply as batch numbers for a given class, and it's usually fairly easy to keep them from overlapping too much.

I do have some tech stuff from various sources, so if I can be of any help, please let me know.

 

[Dukhat]

Are you attempting to use the registries as a chronological reference for purposes of the timeline? The reason I ask is that it's been my experience that it's probably not the best bet, as a lot of the registries for both official and unofficial classes don't make a lot of sense in that respect. It might be easier to treat them simply as batch numbers for a given class, and it's usually fairly easy to keep them from overlapping too much.

I do have some tech stuff from various sources, so if I can be of any help, please let me know.

Unicron: If I understand your question, yes, my theory is that registries are (mostly) chronological. And honestly, they really need to make SOME kind of sense, otherwise why didn't we see registries as off-the-wall as NCC-08 or NCC-FKY429 in TNG?

The reason why I made the timeline in the first place was that I wanted to figure out where the conjectural ship classes fell in terms of age and design attributes. As to your points:

1. I'm only interested in canon registry numbers and ships. There's so much non-canon stuff out there, a lot of which contradicts each other, as you said. And it doesn't help me with figuring out what the conjectural classes might look like anyway.

2. The "batch" idea did occur to me at one point, especially with the Excelsior/Ambassador conundrum (i.e. why Excelsiors in the TNG era have higher 4XXXX numbers as opposed to the more advanced Ambassador class having only 1XXXX and 2XXXX numbers). Sure, I could say that the older Excelsiors were all produced before the newer Ambassadors even though their "batch" numbers are higher. The problem arises when you realize that registry numbers for the Excelsior class are all over the place (2XXX, 1XXXX, 3XXXX, 4XXXX, 5XXXX, and if you count the "Melbourne," 6XXXX). That kinda throws the batch idea out the window.

 

[Unicron]

I agree they need to make some degree of sense, and I tend to prefer the batch system myself rather than trying to use chronological order. I just find it easier, and I think in terms of certain numbers overlapping between classes, at the least one might say that one design is a cruiser and another a frigate.

Admittedly it doesn't solve problems like the Excelsior registries either, because a lot of those were just made up out of thin air. FASA at least attempted to keep its Excelsiors within the 2000 range set by the lead ship, and I find it interesting that not only are most of the canonical ships ridiculously high, few if any of them have a 2000 series number. Kind of annoying, really.

 

[Dukhat]

I agree they need to make some degree of sense, and I tend to prefer the batch system myself rather than trying to use chronological order. I just find it easier, and I think in terms of certain numbers overlapping between classes, at the least one might say that one design is a cruiser and another a frigate.

Actually, the nuttiness with Excelsior and Miranda registries didn't really start until DS9, although the Oberth's registries were pretty much always out of whack thanks to "The Naked Now." I don't have any proof of this, but I honestly think the Oberth class was supposed to be a newer design based on the Tsiolkovsky's launch date, but there wasn't enough money to build a new model so the Grissom model was perpetually stuck with the moniker (which happened time and again whenever a new ship was needed but couldn't be built). Just wait until you read my essay about the Oberth class to see my hatred of that ship

However, most new models built for TNG (Galaxy, Nebula, Constellation, Ambassador, the BoBW kitbashes) are pretty consistent registry-wise. It's the movie models that give me such a headache (see below).

Admittedly it doesn't solve problems like the Excelsior registries either, because a lot of those were just made up out of thin air. FASA at least attempted to keep its Excelsiors within the 2000 range set by the lead ship, and I find it interesting that not only are most of the canonical ships ridiculously high, few if any of them have a 2000 series number. Kind of annoying, really.

That's one of my main gripes with the Excelsior and Reliant models being used in the show, especially during DS9. During the first few seasons of TNG they got it right; most reuses of the movie models gave them extremely low registries in an attempt to show they were just old ships still in service (the Hood's original NCC-2541, the Repulse NCC-2544, the Lantree NCC-1837, the Jenolan NCC-2010, etc. Even the Stargazer and Hathaway had low 2XXX registries). Then, for some oddball reason, in DS9 they started showing up with these ridiculously high numbers for no reason whatsoever (as you say, out of thin air). I have a copy of the old FASA TNG Officer's Manual, and you are indeed correct about the 2XXX regs for the Excelsior.

I understand why they were in DS9. The original physical models existed for them to scan and create the CGI model from. And I also understand why the bulk of the fleet shots showed these two classes of ships more than any other: because the level of detail was better than the detail of the First Contact ships also in the fleet. But what was the rationale for giving them such high numbers?

 

[Londo]

Theorising wildly:
In the first half of the 24th century the Federation seems to face very few threats. It would not be surprising if they dropped the ball a bit when it comes to new construction. The Mirandas and Excelsiors are reaching the end of their operational lives but there is no pressing need to upgrade to new classes so they get a major rebuild to extend their lives instead, this could be connected to the move away from duotronics. They could have come out with their old numbers but someone in starfleet engineering, or perhaps the media department defending their continued use of old designs declares that they are practically new ships and so they get a new number to fit with the new spin.
Thus the Low registry Excelsiors were simply rebuilt before the new media strategy or not rebuilt at all.

 

[Deks]

The thing is that fully upgraded Mirandas and Excelsior class ships ARE more or less new ships.

Problem is, on DS9 (for example), ships were dropping like flies (of older and newer designs) because apparently someone completely forgot about durability of ships in a fire-fight with shields.
Couple of hits and the Miranda class ship is gone.
Excuse me, but that almost reeked of someone letting a completely NOT upgraded ship into a battle ... which is a waste of a perfectly good ship and lives.

On the other hand, an Excelsior class ship, the Lakota, once upgraded was on par with one of the most powerful war-ships in the A.Q., the Defiant.

DS9 was full of inconsistencies, however, putting that aside, I do think a fully upgraded Miranda and Excelsior class ships (not to mention the Ambassador class) would continue to function as intended for longer periods of time in a new era which would bring the said ships up to par with the rest of the Galaxy.

 

[Captain Robert April]

Probably the only chronological aspect to ship registries is when they're issued and assigned a slot in the construction schedule. When that ship actually gets built, and whether or not the number ultimately gets on the ship it was originally intended for, is something for the arcane vagueness of the Starfleet Operations administrative section.

F'r instance, a name and registry is assigned a slot in the schedule. Let's call it U.S.S. Baltimore, NCC-2295. At the time this name and registry is assigned, the plan is for a Miranda class ship.

Sometime before that slot comes up, though, Starfleet decided they don't need that many Miranda class ships this year, so the order is cut. Instead, they want more Oberth class ships, so that slot, along with the name and registry, is now destined to go on an Oberth class ship. At least until something else happens in the front office and the order gets changed again.

(The only downside to this scheme is that it tends to play into the hands of the "Naval Construction Contract" crowd, but what can ya do....)

Makes ya wonder what happened with the Constellation, don't it?

 

[Unicron]

That idea does make some sense to me, as it's the case with some of FJ's ships (his 500 series registries start with the destroyers and continue to the Hermes scouts, which then get 600 registries) and ships in some other sources.

 

[Captain Robert April]

Slight modification to my scheme. Names and registries may not be all that linked together, since there does seem to be a bit of a theme regarding what names go with what class of ship. Constitution class ships get named after famous naval vessels, Oberth class get named after astronauts, Danube class runabouts get named after rivers, etc. So, the name might not be decided upon until the actual ship being built gets decided on.

 

[Timo]

One might also argue that big and important ships get named after previous ships (thus displaying no thematic consistency as such), whereas small and humble ships get named thematically, if only because such small ships are also produced in vast numbers that would stretch the imaginations of ship namers otherwise.

Thus, Oberths get named after space pioneers of all sorts - but if one Oberth/i] does something really impressive (read: dies heroically), her name might be given to a big capital ship later on. Thus we get the Excelsior class Grissom, even though the Excelsior class features no general astronaut theme to its names.

On the issue of Excelsior registries, I see nothing odd in there being both pre-Ambassador and post-Ambassador ones. The Excelsior wasn't replaced by the much larger Ambassador, it seems; instead, Starfleet merely complemented its arsenal with a few large Ambassadors during an era when it was building Excelsiors for its workhorses.

There's no competing workhorse class in the registry range below NCC-50000, really. So Excelsior production could plausibly proceed until that range, and until the 2340s or so, as the main production effort of Starfleet - those ships would still be the hottest news, the best available design. In the 50000 range, ships of Galaxy design style begin to appear, and in the 60000 range we get possible new workhorses such as Akira and Nebula. But individual Excelsiors might well be completed a decade after the end of primary production.

It's pretty much the same with Miranda. The Excelsior didn't replace it; canonically, nothing much in that lower end of the size range seems to emerge until the registries hit 60000 and we get the Sabers. So Starfleet could quite well keep on building the Miranda as its most advanced and foremost small ship type until the 2320s or whatever.

...Although I like to think that the "Excelsior generation" did eventually come to feature its own small ship type - and that USS Centaur was it. I might even slap a class identity to that ship, something we have heard of but haven't seen - say, Renaissance class.

That idea does make some sense to me, as it's the case with some of FJ's ships (his 500 series registries start with the destroyers and continue to the Hermes scouts, which then get 600 registries) and ships in some other sources.

To be sure, FJ's destroyers and scouts are pretty much the same ship, with only minor differences in standard of equip - so one need not argue that production was "rethought" or "rerouted" in the middle of some NCC range. Rather, one could say that Starfleet systematically pushed more and more destroyers out with consecutive registry numbers, perhaps while the same thing was happening to cruisers (but at a different segment of the registry range) and frigates (ditto) etc.

That is, all TOS/pre-TOS destroyer registries would be part of a continuous, chronological sequence; so would all TOS/pre-TOS cruiser registries, and all TOS/pre-TOS frigate registries, etc. But the destroyer registries (say, 400 onwards) would all be lower than the heavy cruiser registries (say, 1000 onwards), etc. This system would prevail over simple chronological numbering until the turn of the 24th century - with minor anomalies here and there, of course.

Timo Saloniemi

 

[Saquist]

I'm writing up some starship-related essays in my spare time for fun, one of which concerns the conjectural Starfleet ship classes. In order to pin down attributes for each class, I needed to find out in what time-frame they were hypothetically built, by using the registry numbers of the known ships of each class, along with the numbers of some other official contemporaneous classes. So by using such research materials as the Chronology, Bernd Schneider's timeline from Ex-Astris-Scientia, the TNG Tech Manual, statements made about the ships in dialogue and other factors, I made a hypothetical timeline of ship production. Not only did it help me out, it also showed some interesting information, such as that production of new ship classes happened in roughly ten-year increments between the 2270's and 2360's, the timespan of the timeline.

Please understand that these are my own unofficial opinions based on my data, and that there has never been official proof that registries are chronological (and even in some cases, my timeline has no good explanation for some classes of ships, i.e. the Oberth production). But I've tried to do my best with the information I had. Please bring to my attention any mistakes or omissions, as I want this to be as complete and accurate as possible. Canon information is in black font, and conjectural info is in blue font.

http://www.box.net/shared/sp4ynx9pm5

I've posted my research before but there is other's you can use.


Constitution Class NCC-1700 USS Constitution
Miranda Class NCC -1837 USS Lantree
[2266]-Romulans destroy Federation Outpost
[2267]-First Klingon War
[2285] Excelsior class NX-2000

-Constellation Class 2893(USS Stargazer)

[2293] -Khitomer Confrence
2311 -Tomed Incident: results in 53 year Romulan isolation
-Earliest Ambassador class (Zhukov NCC-26136)
-Niagra Class USS Wellington (NCC - 28473)

2344 - Enterpise C destroyed at Narendra III
2346 -Khitomer Massacre by Romulans
2347 -Start of Cardassian War
-Steamrunner USS Appalachia (NCC-52136)
-New Orleans class USS Rutledge (NCC 57295)
-Nebula Class NCC 60205 USS Honshu
-Sabre Class USS Yeager (NCC-61947)
-Akira Class 63549 USS Thunderchild

2355 -Stargazer disabled by Cardassian war ship.
2356 -Galaxy Class NCC 70637 (USS Galaxy)

2362 - Massacare at Setlik III
2363 - Galaxy class (USS Enterprise D)

Federation Produces 68,637 units in 78 years. (879 units a year)

2366 -Defiant Class Development Project
2367 -Wolf 359 - Cardassian Truce Demilitarized Zone
-Nova Class USS Equinox NCC-72381

2368 -Danube Class NCC -72452 (Rio Grande)
-Defiant Class (NX 74205)
-Intrepid Class (NCC-74600)

2371 -NCC -74656 (Voyager) / Type 9 shuttlecraft
2371 -Defiant pulled from Storage
2372 -Enterprise E launched
2373 -Battle of Sector 001 / Voyager encounters 8472 / DS9 start of the Dominion War.

( Federation produces 257 units)

2374 -Prometheus Class NCC-74913

Federation produces 314 units a year.

2375 -Briar patch. - Scout 75227 / Voyager encounters (USS Equinox) / End Dominion war.
2378 -Destruction of Borg Collective - Voyager returns to Earth
2379 -Battle of Bassen Rift - USS TITAN (NCC -[80102])

Federation produces 5,189 units in Five Years. 1,037 ships a year after the star of the Dominion War.

Between the start of the Cardassian Federation wars some 1,156 ships are produced a year.
__________________


Also G2k...Darkstar put one together I believe.
http://www.st-v-sw.net/STSWvolumetrics2.html

It's a long long read to comprehend...

Me and DarkStar come out with different figures because I take in account Sternbach and Bermans statement that there is around 30,000 ship in the Federation Fleet from USS HOOD to USS VOYAGER.

 

[T'Girl]

double post

 

[T'Girl]

Constitution Class NCC-1700 USS Constitution

USS Constellation under Commodore Matt Decker was NCC-1017. I think despite the simularity between the two, the Enterprise and the Constellation were of different classes. That's how I personal acount for the Constellation's hull number, so perhaps that class began with NCC-1000.

 

[SpyOne]

Constitution Class NCC-1700 USS Constitution

USS Constellation under Commodore Matt Decker was NCC-1017. I think despite the simularity between the two, the Enterprise and the Constellation were of different classes. That's how I personal account for the Constellation's hull number, so perhaps that class began with NCC-1000.

Yeah, that or call it a typo in canon. I mean, it's a problem, and it needs to be dealt with somehow, and I find none of the ways of fixing it particularly awesome. I prefer to pretend it was really NCC-1710, but I know full-well that I'm boldly going non-canon when I do, so maybe making it an older class is best.
Similarly, a lot of the canon Constitution Class vessels really aren't: they are never shown, and never specifically said to be a Constitution Class, but somewhere in fandom it was decided they were and it stuck. Take USS Republic: It is specifically stated that her "number" was 1371, but her class is never suggested, nor was there any dialogue making her particularly similar to Enterprise.

One of the problems one encounters with the NCC numbers is that their rate of use suddenly jumps. That is, in the first hundred years the Federation used roughly 2000 numbers, but 100 years later they were in the 70,000's. One explanation for this is that they built tens of thousands of ships for some reason. Another is that many of those numbers got skipped for some reason.
For instance, they might have decided they wanted a new class of ship, the Widget Class, and they'd want 60 of them, so they set aside numbers 2530 through 2589 for them. Then the prototype, USS Widget NX-2530 turns out to have a ew major bugs, and Starfleet decides the design was redundant anyway, so they cancel the class and build 60 new Miranda's instead, and give them numbers 2590 through 2649 because 2530-2589 were set aside for ships that never got built.

The number do seem to proceed in some vaguely chronological order. That leads to the conclusion that Starfleet was building new Mirandas and Excelsiors a century after they were introduced, and long after some cool new ships were made. As others have pointed out, there may be some good reasons for this: while some other class had taken the name of "biggest", "coolest", or "most badass", those designs still did certain jobs very well, and were probably easy to build, and easy for new Federation members (who might be just a tad behind the cutting edge of Federation tech) to build.
Whatever the reason, it is pretty apparent that they did keep building them.

 

[Saquist]

One of the problems one encounters with the NCC numbers is that their rate of use suddenly jumps. That is, in the first hundred years the Federation used roughly 2000 numbers, but 100 years later they were in the 70,000's. One explanation for this is that they built tens of thousands of ships for some reason. Another is that many of those numbers got skipped for some reason.

I consider it an arms race.
Or at the beginning it was exploration.
The Federation was expanding exponentially.


If the Federation was only 20x20x20 LY Earth to Vega
Then 2000 ships= .25 That's four ships for every light year

But the Federation even at Kirks time was at least 75 lightyears wide:4 ships every 1000 cubic lightyears
Now that sounds like "we're the only ship in the sector" kind of numbers.

By the time of Picard he says in FC that the Federation spans 8,000 lightyears . I'm just going to take half of that at 1,000 lightyears deep against Voyager's registry: 4 ships in every cubic million lightyears

That...is alot of distance between you and the next available starship. Of course Federation ships aren't spread evenly but at points of interest. But there is still a border to patrol and installations to protect as well as regular trade routes as well as exploration endeavors.

I don't have a problem with the numbers expanding so quickly. Kirk was the beginning of the golden day of stellar exploration. For 8,000 light years I don't think they have nearly enough ships.

------
I originally did these calculations for Star Wars vs. Trek.
The empire could afford to have 25,000 star destroyers over an entire Galaxy because Hyperspace could get you anywhere in the galaxy from a day to month.

In Trek Warp is much much much more slower.
You'd want your border patrol and trade route patrols to be within 5 to 6 light years of response time since Fed ships are faster than almost anyone else. With the construction of ships on boarders routes and installation and planets that leaves a lot of empty space several million cubic light years of no assistance. That's a large desert of inactivity.

 

[Dukhat]

Constitution Class NCC-1700 USS Constitution

USS Constellation under Commodore Matt Decker was NCC-1017. I think despite the simularity between the two, the Enterprise and the Constellation were of different classes. That's how I personal account for the Constellation's hull number, so perhaps that class began with NCC-1000.

Yeah, that or call it a typo in canon. I mean, it's a problem, and it needs to be dealt with somehow, and I find none of the ways of fixing it particularly awesome. I prefer to pretend it was really NCC-1710, but I know full-well that I'm boldly going non-canon when I do, so maybe making it an older class is best.
Similarly, a lot of the canon Constitution Class vessels really aren't: they are never shown, and never specifically said to be a Constitution Class, but somewhere in fandom it was decided they were and it stuck. Take USS Republic: It is specifically stated that her "number" was 1371, but her class is never suggested, nor was there any dialogue making her particularly similar to Enterprise.

One of the problems one encounters with the NCC numbers is that their rate of use suddenly jumps. That is, in the first hundred years the Federation used roughly 2000 numbers, but 100 years later they were in the 70,000's. One explanation for this is that they built tens of thousands of ships for some reason. Another is that many of those numbers got skipped for some reason.
For instance, they might have decided they wanted a new class of ship, the Widget Class, and they'd want 60 of them, so they set aside numbers 2530 through 2589 for them. Then the prototype, USS Widget NX-2530 turns out to have a ew major bugs, and Starfleet decides the design was redundant anyway, so they cancel the class and build 60 new Miranda's instead, and give them numbers 2590 through 2649 because 2530-2589 were set aside for ships that never got built.

The number do seem to proceed in some vaguely chronological order. That leads to the conclusion that Starfleet was building new Mirandas and Excelsiors a century after they were introduced, and long after some cool new ships were made. As others have pointed out, there may be some good reasons for this: while some other class had taken the name of "biggest", "coolest", or "most badass", those designs still did certain jobs very well, and were probably easy to build, and easy for new Federation members (who might be just a tad behind the cutting edge of Federation tech) to build.
Whatever the reason, it is pretty apparent that they did keep building them.

1. There are lots of ways to justify why the Constellation has a lower registry number than the NCC-1700 Constitution, but saying that it's a different class isn't really feasible, since pretty much every official publication of the ship lists it as a Connie, and that was the intention of the people who made the episode. Besides, I'm not really keen on trying to make excuses for the decal guy in 1968 not understanding registry schemes for Star Trek.

2. I do, however, feel that ships that weren't shown (like the Republic) but were referred to in official publications as Connies can be overruled if the ship is shown to be different in a future onscreen appearance. (i.e. the Farragut in ST '09 is obviously one of those older vessels docked at the station, and not a Connie).

3. Re: large jump in registry numbers: I don't have a problem with this, beause it's already been shown that very small mass-produced vessels can have their own registry numbers too (i.e. runabouts, Data's scoutship from Insurrection, etc.). Just because there may be 75,000 registry numbers out there doesn't necessarily mean that there are 75,000 ships of Excelsior or Galaxy size.

 

[ThunderAeroI]

In order to prevent duplicate NCC numbers, I always assumed StarFleet administrtion assigned NCC numbers in production order.

For example the ship we know as 1701-D was probably assigned a different NCC number at production, but was renamed on dedication to something more fitting.

That or each startship has a VIN number which specifically calls out what that ship is and what its fittings are. We know each ship has a code to control it so I'm sure each ship has a serial number, a VIN number and other numbers associated with its fittings.

The NCC number may mean nothing at all and is randomly assigned to ships as they are dedicated.

 

[Unicron]

Interestingly enough, Bjo Trimble suggested that the Enterprise was a Constellation class ship rather than a Constitution class ship in the Star Trek Concordance, in keeping with the Constellation's lower registry. The Republic was depicted as a Connie in the 25th Anniversary game, albeit with a small error in registry (NCC-1373 vs the canonical 1371).

 

[Captain Robert April]

Interestingly enough, Bjo Trimble suggested that the Enterprise was a Constellation class ship rather than a Constitution class ship in the Star Trek Concordance, in keeping with the Constellation's lower registry. The Republic was depicted as a Connie in the 25th Anniversary game, albeit with a small error in registry (NCC-1373 vs the canonical 1371).

The notation of the Enterprise as a "Constellation class" in the '76 edition of the Concordance is a typo; elsewhere in the book, it's identified as a "Constitution class", a reference that dates back to at least TOS' second season, and the '68 edition of the Concordance does identify the Enterprise as a Constitution class.