Talk:Comparison of orbital launch systems

Launch cost

Fascinating that the launch cost columns are almost entirely vacant of data. It would be very interesting if more data could be obtained for comparison, because, apart from success rate and availability, cost is obviously the over-riding criterium for selection. Although I appreciate that cost figures can be highly dubious due to subsidies, marketing, bribes etc, I find it hard to believe that SpaceX can realistically compete with the Russians or Chinese on costs.1812ahill (talk) 18:16, 8 October 2012 (UTC)

Most vendors do not publish a list price for launches. I understand they prefer to negotiate each launch separately, given that we are still pretty much in a customization era for launches so each will cost differently. In summary, I understand the cost data are largely proprietary and not available to Wikipedia. Tarl.Neustaedter (talk) 21:26, 8 October 2012 (UTC)

Is it so hard to get source data from previous launches? I would think that one could compile a list of launch costs from IRS records, or other public annual budget sources that would allow for the averaging out of a LV cost per launch. No I don't have this info, but it must exist, and be realativly simple to get. The NASA budget is public, and it sites commercial contract expenses in it's annual expense reports. the FOIA give Wikipedia some leverage for the data, and it's not like were competing against the providers... Again just another two cents. --DarrenHensley (talk) 01:20, 7 December 2012 (UTC)

It is worse than that. Not only do many of the launcher manufacturers not release this data, the data that could be made available across some subset of the companies where it is released are calculated according to very different accounting standards in various countries. Add in the variety of economic systems (state-capitalist in much of the West, communist in others, etc.) and the problems grow worse. Independent of economic system, there are a a large variety of politically-influenced reasons to report costs differently.

Net: all of the so-called "cost" information, even when it is released, is simply non-comparable. The cost methodology used by one company, or one national government, will not be the same as the methodology used by another. So attempting to compare these (mostly missing) data in a Wikipedia "comparison" article is misleading to our readers of this encyclopedia. Cheers. N2e (talk) 00:57, 23 December 2012 (UTC)

Indeed the quarterly launch reports produced by the FAA include the disclaimer that prices include discounted rates for the US Government WatcherZero (talk) 22:27, 23 December 2012 (UTC)

This problem, after having no major edits/resolution since late December of 2012, finally DID get resolved with the elimination of the several cost columns, per a different discussion (below) on this same Talk page, with a consensus achieved in January 2014. See the section entitled: Launch costs on space shuttle. Cheers. N2e (talk) 05:40, 13 January 2014 (UTC)

SLS

SLS is planned as a family of rockets and not one particular booster. The variants are known as Block I, Block IA, and Block II. These variants have very different planned performance parameters. The SLS is the only thing in this comparison list that gives a range for Mass to LEO (70,000 - 130,000). Splitting this up as three different entries, one for each variant, will be more accurate. Ffejmopp (talk) 19:02, 31 March 2013 (UTC)

I posted the above back in March, but there have been no replies pro or con. Unless someone can give me a reason oterwise, I am going to edit the table to have three different rows for the SLS. Ffejmopp (talk) 18:38, 27 August 2013 (UTC)

I went ahead and did it, creating two SLS Blocks (rather than three) based on a NASA PDF I found on their website. there are a lot of references to a third intermediate block elsewhere, but that wasn't referenced in this PDF. Please feel free to add extra data and references. Ffejmopp (talk) 21:11, 27 August 2013 (UTC)

Changes to cost column

I've twice backed out changes from an IP address which has been changing numbers in the cost column. The first one is changing a cost from Euros to dollars (and putting Euros in parens), which I believe is incorrect - the rate of exchange is not fixed, the cost should be listed as priced, not as converted to some other currency. The second change has been changing numbers to have leading zeroes. Both of these have struck me as inappropriate, I've left an edit history suggesting the editor discuss this on this talk page. Tarl.Neustaedter (talk) 17:40, 21 August 2013 (UTC)

Falcon 9 1.1

Query on the Falcon 9 prices, the base design does feature a 25% increase in thrust however the price listed seems to include the recovery of the engines (which if performed reduces payload by 15% for a water landing and 30% for a hard landing as fuel is reserved for controlled braking) while still claiming the higher payload capacity and price-weight ratio of a non-recovered engines flight. WatcherZero (talk) 01:10, 8 December 2013 (UTC)

Hmmm. I guess I have a couple of comments.

1. All of the price comparisons in this article, done by different folks and different methodologies and with data from very different sources [some producer-produced "costs" (which could include/exclude just about anything) and some with closer-to-market real prices (prices (generally) available to multiple customers under similar terms and conditions)] are essentially non-comparable. It is likely a fool's errand for us in Wikipedia to try to place these cost/kg figures in a single column in this table as if they were comparable—and so the column should probably be nixed.
2. Having said that, the column is in the table today, such as it is. So assuming it stays, then: I seem to recall reading somewhere, since late last summer, that SpaceX has published two prices—one for payloads below a certain level (where, ostensibly, SpaceX will retain the option to fly the booster back and attempt a landing, and if a market materializes for "used" boosters, then fly it again) and another price for a payload above that level, where I am guessing that they would not attempt a fly-back and understand that trashing a booster in the business-as-usual-way (since the governments all did this in the 1950s onward) in order to get the higher payload and/or higher orbit. I don't have time to look for it just now. But look at the sources in the FH and several F9 articles; I believe SpaceX released a bunch of new spec info, including these prices, sometime around last summer.

Good luck. N2e (talk) 03:32, 8 December 2013 (UTC)

Here is the source I was thinking of: SpaceX Capabilities and Services page for the Falcon 9 AND Falcon Heavy. Looks like SpaceX is only publishing one price for Falcon 9, with 4850 kg to GEO: US$56.5 million.

However, they are being quite up front about the difference in prices for going to GTO with the Falcon Heavy US$77.1 million for payloads up to 6400 kg, and a price of US$135 million for GTO payloads of 6401 kg to 21200 kg. Looks to me like they intend to get back some or all of the FH boosters rocket when they send an FH to orbit for only $77M, and perhaps none of it when they send 21200 kg to GTO.

On the F9, and I'm just guessing here, looks to me like they may have cooked in "expendable" to their current price sheet, but that would leave plenty of opportunity for some price lowering action should competitive forces push them to it if they get the SpaceX reusable rocket launching system technology fully developed and in production. Cheers. N2e (talk) 03:24, 12 December 2013 (UTC)

Launch costs on space shuttle

We're having a back-and-forth on the Launch Cost column for the Space Shuttle. The reference triggering this is Forbes, which doesn't fit with the rest of this article. Other launch costs in this table are incremental costs, in some times simply prices charged, which may have nothing to do with the actual costs.

Tarl…

Check the Wikipedia page 'Space Shuttle Program', 'Budget' section, you find the text: “Per-launch costs can be measured by dividing the total cost over the life of the program (including buildings, facilities, training, salaries, etc.) by the number of launches. With 134 missions, and the total cost of US$192 billion (in 2010 dollars), this gives approximately $1.5 billion per launch over the life of the program”.[16] With a link to: ^ Pielke Jr., Roger; Radford Byerly (7 April 2011). Shuttle programme lifetime cost 472 (7341). Nature. Bibcode:2011Natur.472...38P. doi:10.1038/472038d. Retrieved July 14, 2011. There are many additional valid shuttle cost links such as: http://www.space.com/12166-space-shuttle-program-cost-promises-209-billion.html Which prove a final Shuttle program cost of circa $220 billion.. IMHO this table should to fairly/equitably compare commercial/private costs to Govt launcher costs.. Any private enterprise/business MUST cover all costs (e.g. development, staffing, overhead, etc) prorated in their unit product/service costs, or go bankrupt. They can't just subset costs to look good.. Numbers such as the $300 million ignore development costs, KSC shuttle costs (13,100 employees $2 billion/year), MSFC 'astronaut support' shuttle costs, etc. Final/comprehensive Nasa published official Shuttle Program costs numbers are over $210 billion, $1.6 billion/launch... and are over a decade newer than the currently cited 'Futron' 1990-2000 numbers...also even the Futron report STATES IT USES THE INCOMPLETE NASA PREFERRED NUMBER (incremental costs only) from the 1990s... invalid then and obsolete now, not including total end of program costs and lower flight frequencies.. The only fair/valid/equitable method is total program costs / # of flights... even this excludes the 'cost of money' which a commercial effort must cover.. The $300 million number is invalid, misleading, obsolete. Also, since when is Forbes less trusted source than 'Futron'? Calling facts you don't like a 'hit piece'? I can point to plenty of soft-ball Nasa biased 'fluff pieces'.. like Futron. -- this unsigned paragraph by ‎User:Wrwhiteal

The Forbes article doesn't meet WP:NPOV, since it's clearly a hatchet job, doing it's utmost to present the shuttle program in a poor light. The figure it's using for launch cost is the entire cost of the shuttle program, including costs which had nothing to do with shuttle launches. Since this table is about comparisons, the costs presented in that article don't belong here. Tarl.Neustaedter (talk) 05:50, 11 January 2014 (UTC)

Tarl: I've got references from the Wall Street Journal, Forbes,Space.com, which cite Nasa itself as the source of the $210 billion shuttle program cost figures... IMHO the $1.6 billion/launch is understated, as it fails to include time weighted 'cost of money' and/or opportunity costs that a private enterprise would pay... basic cost accounting... I agree that shuttle launch costs are 'out of kilter' with other rockets... THAT'S THE POINT... they are 'out of kilter' because of inherent Shuttle/Nasa massive inefficiency/waste, not because of unfair, improper or misleading accounting.... Just because they are 'out of kilter', far higher than others, or just because some folks don't like the fair cost, doesn't make it wrong... if you want to compare launch system costs with any meaning, IMHO you must compare program costs / # flights... a commercial/private provider MUST cover ALL it's costs..otherwise they go bankrupt... e.g. SpaceX Falcon costs must cover all the Falcon program... they can't just ignore development costs/overhead/staffing/etc..... Govt should to... otherwise, any Govt could subsidize costs, play accounting tricks, ignore development/overhead, and offer the launch for free.... People have or will use this list to judge the relative merits, inefficiencies, value of private vs Nasa provided launch vehicles... we should use strict basic, strict accounting principles.. not ignore costs/subsidies and produce misleading results. Each shuttle flight did cost taxpayers $1.6 billion... -- This unsigned paragraph by ‎User:Wrwhiteal

All you are saying is "my data is good", I'm telling you it's not appropriate for this use. What you seem to be intent on demonstrating is that the program cost the U.S. government a huge amount of money, whereas In this comparison table, the costs are intended to be what it cost the END USER. What did a CUSTOMER pay to launch a payload. The total cost of program isn't of interest in this table, and isn't what ANY of the other prices listed are based on. This table is a COMPARISON, please do not change systems for one line item only, it entirely removes to ability to compare costs to launch customers.

Also, please be careful to always sign your entries (use a quadruple tilde - ~~~~ - which will be replaced by your name and talk page pointer) on a talk page, and please do not intersperse your comment inside someone else's text without off-setting it with colons (as I've done for you after the fact). It's important so that other readers can tell who wrote which words. Regards, Tarl.Neustaedter (talk) 23:33, 11 January 2014 (UTC)

I'll stay agnostic on the debate between Tarl.Neustaedter and User:Wrwhiteal on Space Shuttle costs for now, as I haven't finished thinking about it. I will however note that the entire enterprise to attempt to put some sort of standard launch cost per kg in some table in this article is likely a fool's errand. The data are simply not there, and not provided in any way close to similar accounting methodologies between the various orbital launch systems so as to make them comparable, as discussed in above in the section on Launch costs.

The data that could be interesting might be a standard market price—which in economic and accounting thinking is a very different thing from "cost"—but we simply don't have standard market price data, per launch, or per kg, for this large set of orbital launchers.. Bottom line, we don't have cost or price data that can make the numbers in that column comparable, and therefore we ought to delete the entire column so as not to mislead the readers of this article. Cheers. N2e (talk) 04:15, 12 January 2014 (UTC)

Deleting the columns entirely makes eminent sense. Most rockets don't have the data at all, the ones that do have data which is distorted by governmental policies rather than economics. Deleting the three columns, and while we're at it, the extraneous duplication of name column on the end, is something we should do. Tarl.Neustaedter (talk) 05:40, 12 January 2014 (UTC)

I agree. If you compare the PPP of the total program cost of the Zenit launch program to that of the Falcon 9 program, the economics of the Zenit look stupid. But if you add all of the free R&D and below cost tech support that SpaceX got from NASA, things rebalance a bit. So corporate vs government programs can't be compared. And you run up against similar issues trying to compare rockets developed in a planned vs market economy. There is simply no reasonable way to add up all the various costs. They are uncomparable. I say delete the columns because we'll *never* be able to make those costs comparable. — Gopher65_talk 14:34, 12 January 2014 (UTC)

Looks like an emergent consensus exists that we should delete the cost columns, because the data are not comparable. As far as I'm concerned, it would be okay now for some editor to begin to boldly make those edits, based on this consensus.

I will add a comment to "The wider problem" discussion below in a bit; but I don't think that need necessarily delay the step 1 on a subset of that on which there is pretty decent agreement. N2e (talk) 04:21, 13 January 2014 (UTC)

Done. Revert and discuss if adjustments are desired. Tarl.Neustaedter (talk) 05:39, 13 January 2014 (UTC)

The wider problem

I only noticed this discussion today. I think the issue with the prices is part of a wider problem - not just with this article but with those like it. Absolutely everything in this articles are subjective, yet presented as fact for the purposes of making it "comparable". Let's start with the subdivision of the lists - How do we define which rockets should share an entry and which should have separate entries? Some rockets are split down to configurations - such as the Delta II - some are split even beyond this - Ariane 4, for example - while others - Taurus, Antares, etc, have a single entry. Many aren't subcategorised by upper stage - Soyuz-2, for example.

The "manufacturer" and "origin" columns may seem neat enough, but there are plenty of cases where this isn't in reality the case. Zenit-3F is a good example. We list it as Ukrainian. Yes, its manufacturer is based in Ukraine, however the rocket is used by Russia, has a Russian upper stage, and the Zenit design itself dates back to the Soviet Union - using Russian technology and know-how. Ukraine never developed the technology to build rockets indigenously, it inherited that technology from the USSR. Then there's Antares. We list it as an American rocket built solely by Orbital Sciences. No mention of the Ukrainian (Zenit-derived, i.e. former-Soviet) first stage with Russian (also former-Soviet) engines. The upper stage was built by ATK. But we just list OSC.

Next, there's the question of payload capacity. The table just gives a single figure for LEO payload - which is completely meaningless. LEO isn't a specific orbit, it's a whole range of many, many different orbit types. A rocket will have a significantly different payload capacity to a 150 x 150 km x 0° orbit as to a 1,999 x 1,999 km x 180° one, yet we're tarring them all with the same brush - there is no standard. It also ignores other factors - launches from some sites require more energy, either due to the high latitude of the site or having to dogleg during ascent. And even then, if identical Soyuz rockets launched to identical orbits from Kourou, Baikonur and Plesetsk, you'd find that all three would have different payload capacities.

The same goes for payloads to other orbits. The rest's just as bad - I won't go into costs, but obviously launch counts and first/last flight dates are dependent on how you split the list and what tests flights you include/exclude - as well as how disputed launches (such as the February 2013 Safir) are included. Status is probably the only thing that isn't completely subjective, but again it does depend on division, and also on whether a system's retirement/abandonment/cancellation is publicised or if the rocket is simply allowed to drift into obscurity like China's KT-1 appears to have.

So in short, what is the point of having a comparison when the data simply cannot be made comparable? Maybe we need to review whether this article continues to add value to the encyclopaedia, or whether we should look for better ways to display this data. --W. D. Graham 16:26, 12 January 2014 (UTC)

The origin of this list came about from a merger of several ad-hoc lists of launchers (Heavy Launcher, Mid-sized launchers, ...). The intent was to provide a single list of launchers without the complications of arguing about which launcher belonged on which list. It appears to have received the name "Comparison" as a way to add value. I think the list still has value as simply a comprehensive list, although I doubt the utility of all the single-launch variants of one booster. Tarl.Neustaedter (talk) 17:18, 12 January 2014 (UTC)

It was a comparison from the beginning - and while the merger did cut out some of the problems, it only addressed the tip of the iceberg. I agree that a comprehensive list would be useful, but I think the format needs to be simplified, and we should probably consider a merger with List of orbital launch systems. --W. D. Graham 17:42, 12 January 2014 (UTC)

I agree with the general thrust of the discussion above in "The wider problem". Yes, such a problem exists, and yes, I hope a consensus can be reached to do something about it. Might I suggest someone make a specific and conscise proposal? Not sure where that ought to be made, as it would affect a number of articles. But the folks who've commented on this Talk page could be invited with Talk page messages, and perhaps a note on WIkiProject Spaceflight could be left.

One other thought: since it may be a bit more challenging to develop a consensus on a broader set of articles with the same or similar problems, and that consensus may or may not ever be reached, I don't believe that the discussion on "The wider problem" need necessarily hold off action on the more limited consensus we've already reached on the cost columns in this particular article. Cheers. N2e (talk) 04:29, 13 January 2014 (UTC)

While the figures in LEO and GEO aren't precise due to a number of variables, they do give us order-of-magnitude information. There's a big difference between a rocket which can launch 100 kg to LEO vs one that can launch 100,000 kg, which we can see in this table. And by the way, in the launch business, "to LEO" generally means "to the lowest orbit that won't fall on our heads" - it's a short-term orbit, either a transfer orbit or a short-lived satellite. The higher orbits that are possible under the umbrella "LEO" aren't generally used in measurements of launch capacity of LEO. — Preceding unsigned comment added by Tarlneustaedter (talk • contribs)

Not all LEO mass figures are for a minimum orbit; some are typical orbits for those rockets; such as sun-synchronous or 500km circular orbits. --W. D. Graham 12:26, 13 January 2014 (UTC)

Ah yes, the "which LEO?" issue again. As an example of a (purposeful) source error, SpaceX's Falcon Heavy page use to have some figures comparing its 53 tonne launch capacity to the capacity of competitors. From a marketing perspective it looked really impressive, but they also had the poor sense to include both the orbital inclination and orbital altitude information for each rocket. I think it suffices to say that there is a big difference between launching 53 tonnes to 160km @23 degrees and launching 53 tonnes to 350km @60 degrees:P. In addition to inclination and altitude, another far more minor issue is that the launch site choice matters (slightly) to the total payload capacity of the rocket.

I wonder if there is a formula to convert payloads back and forth based on those variables? — Gopher65_talk 13:24, 13 January 2014 (UTC)

whoa boys.... I believe that the previous $300m shuttle costs were wildly misleading, but throw the baby out with the bath water? remove the entire (estimated) cost columns? I would like to be able to compare launcher cost (efficiency), a preeminent attribute of a space launcher... cost per lb to orbit is crucial... IMHO a valid ESTIMATED cost should be provided, but just calculated properly and noted as what it is... apparently we do have substantial information to govt subsidies, hidden costs, and estimates can be made and be added in (for Zenit, for example). Nasa itself provides the total shuttle program cost.. if someone wants to include price (which probably wildly varies), then why not add another column...I believe we could have a fair, balanced estimated cost/launch, based on best information... once we agree on a methodology of covering all known and estimated costs.... IMHO this is very pertinent and useful information... however, I go with the group consensus..Wrwhiteal (talk) 15:01, 13 January 2014 (UTC)

"Once we agree on the methodology", the article is in danger of becoming original research. I'd argue the same could be said about Gopher's suggestion above regarding converting payload masses. --W. D. Graham 15:07, 13 January 2014 (UTC)