From a metagame perspective the player has invest quite a lot in nanofabrication. On the flip-side, without the actual blueprints for a CM, this could really bog the game down without access to accelerated simulspace (x60 to get the design to under 5 hours, assuming no major failures on roles) If simulspace is available, he could also fork into it and do the designs while doing other things. The big question is, what does the player want the CM for?

I would be inclined to allow this on some level. The first cornucopia machines had to be built without such convenient technology. As such, I would rule that it is possible for a player to build a cornucopia machine; it would however require plenty of work, proper know how, and the right tools. What kind of nanofab is the character trying to work with? I don't think you can do what your player is trying to do just with just any nanofab. A low quality maker (makes awful tasting food paste) probably wouldn't have the mechanisms needed to produce anything resembling a nanofab. Various fabbers probably have their mechanisms optimized to build certain kinds of materials (like carbon or electronics) as a way to keep costs down. This would have the added benefit (or drawback) of making it difficult, if not impossible (some fabbers might be better than others), for a fabber to produce the next step up. If the player intends to design blueprints to create such a high tech device on a lower quality machine, then the player might suffer a penalty in the designing process as they try to figure out some way around the hardware limitations of the device. A -30 penalty might do well for any fabber not built specifically for this job.

Erenthia wrote:

From a metagame perspective the player has invest quite a lot in nanofabrication. On the flip-side, without the actual blueprints for a CM, this could really bog the game down without access to accelerated simulspace (x60 to get the design to under 5 hours, assuming no major failures on roles) If simulspace is available, he could also fork into it and do the designs while doing other things.

Forking to get the 5 weeks of work done in 60x acceleration simulspace will work, but it creates problems with merging your forks back together. In 14 hours, the fork would have been separate from the original fork for 5 weeks. In my books, that would count as being separate for over a week for merging purposes. That will cause a -60 penalty to fork merging and will guarantee some memory loss and stress damage even on a success.

You might want to bring up the energy issues too. An industrial scale fabber is going to be a real power hog when it's running. You might not be able to get permission to plug it into the local grid, and serious private power generation can be tricky to obtain.

Shredicine wrote:

Another train of thought, is the classic quality-loss in recording over another recording, as the case with audio cassettes. Imagine that once an audio cassette gets to its 3rd or 4th generation in recording, the quality is so bad, that it almost sounds like static. This same generalization can be applied to this concept for Nanofabricators producing endless Nanofabricators. Perhaps after the 2nd one is made, the duplicates become so corrupt, to the point of producing undesired and adverse prints. This could actually seed some good adventures, IMO. Just something to think about :)

This would imply that there is a manufacturing method that's more precise than any nanofabrication that is used to create the first generation fabbers.

I don't think that cornucopia machines will have the same issues with quality loss over many generations of copying a copy as cassette tapes would have. Copying cassette tapes suffered from the problems of imperfect copying methods combined with the lack of error checking and fixing. While its unlikely that the copying methods for CM would be perfect, it would be better than cassettes. In addition, unlike cassettes, CMs would have error checking to test how similar or correctly made something is. A combination of hardware/software checks could find some errors, and actual attempts to produce complex objects (whose sole purpose is to test accuracy) might find others. Examples might include producing: -Tests to see if the machine can assemble a pattern over and over. For instance, it might use combinations of simple patterns (like 1 2 3 1 2 3) to complex patterns hundreds of atoms long. -Tests to see if the machine can use every square inch of the printing surface. For instance, printing an entire block of carbon (like diamond). -Tests to see if every part of the machine works. This might require a special program that overrides the normal printing optimization programs (I'm assuming they have optimization programs) and forces the machine to use every gear it has to do something. It probably needs to flush its system afterwards so whatever is produced is probably gibberish (and likely varies depending on the differences between models). It should be noted that one of the reasons why technology is valuable is because it is reliable. You don't get a ticket to fly on a plane if the odds of landing is 50-50. More advanced forms of technology tends to be more reliable. This has already happened with music cassette tapes to digital music files. You can expect a file copied to be the same as the original. If you are unsure, you can find tools to test how similar files are (and what the differences are).

We make tools using tools. Where there is degradation is when the process 1) is approximate (analog copies with noise), 2) the new tools are made out of inferior materials, or 3) there is a lack of skill in applying the tools. Nanotechnology gets around 1 by being digital, 2 by using discrete atoms, and 3 by running automated scripts. ...except that in practice 3 can be a headache when rolling your own CM. You will need to make something that reads the blueprints and implements them as actions undertaken by the component assemblers. A lot is going to be open source software libraries, but the final part will be gluing them together into a system that can take standard blueprints and get this particular architecture to implement them. This is where there is a myriad of hairy details: nothing impossible or even hard, just loads of details. Just consider dealing with the no doubt multiple blueprint formats, the special cases of handling different materials, dealing with different kinds of feedstocks, DRMed and obfuscated blueprints, adding security (there are rather nasty malware for fabbers out there... especially hidden in modules used in making free CMs) and so on. The PCs skillset sounds about right for the job, but it is a lot of work. I would rule it is doable, but the quality of the product depends of a big time investment: kludging something together that prints basic blueprints might be quick, but expect it to crash if the blueprint uses the wrong color or you feed it a slightly off feedstock. Hmm, malware. I think Ken Thompson hacks for fabbers are a real possibility: http://c2.com/cgi/wiki?TheKenThompsonHack http://cm.bell-labs.com/who/ken/trust.html Sure, after the Fall everybody went over their code and assemblers with a fine comb, but there are so many interests out there who might want to reintroduce them. This might be a good reason to make your own CM. As Ken said: "The moral is obvious. You can't trust code that you did not totally create yourself. (Especially code from companies that employ people like me.) No amount of source-level verification or scrutiny will protect you from using untrusted code. In demonstrating the possibility of this kind of attack, I picked on the C compiler. I could have picked on any program-handling program such as an assembler, a loader, or even hardware microcode. As the level of program gets lower, these bugs will be harder and harder to detect. A well installed microcode bug will be almost impossible to detect. " Unfortunately, the way around this kind of hack is to go deeper - not just stringing together open source libraries and blueprints but writing them. Not just printing a sheet of nanoassemblers but making the seed assembler and have it make the rest. And so on - the price of total security is endless toil.

You didn't mention whether they had a blueprint for a Corn Machine. They've got to acquire that, either by cred or by rep. From there, fabbing it should be possible, provided they have Hardware 'lectronics to put together the pieces. If they're in the inner system, the IP cops will kick in their door waving the four-four. If they're in the outer system, and that shit gets abusive in your campaign, the PC has been known to mount punitive IP enforcement strikes rimward, just because they're cockmasters like that. Remember: outer system post-scarcity is awesome, but from a game balance standpoint, it can always be countered by inner system corporate interests being extraterritorial dickheads about it.