|
Post by CELS on Mar 13, 2005 17:34:44 GMT -5
My apologies for my long absence from this thread. Obviously, the discussion has gone down a different road than I had in mind. Warp drives designed by computers the size of mountains with A.I.... Ok...
But I will say this...
On viruses... I agree that because an internet like the one we have today is terribly vulnerable, it is not commonly seen in the Imperium. I mean, the viruses and virus-creating programs we have today would probably have caused a lot of problems a few years back.
On servitors... I personally think that this is an important subject. Apparently, everyone thinks this is just me rambling...
Once people are ready to get back to working on the Adeptus Mechanicus, its position in the imperium, its non-military divisions, etc, I shall return with great enthusiasm. (Or, perish the thought, someone offering a coherent argument to explain why servitors are used instead of robots)
|
|
|
Post by Kage2020 on Mar 13, 2005 18:33:37 GMT -5
<swallows> My apologies for my long absence from this thread. Obviously, the discussion has gone down a different road than I had in mind. All threads do that. It is one of those things that you must but accept... Warp drives designed by computers the size of mountains with A.I.... Ok... Not gonna happen, so you don't have to worry about that. I agree that because an internet like the one we have today is terribly vulnerable, it is not commonly seen in the Imperium. I mean, the viruses and virus-creating programs we have today would probably have caused a lot of problems a few years back. That's fairly... strange. If you have any 'system' there are always going to be peole that work against that 'system' and try to find ways around it, through it, or who just try to undermine it. As technology improves, so does the means to destroy that technology. In other words, just because you have 'viruses' in the modern world doesn't mean that you're not going to have an "internet" in a future world because of those 'viruses'. That is not to say, however, that one cannot offer a caricature of the real world and say that this is the cse. With reference to 40k, however, methinks that given the descriptiosn present in those novels (etc.) that tend to move away from the "battle report" style, it is reasonable to assume that worlds with advanced technologies likewise have advanced communications. I'm not saying that it is the same as the 'Net - with MSN v.41000 or whatever - but rather that limited extrapolations of 'virus=bad' may not be the way to go. In other words...? Let us keep an open mind. I personally think that this is an important subject. Apparently, everyone thinks this is just me rambling... On another thread you told me to "give it up", with attendant smilie. I suggest the same thing is appropriate here. No one is questioning that servitors are used, merely that the ab initio revision of servitors might be the simplistic - and boorish - response. Rather, itis more interesting to find historical, political and, well, just plain human reasons why the current situation exists. I shall reiterate for clarity: The current Imperium uses servitors almost exclusively. See? That's not too difficult. What is more difficult is seeing beyond the Revision, and integrating it... Actually, I lie. It's not difficult at all. The method by which this can be achieved has already been suggested: we utilise it already in reference to the factions, philosophies, paradigms and cults of the adeptus mechanicus more so in reference to a compartmentalised Imperium. The 'question' of servitors is not a question in terms of representation. The complexities derive from the above and the fact that they are merely robots by another name. Or, perish the thought, someone offering a coherent argument to explain why servitors are used instead of robots This has already been done in reference to the 40k universe, CELS. Simply put the servitors are one of those things taht you have to accept given the Image of the 40k universe. Given the original 'fluff' on robots, however, it behoves us to explain the paradigm shift beyond GW's own shift from sci-fantasy to technofantasy (in terms of genre). In short, let us give a "political", or "cultural" reason for an entire shift to servitors. Let us not disclude robots, but minimalise them. And let is most certainly not black box them... Age of StrifeThis is always more problematic, when taken into account with the latest 'fluff'. What options are there other than Philip's taking of Herbert and Anderson's universe of the great Frank Herbert? (E.g. the whole Omnius approach.) While a reliance on technology can be seen as an interesting caricature of current humn society, taking it to extremes is not the way to go. In other words, let us not at all put AIs " in control". As cliched as it might be, though no more so than the alternative, let us examine means by which the 40k set up can be explained without necessarily "plagerising" Herbert/Anderson's approach.. Origins of the adeptus mechanicus... The failure of "rad shields" is problematic only insofar as it ignores potential 'realistic' colonisation of Mars. It is a part of the 'fluff', however, and thus must be utilised. Ergo, failure of the terraforming systems drove the people back into the original (i.e. underground) habitats. They had to abandon the surface ones as the 'terraforming' failed. (I hate it, but instead of 'natural' terraforming, i.e. alteration of albedo, etc., we'll have to stick with some horribly 'magical' technological transformation.) What does this mean? We modify the 'fluff' only minorly. We maintain the assumption that 'technological' means of terraforming would be used (anyone fancy research the different methods?) and that their failure drove people into the old habitats (i.e. underground). You get the same approach but without the mystical nonsense. The major place that it woud diverge is the mystification of technology, and this is something that I see as important. I find Philip's approach that technology is outside of the hands of 'humans' to be boorish. Rather, human nature comes to the fore in the Age of Stife on Mars. Rather than the scientists be excluded, we have the "Rise of the Nerds!" (speaking as a 'nerd' himself). They find themselves in a place where their knowledge is reinforced as necessary... So rather than having those without facility trying to grasp the knowledge of the past, we have those in knowledge trying to maintain their position... Kage
|
|
|
Post by DesertGhostExarch on Mar 13, 2005 22:46:06 GMT -5
(anyone fancy research the different methods?) Eh. Most terraforming theory (after a quick search) seems to follow albedo manipulation or more general "heating" methods (such as fanciful means like giant mylar mirrors or "somehow" introducing lots of greenhouse gasses, or the inevitable "nuke the polar ice caps" solution, which might very well be why even a terraformed/"atmosphere-thickened" Mars has problems with radiation, although that might depend on how "clean" the nukes were...), which relate also to the idea that "atmospheric thickening" would be a good idea (which also has its own theories associated with it, like hydrogen dumping). I imagine others include artificial creation of plate tectonics or magnetospheric manipulation of some sort, as plate tectonics and magnetospherics seem to be important for some reason or another. There's also the typical nanotech theorizing.... And that's just to make it suitable in the long-term for anaerobes and lifeforms not quite as complex as plant-life, I think. From a quick search....
|
|
|
Post by Philip on Mar 14, 2005 3:57:32 GMT -5
Hi Kage, I thought we weren't going to discuss my ideas here? What options are there other than Philip's taking of Herbert and Anderson's universe of the great Frank Herbert? (E.g. the whole Omnius approach.) *raises eyebrow* While a reliance on technology can be seen as an interesting caricature of current humn society, taking it to extremes is not the way to go. In other words, let us not at all put AIs " in control". As cliched as it might be, though no more so than the alternative, let us examine means by which the 40k set up can be explained without necessarily "plagerising" Herbert/Anderson's approach.. plagiarism? Moi? I find Philip's approach that technology is outside of the hands of 'humans' to be boorish. Ah, this is where the problem lies: you aren't listen to me, you trying to do that whole 'pigeon hole' thing that you do ;D However, I would point out that I never said the technology was outside human control... Seeing as I'm being accused of plagiarism I think I should make it clear that I am not, and explain this point of control/ Artilects a little further; Premise – humans in control To give a contemporary example: I used to do all my banking by actually going into the bank armed with bits of paper and conduct all transactions via a clerk. Now I use direct debit and online banking and it is much quicker and easier.
Direct Debits: The Bank's computers use my account to pay bills automatically - the Bank's computers run the whole thing, but I wouldn't say that the Bank's computers are 'in charge' or 'in control'. I can override my use of direct debits and do it the old fashioned why if I wish too.ArtilectsThe same would hold true for the Mars-Artilects or even the STC-Artilects. The point is these Artilects provide a service and automate many systems to make them run smoothly. Humans would be kept informed of what is happening, an could override the system. For example lets use City Design: the STC-Artilect could not only design an urban regeneration programme, new buildings, roads etc, but it could also cost everything, contact suppliers/ contractors and gain quotes (actually it would have the contractors/ suppliers information stored as part of its memory, so the Artilect would have 'accurate' information compared to the humans tendency to give low quotes), and as it is tapped into everything, give the most comprehensive 'plan' fully integrated and optimised for that society it is designing for. It would also include a 'world simulation' and any changes would come up in real time. All the human has to do is take the plan 'paint by numbers' (or hit the enter key) Now a human could override the system, but why would they? The Plan they are given is beyond anything a human could ever hope match in both scope and detail. With time I think the humans come to trust the Artilects. Sure the humans can override the system but after a while they just don't (it's too expensive, and explaining why you screwed up the Artilect's design gets a little bit embarrassing – I can see it now; 'I thought...' 'What? That you where smarter than the Artilect?' Your fired). After a few thousand years, maybe they humans just accept Artilects are part of the system, I don't think a human would ever think of these Artilects as a threat, no more than cash-machine are a threat or direct debit. Lock downThe STC-Artilects recognised that there was a Psyker problem and would instantly recommend isolation of the 'infected humans'. Which I think the humans tried to do, in a humane way, but in which they would obviously fail. I think the final STC-Artilect lock-down would be an embedded order from the military to secure a world in dire-straits. In essence the STC-Artilect didn't make the call, when a certain situations parameters are reached, 'military override' kicks in. However that don't stop the humans blaming the Artilect. Seeing as the lock-downs occurred after warp-travel was in effect (has to be, in this scenario is to the trigger for the increase in Psykers) the STC-Artilect would issue a warp distress signal for back up. This is the ironic part, as I'm sure this lock-down protocol would have been added later after warp-travel had been invented as a 'refinement'. Before warp travel, I think that the STC-Artilects where independent, and if a similar 'Psyker infection' situation occurred before the warp-age the implementation of the lock-down would have been left entirely to human discretion on any given world. I think that during the warp-age and the consolidation into a great 'empire/ federation' lead to the inclusion of the military's 'lock-down' protocol as the military always figured that they could get to the world via warp travel. In a universe before the Psyker problem, the humans of the DAoT really didn't think normal humans could spontaneously transform into Alpha Psykers, that was the stuff of fantasy and was just 'crazy talk' among the practical, wise, seen it all/ know it all society that existed during Golden Age of Technology. Betrayal and focus The machines never betrayed mankind: man betrayed man, the way it has always been. In this scenario I have outlined, the Artilects are the real bystanders; looking on as humans evolved into something beyond its comprehension. Humans are centre stage (aren't we always) and become even greater than the machines they had come to rely on. Nothing dazzles brighter than an Alpha class Psyker. Philip
|
|
|
Post by CELS on Mar 14, 2005 6:26:39 GMT -5
All threads do that. It is one of those things that you must but accept... Well, you can always try and push it in the other direction Of course. A billion servitors can still be killed if you have biological weapons. But then, biological warfare is a nasty thing on any world. I beg to differ, but I'm glad that you don't question it. But that's what I'm looking for, my dear Kage. Then please show me the way, for I am quite lost. This explains why the planet Mars changed, but doesn't explain why the culture changed, or how the Adeptus Mechanicus came to be. And what events led to the "rise of the nerds"? How did they seize power, and why did it only happen on Mars? And why did they lose touch with the traditional way of science, and start with all that mystical mumbo-jumbo? Why did they suddenly get all religious? Where did the Machine God come from? And.... what about the servitors? ;D
|
|
|
Post by Philip on Mar 14, 2005 7:41:11 GMT -5
Oh, I got a little side tracked, back to the thread; (Or, perish the thought, someone offering a coherent argument to explain why servitors are used instead of robots) I would say, ‘cost’, building a machine equivalent of a human mind would be expensive and very difficult, whereas growing a generic bio-engineered clone would be much easier. The Imperium can integrate biological and mechanical with ease, so there really is no need for a machine mind for a servitor, a human mind will work just as well (if not better) for it’s function and cost virtually peanuts in comparison. Also growing these ‘bio-engineered clones’ would be relatively easy in comparison to the complexity of chip creation, no more difficult than growing pumpkins, well, maybe a bit. Growing these clones would be easier and less complex that modern day CPU creation, and modern CPUs are pathetic compared to an actual machine/ artificial intelligence. All the really high-tech stuff would be in the programming of the servitors bio-mind and the machines to do this are reusable on site systems, and can manage it on a production level scale. The really high-tec part is where it is needed, and only where it is needed: in an Ad-Mec facility, not wondering around in the wild as a servitor (If a servitor is captured do those that have captured them gain an insight to the Ad-Mec’s technology? No! They get a squiggly bio-mass which they can’t access, the only interesting bit would be the interfaces – but that probably explains the dodgy back-street bionic upgrades I’ve heard tell about). Machine intelligence isn’t needed, and since the fall of the STC, err, that’s going back to my ideas, so I’ll cut that off right there. Suffice to say, Artilects/ machine intelligence etc is only needed at the high end, and they would be under wraps in a secret Ad-Mec facility (on the edge of town… err, maybe not ) Servitors make a lot of sense from an Ad-Mec point of view, but more importantly could a world really compete with the super cheap servitors of the Ad-Mec with a machine intelligence powered unit? The Ad-Mec produced servitors are going to be less than a fraction of the cost in comparison to an MI/AI equivalent and it will work just as well (and be more robust in the long run – bio-system have good staying power). Bio servitor = 200 credits + options MI servitor = 20 billion+ (no, more, way more, a world couldn’t afford it), massive restrictions, classed as military grade munitions, huge upkeep cost, crashes, everyone will want to steal it, huge insurance costs, needed armour storage facility to house it just to get the insurance in the first place etc. It is the difference in cost of having a child (they are basically ‘free’) and actually building a child from scratch in a factory via artificial means – ie building cell walls, injecting DNA with machines, oh wait, making DNA strand from scratch! Bloody hell… oh that’s hard (unbelievably expensive and virtually impossible). In this instance both would have the same upkeep costs In conclusion biology is far simpler and a far more elegant solution and a very effective way to build ‘impossibly complex’ structures ‘god’ was a great designer! And the Ad-Mec are smart. So servitors are a reasonable and cost effective way to meet the design requirements, not just because ‘GW says so’. Simple really, hope I didn’t step out of fluff. ;D Philip
|
|
|
Post by CELS on Mar 14, 2005 12:00:49 GMT -5
Thanks, Philip. At the risk of being rude, that's the first time in a while that you've made a post that I found useful. I hope you stay involved in this discussion. In nothing else, it might motivate me to take a look at your threads in the general forum.
Now, I'll try to reply to your argument. Unfortunately, I've got a pizza in the oven, so I have to go in five minutes. I'll try to be quick.
I realise that servitors are much cheaper than androids with artificial intelligence, or robots with machine spirits. The question is, why does the Imperium use so many servitors. Your argument explains some of the servitors in the Imperium, but far from all of them. What about the servitors onboard Imperial space craft? The servitor gunners, for example, or the many servitors that made up the crew of the sprint trader in the Eisenhorn novels. The Essene, was it? Computers could probably do all those things, especially operating guns, a lot better than servitors. Unless, of course, you're installing those computers into servitors, but then you have to wonder about the point of servitors in the first place.
Why use servitors in factories, as human lift-trucks? Surely, a primitive and cheap robot could do the same. You don't need artificial intelligence to assemble a tank, but still you have servitors doing that kind of work. You have servitors carrying people's luggage, you have combat servitors. Combat robots, I dare say, are nigh on non-existant in the Imperium. Why?
|
|
|
Post by Philip on Mar 14, 2005 12:59:47 GMT -5
Don’t forget cogitators!
As for using bio-matter, the Ad-Mec wouldn’t care if it fell into the wrong hands, what is a techno-thief going to do with it? Reverse engineer it?
As for a fork lift truck servitor, well you can tell it where to go, it is reasonable smart and can improvise. May not get a good conversation out of one, but it will do its job with ease.
As for Gun-servitors, they are good at target recognition (visually without any radio indent etc.), could be completely isolated and only recognise voice prints, and with more of the brain matter devoted to targeting and actually linked into it's weapon it could be unbelievable good shoot against moving targets (not that the fluff would have you think this!). Also humans can make decisions very fast, and if piped into the reptile part of the brain (automotive response) can use reflexes.
So; Dumb as a plank of wood = Cogitator Any type of intelligence or decision making = bio based servitor.
I think that if it is complex, a bit of bio-matter goes a long way. Even the ‘brain matter sandwiched between two slide of glass with life-support/ recycling’ is the way to go when things get small.
Servitors are just humans with no emotions and are very focused. With the lack of emotion I guess chaos has a hard time affecting them (though not impossible). So a servitor in its field is as flexible as a human, but without all the overhead.
Oh, and servitors don’t crash, and recognise what side they are on: there is no way to remotely tamper with a servitor, to change its target recognition for instance, and unlike a human you can’t bluff it or plead with it as it has no conscience.
So in the case of a ship run by servitors, what would you hack?
Philip
Edit: Added And they're cheap!
|
|
|
Post by Kage2020 on Mar 14, 2005 23:32:55 GMT -5
I've lost the energy to continue to reply. It ends here despite my knee-jerk reaction and not being last-worded. Thus I will not reply to any of it, even the misrepresentation. Enough. As to the thread... And a computer virus 'destroys' computers, whether software or, in many 'cyberpunk' style narratives, hardware. It's a case of "six of one, half a dozen of the other". Either can be used to justify the exclusion of the other but, plausibly, the justification is fairly meaningless. Thus other reasons must be analysed, which is what you're trying to do. Again, though, ab initio removal of 'robots' merely because of a lack of comfort...? That, to me, doesn't entirely make sense but then again the arguments thus far applied for them existing in exclusion to robots other than "GW says so" lacks strength. It would be just as reasonable to add "So nyuh!" on to the end! As previously suggested, robots were a part of the 'fluff' and - all things considered - a reasonable part of the 'fluff'. But they are now considered to be somewhat... gauche. All that is requested is that something other than, "Erm, GW says so" approach to Revisionism be avoided and rather a reasoned 'narrative' utilised. Possibilities have been othered, even if in brief or even tenuously, but no mention has been made to a timeline. The 'heresy' of robotic intelligence could, for example, have been enforced within the first millennia (few millennium, or whatever) of the Age of the Imperium. The possibilities that allow the integration of the 'fluff' are manifold, rather than an exclusion based upon the current Image and then sweeping back into everything that was done before. It would be like taking a few lines of the 4E rulebook and, from there regardless of the context, stating that both Navigators and Astropaths are required for warp jumps and then creating ad hoc reasons why this might be so (i.e. rather than actually questioning the 'fluff'). So, the basic suggestion absent of specific features would be: (1) robots existed in the past; (2) robots do not exist in the "Imperium" in the present (though they can still exist, and may even be produced by 'radicals'); (3) they are considered 'heretical' from a given point in time such that the adeptus mechanicus distances themselves from the technologies; (4) that 'time' could have been up to 10,000 years ago. In other words, story and narrative rather than changing assumptions/impositions. Of course. Forgive my presumption... See the above. It is not the be-all-and-end-all but it is a possible starting point. If it truly is completely, totally and without question to have both 'robot' and 'servitor' without the use of history or culture, then I'll say "Yes." But I really don't think that it is impossible and, indeed, the possibility (possibilities) increases the depth of the universe. Depends on the assumptions that you're going to make about the adeptus mechanicus. From your posts elsewhere it seems reasonable to assume that you do not believe the TechPriests, well all of the TechPriests, are prating morons with little understanding of anything other than the dogma and rote activities that they engage in. That is, they understand the technology. That's fine. I like it. I believe the same thing. Yet the official 'fluff', or at least Image (and this is the same thing that applies to the servitor question), that we merely accept it as truth: the TechPriests are prating morons with no to little understanding of what they're doing, that they merely collect and archive technology (as well as following some basic blue prints in a hit and miss attempt to recreate it)... etc. So what to do? As above, instead of making them 'incompetents' who collected technology you make them people who understood it and used it as a means by which power could be grasped. The 'incompetence', or the religion, is the means by which power is maintained... "Adeptus Me-can-icus" and "Adeptus Me-can't-icus" are held in balance. The initiators of the adeptus mechanicus then, perhpas, become those few individuals that had the means and knowledge to be able to maintain the appropriate technologies... Technicians and scientists rather than librarians. Possibly. Again, how does any group seize power. In this case, perhaps, by the control of a 'thing' that others desire. Power, wealth... etc. But is it mystical "mumbo-jumbo"? The 'fluff' is clear that it serves a purpose, even if it does tend to have tongue-in-cheek references to the "Ritual of the Hammer" and the 'awakening' of a machine spirit by bashing it with a mallet. Just because a 'lower' person doesn't understand the ritual doesn't mean that the ritual is not understood... Rituals, after all, serve several functions. On the one hand they are a somatic/physical codification of prescribed actions for a given effect. On the other hands, they are a means of obfuscating that same knowledge... And, of course, one might consider " tao": they are a means by which a way/path can be placed in front of an individual. Or, if you want to couch it in sci-fi: "You'll learn as I did that it's not knowing the path but walking the path" (paraphrased!). I'm biased here. For me religion is all about control. (Faith is another thing entirely.) Thus the genesis of the 'religion' was an act of control, either of the restriction of knowledge to the adeptus mechanicus or the restriction of knowledge within the organisation. Deification of ideals. <grin> See the above for a potential. Is this not missing the point? Are we talking about 'intelligent' robots here? I would say not, or at least not exclusively so. Servitors, as with the question of AI (PI, MI, etc.), cover the entire spectrum even if they are sometimes represented as drooling, lobotomised people. (E.g. ? Deus Encarmine, or is it Lord of the Night?) Is it more expensive to 'mass produce' a calculator, with the expense balanced out overall, or lobotomise some poor sap of a human and install an abacus where there intestines used to be? To then selectively remove and add in bits of 'memory' - or whatever - to make the servitor-calculator work?
|
|
|
Post by Kage2020 on Mar 14, 2005 23:33:21 GMT -5
Arguably servitors make the Imperium even more advanced than it is already... or perhaps needs to be. Of course, there is also the concept of 'hidden technology': servitors are merely the outwards face of technology... Rather than having disembodied voice 'controlling' a gun, you plug in a servitor so that you can see who is doing the firing. It's ludicrous, but heck... It still doesn't change the complexity of the systems involved. Who would put a 'machine mind' in a servitor? That's just daft. Of course, would you then array thousands upon thousands of servitors as a 'starship computer'...? Maybe a lot. And sounds a bit chicken and egg-like to me, but then again I'm expecting a "STC! hallelujah" cry from this one. Maintenance systems, direct neural interface systems, etc. If all you need is a human to do a job, just use a human. That's even cheaper than a servitor. And no technology involved at all... This is not about worlds vs. the adeptus mechanicus. Your assumptions in this regard alone. Enough on that one. Indeed, they are. So even the 'trucks' used to transport goods are servitors? Do all vehicles have servitor bodies implanted into them? Or brains? That last was the preference of Tobias... Why not just use a cargo lifter? Just because you could automate it doesn't mean that you would want to. Having a mind-numbing job doesn't mean that they're going to automate it. Any archaeologist who has had to hand move tons of earth, despite the presence of mechanical and even robotic means for achieving the same, will be able to tell you this. Of course, just because there are both doesn't mean that you must use an archaeologist. That is just a choice. Now they are... Darn them not needing any technological compone... hang on. DNI, cybernetics, etc., etc. Same thing applies to a fork-lift truck. One must remember that the incorporation of servitors into the 'fluff' is as much a matter of Image as anything else. And darned good eyes as well for those spaceship gun servitors. And even capable of determining whether a ship is an enemy just from looking on the outside of it as well! Well, that'll be part of a problem. That is not remotely how I view servitors, nor how they have been described in the 'fluff'. (Although there have been examples of 'servitors' that are given near-human or human-level consciousness/mental faculties.) Yet biological systems do... The servitor. Or the human, since the way that you're describing them there is little to no difference. And, of course, you could just replace 'servitor' with 'robot' at practically any point and, when not, come up with counter-arguments and examples. Kage
|
|
|
Post by Zholud on Mar 15, 2005 3:55:28 GMT -5
Some additional tidbits I thought out on robots vs servitors. Robots are extremely durable and this durability is a negative factor for Mechanicus when they lend their resources. Once some Imperium body receives good robot, self maintaining one with 1,000-year warranty, it does not need the AM. And the first goal of the AM as any other organisation is to be needed. Thus they supply servitors, which last for a decade or two and wear off. On expensive and cheap parts. We have to set the prices and costs first. Philip often misuses price, or what people are ready to pay for a good and cost, or how much you spend to produce the good. Robots may be cheap to produce but Mechanicus as monopolist may set super high prices on them. Even just to push you to buy servitor instead.
|
|
|
Post by Philip on Mar 15, 2005 10:25:36 GMT -5
Production line. As far as I know, automated systems and production line robots do still exist on forge worlds. These come under ‘auto-systems’, even Space Marines have a form of computer system in their power-armour in the form of the auto-senses. Star ship gunner I think star ships would still have auto-systems, even auto-aim. The servitor plugs into the ‘auto-sense/ aim’ and acts as an operator, the servitor’s form just allows a deeper/ immersive interface with the machine. The servitor could be the human element in the system that has to ‘OK’ an action, the last line of failsafe before terminating a target. Combat robots In the discussion in this thread I assumed the robots we were talking about were the combat type ones that could replace servitors in form and function (the typical humanoid style combat-droid). Apologies for any confusion, I have always seen 40K robots as ‘servitors’ in one form or another, so when I hear the servitor vs robot argument I though people were advocating bringing in ‘computer based’ robots. 40K Robots always had a ‘Bio-Brain’ As a design point of note, no matter which edition of 40K you pick up: robots have never had a machine intelligence, or any form of computer based mind. Even in the very beginning the 40K robots had a cortex with is basically a bio-mass mind, so, technically robots in 40K have always been ‘servitors’ of a sorts. Machine Intelligence It seems that in 40K MI/AI isn’t suited to automating a humanoid type robot. I think the technology to build a fully artificial (non-biological) mind is impossibly complex, and is keep as a secret by the Ad-Mec. Why the current form of servitors? Currently servitors are ‘open-design’ showing the biological elements rather than the previous closed-design where the servitor was covered in a metal skin. However servitors are usually much smaller than the old style robots, and a comparable form to the old style robots would be the dreadnoughts. What’s the difference between a dreadnought and the old style robot? Apparently not much, both have a machine body with a bio-mind (one human one cortex). Looking at the reasons for taking out the old style robots in preference to the dreadnought, is the dreadnought is just ‘better’. Why have an artificial bio-mind that’s a bit stupid why you can put a hard-core veteran in the machine body instead. Also from the original fluff, if I remember, the Ad-Mec would scavenge parts form the old style robots to keep the dreadnoughts running. Conclusion In conclusion there are no computer based humanoid robots in 40K, 40K has only ever used biological minds. I even suspect that auto systems have biological components. The biological base is what Gav was referring to why he talked about ‘machine spirits’ – anything biological and is alive has one of these machine spirits. Traditional automated production lines have robots probably very similar to the ones we have in factories now-a-days, that are used to produce components like laz-gun and what-nots. The distinction between these production-line robots (in 40K) and the classic sci-fi humanoid robot (not in 40K) is by calling the production line robots ‘auto-systems’ Philip
Edit:added Originally posted by Philip: As for using bio-matter, the Ad-Mec wouldn’t care if it fell into the wrong hands, what is a techno-thief going to do with it? Reverse engineer it?Darn them not needing any technological compone... hang on. DNI, cybernetics, etc., etc. Hence this; (If a servitor is captured do those that have captured them gain an insight to the Ad-Mec’s technology? No! They get a squiggly bio-mass which they can’t access, the only interesting bit would be the interfaces – but that probably explains the dodgy back-street bionic upgrades I’ve heard tell about).
|
|
|
Post by Sojourner on Mar 15, 2005 12:32:26 GMT -5
With regards to servitors and data processing:
A single human brain is many orders of magnitude more powerful in processing than the best CPUs today. not so long ago it was announced that CPUs had reached the approximate computing power of a bumblebee.
If we suggest that human neurones operate on yes/no mechanics like microtransistors do, it could be possible to reprogram a human brain to calculate using logic gates, just like a CPU.
You'd only need one...
|
|
|
Post by Zholud on Mar 15, 2005 13:05:07 GMT -5
What’s the difference between a dreadnought and the old style robot? Apparently not much, both have a machine body with a bio-mind (one human one cortex). The question is was is that bio-mind. What we have is: “ What makes a Robot different from an unoccupied Dreadnought suit is its cortex. This is an artificial brain of sorts, which is constructed from artificial proteins and enzymes. This cortex is imprinted with simple maintenance and movement routines - a rudimentary 'mind'. These enable the Robot to obey simple instructions ("Open the Weapon Bay Door, Please ... Move Ahead to the Holding Area" etc) when away from the battlefield. These 'firmware' routines (so called because they are 'wired in' software) are often patterned after living creatures, and a Robot may develop a dog-like devotion to its technician-master” – Compendium,. P. 97. Proteins and enzyme does not necessary mean a brain similar to that of human/dog/etc… currently there is R&D on bio-storage systems, which can replace mechanical storage… 1 cubic cm that holds tetrabites of data… but it is quite dissimilar from brain as we know it (and we don’t know it much)
|
|
|
Post by Philip on Mar 15, 2005 14:09:20 GMT -5
If we suggest that human neurones operate on yes/no mechanics like microtransistors do, it could be possible to reprogram a human brain to calculate using logic gates, just like a CPU. Unfortunately neurons don’t act like that: each one has multiple inputs, and all output signals depend upon the number and strength of inputs (even the outputs are variable in strength and can split). Oh, and inputs/ outputs are reversible too! A single neuron output is ‘analogue’ and is infinitely variable. A whole CPU isn’t as complex as even a single neuron! I forget how many logic gates you need to duplicate a neuron and its analogue nature, but it a lot because modern CPU’s are basically binary (yes/ no – open closed even the Or, AND etc gates follow this principle) whereas a single neuron is analogue. Every neuron has between 1,000 and 10,000 synapses (reversible analogue input/ output connections) so that’s 1,000-10,000 bi-directional analogue gates, a single analogue gate would require thousands of normal silicon based logic gates to replicate (to cover anything like the full range of signal – but obviously being binary there will be still be gaps basically its going to be a D/A-DSP-A/D per gate!). That’s like having a (very basic) soundcard per gate, having 1-10,000 DSP in one area is a lot of chips, modern DSP is around 10 million transistors, but even a very basic one is 50,000. So a single neuron would be 1,000 (synapses) x 50,000 (DSP transistors) = 50,000,000 transistors total – 50 million Or 10,000 x 50,000 = 500,000,000 – 500 million A modern CPU has around 60-120 million transistors (2-4 transistors per gate) so each neuron would be made up of one half to four or five P4s. Oh and some neurons have 20,000-30,000 synapses. Double Oh, and there are around 100 billion neurons in our nervous system.
Edit:Added The question is was is that bio-mind. What we have is: “ What makes a Robot different from an unoccupied Dreadnought suit is its cortex. This is an artificial brain of sorts, which is constructed from artificial proteins and enzymes. This cortex is imprinted with simple maintenance and movement routines - a rudimentary 'mind'. These enable the Robot to obey simple instructions ("Open the Weapon Bay Door, Please ... Move Ahead to the Holding Area" etc) when away from the battlefield. These 'firmware' routines (so called because they are 'wired in' software) are often patterned after living creatures, and a Robot may develop a dog-like devotion to its technician-master” – Compendium,. P. 97. Proteins and enzyme does not necessary mean a brain similar to that of human/dog/etc… Err, in this instance it seems it does, in the Compendium quote you gave it says “These 'firmware' routines <snip> are often patterned after living creatures” so it would seem that these cortex minds are based of living creatures, in the next line it alludes to the type of creature “a Robot may develop a dog-like devotion to its technician-master”, the cortex minds are based on a dog's mind or at the very least it develops a dog like set of social skills. May be, the cortex mind is at the same level of training and intelligence as a ‘police dog’. Whereas a dreadnought thinks like a human.
|
|