Wikipedia:Reference desk/Archives/Science/2016 October 2

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

October 2

Mind uploading

When do people predict that mind uploading will become possible? — Preceding unsigned comment added by Uncle dan is home (talk • contribs) 00:08, 2 October 2016 (UTC)

We have an article on mind uploading. Long story short - this technology will require so many things that haven't been invented or discovered yet, it's impossible to give a meaningful prediction. Someguy1221 (talk) 00:22, 2 October 2016 (UTC)

It already got invented. See Pensieve. Nyttend (talk) 03:46, 2 October 2016 (UTC)

Probably never, per This article which has some good reminders about the danger of badly-conceived analogies for the human mind. --Jayron32 23:28, 2 October 2016 (UTC)

Jayron, that was a *very* good link. Thanks! Robinh (talk) 07:11, 4 October 2016 (UTC)

Excellent link, Jayron. The problem with asking about uploading your mind, is that it is a category mistake. Examples of category mistakes would be questions like, "Can I change my biological father?" or, "Which weighs more, the color blue or the color yellow?"

Your mind is not a physical substance, it is not a series of ones and zeros; it is a faculty which allows you as an organism to apprehend the forms of things in your environment without physically ingesting them. (The converse of this would be digestion, where we physically ingest the substance of food, but do not the acquire its form.) Just as a computer simulation of a Hurricane is not a hurricane, a computer simulation of your mind is not your mind. The "mind" is simply not "uploadable".

What might potentially be possible would be a Ship of Theseus situation, where nanobots might repair or replace damage tissue in your body, including your brain. You'd simply slowly become a cyborg or a pod person. You would not experience death by aging in that scenario, and might not experience any noticeable changes. But even then you would almost certainly slowly forget your past in the way 12 year-olds can't remember what they did when they were two. μηδείς (talk) 18:20, 4 October 2016 (UTC)

Excellent link? It reads like a gigantic strawman fallacy to me. The author treats the neuroscientist's analogy of the human brain to a computer too literally. Someguy1221 (talk) 21:29, 4 October 2016 (UTC)

Probably within the next few centuries, despite skeptics saying that the brain is not a computer. Whether the brain is a computer is not relevant, what matters is that the laws of physics are computable. The Church–Turing–Deutsch principle says: "a universal computing device can simulate every physical process". This means that brain (whatever it is) can be implemented by a computer regardless of whether we consider the brain is itself to be a computer or not.

Another way to see the irrelevance of what the brain is, is by considering thought experiments where you replace the neurons by electronic devices one by one such that functionally it operates in the same way. What matters is that there exists a one to one mapping of the physical state of the system described in terms of its new components to the physical state of the old system that is preserved under time evolution under the influence of external influences. This process would then transform a biological brain into an electronic one. The person would behave in exactly the same way as before in all possible situations. The question is then if the person is conscious and has the same consciousness as the biological person. The person would certainly say so if we asked (if not then the replacement of neurons by electronic devices did not preserve function after all, but known physics implies that it is possible to do so). So, a counterargument could be that the robot is a zombie. But then how do I know that I'm not the only real conscious person in the universe and everyone else including μηδείς and Jayron32 are zombies?— Preceding unsigned comment added by Count Iblis (talk • contribs)

You don't. --Jayron32 19:27, 4 October 2016 (UTC)

The problem with Count Iblis' speculationn is (1) he hasn't studied the philosophy of mind, or he'd know that his "replace the neurons by chips, one by one" scenario is exactly what the Ship of Theseus allusion was in my post, and that (2) neurons aren't wires and aren't digital and don't work digitally. If anything, it's much more likely we'll have Mentats and Guild Navigators in the future than Oracs or Marvin the Paranoid Androids. Dr. Piter will see you now. μηδείς (talk) 21:16, 4 October 2016 (UTC)

The greater issue is that in order to completely replicate the brain in the manner in which Iblis proposes, you'd need a machine that replicates the complete information of every atom in the brain: it's location, bonding, motion, and relationship to other atoms. The problem with that level of knowledge is both 1) some of that information is inherently unknowable and 2) even if it were knowable, it would require a machine with exponentially more atoms than the brain itself: each individual atom has a bewildering amount of information to fully describe its state; and every one of those pieces of information has to be stored somewhere. The fundamental problem of the clockwork universe is that in order to build a computer to completely describe, and thus predict, the behavior of such a universe would require a computer bigger than the universe itself. It's an many-body problem where n=all the atoms in the universe. --Jayron32 21:34, 4 October 2016 (UTC)

I think you may have meant to say, where N > than all the atoms in the universe. μηδείς (talk) 22:21, 4 October 2016 (UTC)

Right, there are multiple problem with Iblis proposal. A) Is it possible to measure the state of a human brain with sufficient accuracy to model its behavior? B) Is it possible to model the human brain's behavior with something other than a human brain, that can actually be built by humans with current or future technology? C) How do you solve the hard problem of consciousness? Even if we assume that there is a way to do what Iblis wants, you always land at problem C no matter what, at least if your plan is to extent life, rather than just settle for a faithful copy. Someguy1221 (talk) 21:40, 4 October 2016 (UTC)

Thing is that right now I'm not a good copy of the person I was yesterday, a lot of things have changed in my brain. So, an approximate copy will do, we don't need to get every last detail exactly right. Count Iblis (talk) 21:15, 6 October 2016 (UTC)

I've doubtless proselytized this here too much already, but my personal opinion is that qualia is a paranormal phenomenon, and like all paranormal phenomena it is a consequence of precognition - in particular, the ability of precognition to induce causality violations. I think that macroscopic precognition is rare because it is very destructive, I mean 'shalt not suffer a witch to live' destructive, and it is also an inherently miscellaneous sort of phenomenon. But when it occurs in a short term, something like pulling a number out of a machine and putting it in to be scanned and sent back to the past a second later, the effect is a free will that is not random and not determined. I think if you replaced the memory of a person with a prosthetic hippocampus in which these phenomena did not occur, the (former) person would seem virtually the same but would not 'really' feel anything. Wnt (talk) 01:15, 7 October 2016 (UTC)

Lithium-ion batteries

When Lithium-ion batteries fail, what is actually happening? Is it just a runaway discharge that overheats the battery and sets it's casing and then it's surroundings on fire? Or is there a true explosion occurring? 49.197.15.137 (talk) 07:49, 2 October 2016 (UTC)

An internal short circuit, or unusually high temperature operation, generates enough heat to cause the battery to burst open and ignite typically the organic solvent that was used to dissolve the lithium. You can read more at Lithium-ion battery, thermal runaway and maybe this article. Someguy1221 (talk) 07:58, 2 October 2016 (UTC)

Thanks Someguy, I had tried both those articles, but it wasn't clear to me why people are talking about "explosion". An organic solvent burning in air isn't really an explosion is it? (OP)49.197.15.137 (talk) 08:13, 2 October 2016 (UTC)

That's not an explosion, no, but the pressure inside can built up so high during a short that when the casing finally gives there really is an explosion. There are videos on youtube of people forcing lithium-ion batteries to explode so you can see what happens, and that's with a tiny one. Someguy1221 (talk) 08:30, 2 October 2016 (UTC)

Anything combustible can generate an explosion if it combusts rapidly enough. See dust explosion. A Web search will give you myriad videos of people exploding coffee creamer, flour, etc. As the previous poster noted, if the material is contained in a confined space the explosion can be even more dramatic. --47.138.165.200 (talk) 10:59, 2 October 2016 (UTC)

• Deflagration is usually closer to the mark and it's very much not a detonation. Still serious though.

I think the 2007 Veolia waste storage plant fire [1] [2] [3] is still the largest single incident. Andy Dingley (talk) 11:56, 2 October 2016 (UTC)

NTSB's report, AIR1401, Auxiliary Power Unit Battery Fire Japan Airlines Boeing 787-8, JA829J, found that "an uncontrollable increase in temperature and pressure" was caused by internal short-circuits within the individual cells of a lithium battery.

Uncontrollable increase in pressure is sometimes termed an "explosion," although there is an important distinction between a detonation and a deflagration - detonations are usually supersonic pressure waves.

The NTSB found that internal short circuits can occur in lithium-ion batteries as a result of foreign object debris introduced during cell welding (part of the manufacturing process); short circuits could also be caused by perturbations in the cell winding process. A single cell short-circuit causes thermal runaway, which causes other cells to overheat, creating more internal short-circuits, ... and the resulting rapid increase in thermal energy presents a risk of "fire" and "explosion," per the NTSB's report.

There is more chemical energy in your battery than you might want to think about: the off-brand replacement for an ordinary laptop battery has about half as much chemical energy as the standard-issue hand grenade. Fortunately, we do not commonly see half-a-hand-grenade on an airplane, but we often see at least 2.5 laptops.

Nimur (talk) 16:33, 2 October 2016 (UTC)

Thank you all for your replies, I can understand the distinctions clearly now (OP)49.197.92.239 (talk) 20:26, 2 October 2016 (UTC)

Feynman Lectures. Lecture 27 [4]

Feynman says that for Fig. 27–2 we can use a spherical surface instead of a fourth-degree curve. Explain please, why did he decide that a spherical surface provide focusing in one point O'? I tested it in AutoCAD and it seems (for small angles) lines really meet on the axis. But how to prove that? https://s.sender.mobi/u/image/2016/10/2/NCKjgD5Jh/-.JPG https://s.sender.mobi/u/application/2016/10/2/9t8RE-RBb/autocad_dr.dwg Username160611000000 (talk) 16:21, 2 October 2016 (UTC)

He said it right there in the lecture: "the student may entertain himself by trying to calculate it by analytic geometry..." Are you not entertained by such calculations? If you are, have at them; and if not, then accept the argument from authority!

Alternately, here's an even better book: Meyer-Arendt's Introduction to Classical and Modern Optics. Chapter 1-5 run the formal analytic geometry calculations for various mirror surfaces; and the subsequent chapter deals directly with optical aberration, including aberrations of spherical and aspherical elements. I'd lob some equations at you, but they take up the better part of a five-hundred-page book. The authors recommend that you read this book over the span of a full year of formal study, after you have completed two to four years of preparatory formal study of physics. Optics is a difficult topic.

I found used copies for under $30 on Amazon.

Nimur (talk) 16:41, 2 October 2016 (UTC)

Are you not entertained by such calculations? . I know how to do that , but calculations is very long. According to image png dwg I write down equations:

${\sqrt {(Y+y)^{2}+x^{2}}}+n{\sqrt {(Y'-y)^{2}+x^{2}}}=Y+nY'=const$

From this equation I can derive $y=f(x)$ by squaring two times . That gives too many terms, so I left that, understanding that I can finish it if necessary.

But Feynman does not connect the solution $y=f(x)$ with the approximation of the spherical surface. Yes, he makes the assumption and than derive the equation Eq. (27.3). But first he had to guess the correct assumption, i.g. why he did not take parabola instead sphere? So he did some check for sphere for small angles before use it. I want to know, what is this check. Username160611000000 (talk) 18:25, 2 October 2016 (UTC)

We've been over this many times before: these lectures are not a very thorough or well-laid-out introduction to these topics.

If you want presentations that don't skip steps, and don't use sloppy math, and that don't assume you've seen the material before, you need a better book.

How can I make this any more clear? You're looking at material that largely is not even written by Feynman, but was republished by Caltech after a lengthy and unpleasant legal battle over international copyright terms, intending to reassert the university's ownership of copyright for work created by Feynman and other deceased professors. These lectures were only published essentially because an international publisher issued an ultimatum intending to republish those copyrighted works without permission. The lectures now exist for free online in order to settle a legal dispute, and to assert ownership of copyright, but what you're reading is not a great resource for students of physics. In fact, the authors even say so in their preface. This edition exists because "the complex legal and contractual details were worked out by Caltech’s Intellectual Property Counsel," not because it's a great textbook for physics students.

If you really care to learn this material, get a better book.

I say this as a Wikipedian volunteer who has no stake whatsoever in how well you succeed, nor in how much money you spend on books; but take some advice from a guy who has read - and compared - so many physics text books that if you only name a topic, I can walk to the shelf here in my room and pick out the best from multiple books to provide you with many options to help you learn better.

If you really value Feynman's work, stop reading those lectures and read Judging Books By Their Covers, a story that also appears in his memoirs. Think about it. Are you reading the Feynman Lectures just because they have a prestigious name slapped on their cover, or do you really really believe they're a good resource? How can you make that judgement, unless you've tried some alternatives?

Nimur (talk) 19:04, 2 October 2016 (UTC)

I saw alternatives Landau's course , Physics for the Inquiring Mind and many traditional university textbooks. Landau's course is very hard, E.Rogers' course is very simple and incomplete.

Again, I like these lectures and I shall continue read them. Many physicists recommend these lectures. And besides these physicists give not very high rating to wikipedia's articles on physics.

Username160611000000 (talk) 19:36, 2 October 2016 (UTC)

Landau and Lifshitz is a great series, but based on the questions you regularly ask, I believe you will find that you have not yet studied enough math to make sense of their presentation. I really think that at this time, you would benefit more by reading a book appropriate to your level, and build up over years toward more advanced material. There is no shame in admitting that physics is hard and that very smart full-time students require many years to appreciate its subtleties. Nimur (talk) 19:43, 2 October 2016 (UTC)

I have not many years. Does this concrete question about sphere need many years to understand it? I don't think so. So please answer if it's not inconvenient to you. Username160611000000 (talk) 19:54, 2 October 2016 (UTC)

Where do we start? The intent of emphasizing paraxial rays is so we can use the small-angle approximation; the intent of assuming monochromatic light is so we can simplify the Abbe number towards the ideal material; we assume a perfect application of Snell's law at the refractive surface; now you just have to compute Snell's law for small angles. It does not get more simple than this. You're already looking at the equation for it (27.2 and 27.3 in the lecture note you linked) and you're already reading the derivation (the paragraph immediately preceding that pair of equations). I can quote pages from yet another book; I can link you to the article we have on the lensmaker's equation; but if you aren't seeing how these equations come directly from Snell's law and basic geometry, ... my recommendation is exactly the same: perhaps you need to spend more preparatory time studying analytic geometry. You simply can't do advanced physics if you aren't exceptionally comfortable manipulating equations in this way. To directly answer your question - does this question require "many years" of study to answer it? On average - yes, absolutely, it does take many years for even a very smart student to become comfortable with this type of math. I have cited many books, websites, and curriculum guides from various sources that confirm this: they all recommend two to four years of formal preparatory physics and math before you dive into this level of material, whether you consume it from Feynman Lectures or from any other source.

A good introductory book will have you work several easy problems, solving Snell's law on simple boundaries, long before asking you to solve it on curved surfaces. As you can see, when you read Feynman Lectures, it is assumed that you are already an expert at solving Snell's law problems and that you have already memorized all the trigonometric identities, all the formulae for common conic sections and other analytical curves; and that you can recall these basic facts from memory, on command, even when prompted with subtle hints or unusual presentations. How well do you have those materials memorized? Perhaps it is worth spending eight or ten months reading Paul Tipler's book first, and working several hundred or a few thousand of his homework problems? That is what physics students do to prepare for more advanced material. If you try to skip this step, you aren't saving any time: you're only losing yourself in details you don't understand.

I cannot distill a complicated physics equation that builds on years of physics and math into a simple one-paragraph explanation - nor can any book. You have to build up to these materials over a long time.

If solving literally thousands of preparatory mathematical physics homework problems doesn't seem possible - perhaps even easy - for you, you need to profoundly reconsider whether you will ever succeed as a student of advanced physics. Could a student of literature survive if they were intimidated by a thousand pages of reading material? But we don't begin by handing Dostoyevsky to kindergarteners.

Nimur (talk) 21:20, 2 October 2016 (UTC)

How well do you have those materials memorized? I have already studied trigonometry and analytical geometry few years in school and in university. I remember material. Besides, Feynman gives such topics as vectors, complex numbers etc. So I think Lectures could be understood with very small additional knowledge. Some physicists recommend Lecturers as transitional course before more advanced courses (like Landau's). Of course if I shall study Landau first, I easily understand Lecturers. Anyway, I am grateful to you for your help. Username160611000000 (talk) 06:31, 3 October 2016 (UTC)

According image above https://s.sender.mobi/u/image/2016/10/2/NCKjgD5Jh/-.JPG we can write next.

If angles are small then QQ₂ = 0. So we must prove that QO'=Q₂O'.

$\angle QO'P=90-(90-\alpha +{\tfrac {\alpha +\beta }{n}})=\alpha -{\tfrac {\alpha +\beta }{n}}$;

$QO'={\tfrac {PQ}{tg(\alpha -{\tfrac {\alpha +\beta }{n}})}}={\tfrac {Rsin(\alpha )}{tg(\alpha -{\tfrac {\alpha +\beta }{n}})}}={\tfrac {R\alpha }{\alpha -{\tfrac {\alpha +\beta }{n}}}}$;

$Q_{2}O'={\tfrac {P_{2}Q_{2}}{tg(\alpha _{2}-{\tfrac {\alpha _{2}+\beta _{2}}{n}})}}={\tfrac {Rsin(\alpha _{2})}{tg(\alpha _{2}-{\tfrac {\alpha _{2}+\beta _{2}}{n}})}}={\tfrac {R\alpha _{2}}{\alpha _{2}-{\tfrac {\alpha _{2}+\beta _{2}}{n}}}}$, where $\alpha _{2}=\angle Q_{2}CP_{2},\beta _{2}=\angle Q_{2}OP_{2}$.

For small angles we can write :

$\alpha _{2}={\tfrac {P_{2}Q_{2}}{PQ}}\alpha ,\beta _{2}={\tfrac {P_{2}Q_{2}}{PQ}}\beta$

$\implies Q_{2}O'={\tfrac {R\alpha _{2}}{\alpha _{2}-{\tfrac {\alpha _{2}+\beta _{2}}{n}}}}={\tfrac {R{\tfrac {P_{2}Q_{2}}{PQ}}\alpha }{{\tfrac {P_{2}Q_{2}}{PQ}}(\alpha -{\tfrac {\alpha +\beta }{n}})}}={\tfrac {R\alpha }{\alpha -{\tfrac {\alpha +\beta }{n}}}}=QO'$.

Is it correct?

Username160611000000 (talk) 07:48, 3 October 2016 (UTC)

I guess Feynman chose the sphere because a spherical lens is easy to grind (nothing to do with theoretical physics, just the way lens grinding works). For paraxial rays, the sphere gives the same results as the paraboloid or the exact fourth-order surface, as can be seen by Taylor-expanding the respective equations in terms of distance y from the axis (the series for the paraboloid and the fourth-order surface are of course trivial). All these Taylor series start with a term a y² with higher orders (O(y⁴)) negligible for small y. If you match the surfaces such that they all have the same coefficient a, then they will have the same optical properties in the paraxial approximation, and the sphere will do as nicely as the exact surface. When that approximation is not valid, you'll get spherical and other aberrations, and that's when optics really becomes difficult. --Wrongfilter (talk) 20:08, 2 October 2016 (UTC)

I have expanded the equation:

$y(8Y^{2}Y'n-8Y^{2}Y'n^{2}-8YY'^{2}n^{3}+8YY'^{2}n^{2})+y^{2}(4Y^{2}n^{2}-4Y^{2}+4YY'n^{3}-8YY'n^{2}+4YY'n-4Y'^{2}n^{4}+4Y'^{2}n^{2})-y^{3}(4Y-4Yn^{2}+4Y'n^{2}-4Y'n^{4})-y^{4}(n^{4}-2n^{2}+1)+x^{2}(4Y^{2}n^{2}+4YY'n^{3}+4YY'n+4Y'^{2}n^{2})-x^{2}y(4Y-4Yn^{2}+4Y'n^{2}-4Y'n^{4})-x^{2}y^{2}(2n^{4}-4n^{2}+2)-x^{4}(n^{4}-2n^{2}+1)=0;$

$y\cdot A+y^{2}\cdot B-y^{3}\cdot C-y^{4}\cdot D+x^{2}\cdot E-x^{2}y\cdot C-x^{2}y^{2}\cdot 2D-x^{4}\cdot D=0;$

Do you mean we can use Taylor series for $f(x,y)={\sqrt {(Y+y)^{2}+x^{2}}}+n{\sqrt {(Y'-y)^{2}+x^{2}}}-(Y+nY')$, then solve $T_{2}(x,y):=f(0,0)+xf_{x}+yf_{y}+(x^{2}/2)f_{xx}+xyf_{xy}+(y^{2}/2)f_{yy}=0$ for $y$? And then use Taylor series for $f(x)$? It seems $T(x)$ really gives some number multiplied by $x^{2}$.

Using Mathcad I've found that curve as intersection of two surfaces PNG xmcd. So required curve is convex and can be substitute by circle of same curvature. — Preceding unsigned comment added by Username160611000000 (talk • contribs) 21:19, 3 October 2016 (UTC)

Impact force distribution

What can be inferred from a photograph of a broken windshield glass regarding the impact force (and velocity) distribution in an accident caused by encountering a pedestrian? Can be the state of injury of the pedestrian be assessed from an analysis of the photograph of the broken glass?--82.137.14.14 (talk) 18:55, 2 October 2016 (UTC)

Not easily. However if the airbag went off then the vehicle speed and velocity trace is recorded, which would simplify he task somewhat. Of course, it might be simpler to get the medical records of the patient rather than reverse engineering the accident.Greglocock (talk) 10:45, 3 October 2016 (UTC)

What makes B complex vitamins smell so awful.

I take several vitamins, but have trouble swallowing the B-complex, as the smell of it makes me gag. I was wondering what is the chemical culprit in this malodorous mixture? — Preceding unsigned comment added by 50.27.61.236 (talk) 19:47, 2 October 2016 (UTC)

I don't know the specific one (or even what specific chemicals are in your particular tablet). However, chemically speaking, among the B vitamins are several thioethers and many amines, two classes of chemicals that can spell bad. When I open a bottle of pure thiamine, it does smell sulfur-like or amine-like to me (can't remember which, and that's the only one I have in pure form as a datapoint). DMacks (talk) 21:26, 2 October 2016 (UTC)

I didn't know that chemicals were bad spellers! ;-) DrChrissy ^(talk) 21:46, 2 October 2016 (UTC)

Apparently my stuffy nose has managed to affect my typing-pronunciation too:) DMacks (talk) 21:49, 2 October 2016 (UTC)

I find this type of question really interesting. I’m a supertaster that can't eat cooked cabbage (by can't eat, mean that even if I force myself to swallow, my oesophagus returns it to it place of origin and quite quickly). Yet, I quite like the smell of B complex (it is the B6 that I can smell. B 1 mentioned above has only a slight detectable smell to me). As supertasters are a result of genetics maybe the OP and others that find the smell of B com unbearable have a genetic trait too? Doctors usually don't prescribe B 50's but sugar coated formulations that don't smell nor have a taste but the B range and dosage is a lot smaller that B 50's which are not sugar coated. In view of this, I know there is little point in suggesting you pinch your nose, before trying to swallow, because I doubt that will help in any way at all. Maybe brushing them with honey first might help but have never tried coating cabbage with honey -so I'm just guessing.--Aspro (talk) 00:08, 3 October 2016 (UTC)

There's always the old-fashioned approach of holding your nose. And by the way, cooked cabbage is often wretched, and not worth holding your nose to try to ingest. But vitamins are important. ←Baseball Bugs ^{What's up, Doc?} carrots→ 00:13, 3 October 2016 (UTC)

• The question seems a bit odd, since I have either been given a shot, or taken a B vitamin supplement, neither of which had a smell. But that is merely anecdotal, given I had major sinal surgery 30 years ago. μηδείς (talk) 02:19, 3 October 2016 (UTC)

Are you sure that wasn't vitamin D ? --Aspro (talk) 03:18, 3 October 2016 (UTC)

I have actually had both, but a B complex shot most often. Unfortunately I only have 1/3 of a colon. μηδείς (talk) 09:54, 3 October 2016 (UTC)

Funny you should ask, since I am on a weekly dose of 50,000 units of vitamni D! One of the other drugs I take prevents sun exposure. I used to drink 5 gallons of milk a week, but as a type II diabetic, that's problematic. μηδείς (talk) 09:49, 3 October 2016 (UTC)

This [5] claims thiamine tends to be the cause, as biotin is present in smaller amounts. Nil Einne (talk) 07:22, 3 October 2016 (UTC)

Wow! I always thought B vitamins smelled of brewers yeast as that is what they were made from. --TammyMoet (talk) 14:53, 3 October 2016 (UTC)

Here in the UK we have a spread called Marmite. One of the selling points is it is a rich source of vitamin B complex. Is it just a coincidence that it's advertising slogan is "Love it or Hate it". DrChrissy ^(talk) 19:27, 3 October 2016 (UTC)

Marmite is also available in Texas from here so if the OP is glutton for punishment, he could try that to. Suggest spreading it very thinly on piazza to start with or add to stews etc. Bound to be a lot cheaper than vit pills, as little Marmite goes a long way, since it not diluted by the stabilizers and bulking agents that constitute most of the physical size of the pill formulations. However, he must run this past a qualified fizcian first as Marmite doesn't contain the all important B6 (same for most yeast tablets). If one needs B complex, one nearly always needs B6 as it is a precursor to aid the metabolism of the other B's. Yet, if he hates it too, it is probably not down to the B6. --Aspro (talk) 20:53, 3 October 2016 (UTC)

I thought it was illegal in the States, but it turns out that was just a rumour about Vegemite. Both of them are illegal in Denmark, though. Tevildo (talk) 21:58, 3 October 2016 (UTC)

Our article Marmite says explicitly, with refs, that Marmite is no banned in Denmark. DuncanHill (talk) 20:16, 5 October 2016 (UTC)

To me, B-complex tablets have a rather pleasant smell. --Guy Macon (talk) 19:40, 3 October 2016 (UTC)