Wikipedia:Reference desk/Archives/Science/2010 May 17

Science desk
< May 16	<< Apr | May | Jun >>	May 18 >

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.

May 17

Tomato and Corn

Why is a tomato a fruit? And is corn a vegetable or grain? wiooiw ^(talk) 01:55, 17 May 2010 (UTC)

See Tomato#Fruit or vegetable?. Dismas|^(talk) 02:04, 17 May 2010 (UTC)

(edit conflict) The term "vegetable" has no definition in horticulture; it is applied somewhat arbitrarily. Fruits, by definition, are the mature, seed-bearing ovaries of angiosperms (flowering plants). Grains (cereals) are those grasses of the family Poaceae. Here, I am assuming that by corn you are referring to maize; be careful with your use of the word, since "corn" can be used to refer to the principal cereal crop of a region. Intelligentsium 02:07, 17 May 2010 (UTC)

Most of the world's growers and consumers of corn mean corn, the tall stalk with big ears, when they say corn. Not "maize." Edison (talk) 03:48, 17 May 2010 (UTC)

The big, international seed companies use the term "maize" to avoid confusion among the English-speaking portion of their audience, which mean various things when they say "corn". Non-English speakers have other terms. In Spanish, choclo or cereal are used for the generic term "corn", whereas maíz is maize, trigo is wheat, cebado is barley, centeno is rye, etc. ←Baseball Bugs ^{What's up, Doc?} carrots→ 12:00, 17 May 2010 (UTC)

This is my favourite article on the tomato subject: Nix v. Hedden.Aaadddaaammm (talk) 15:01, 17 May 2010 (UTC)

Yes, referring to maize, thankyou. wiooiw ^(talk) 02:41, 17 May 2010 (UTC)

The more generic query

On the subject of fruits vs vegetables, back in high school (several decades ago) we were told the question was nonsense because biologically one is actually a subset of the other. You might as well ask, when seeing a four-wheeled vehicle on the street, "Is it a Ford, or is it a car?"

So, quite some years later, the question: is there any truth in that analogy, or is it just another "Go-away-kid-you-bother-me" response? DaHorsesMouth (talk) 02:47, 17 May 2010 (UTC)

Interesting perspective. wiooiw ^(talk) 02:52, 17 May 2010 (UTC)

"Vegetable" often refers to any kind of plant matter (consider the classic animal, vegetable, mineral trichotomy), so in that sense, it's totally correct. But culinarily, "vegetable" has a variable and vague, yet more specific, meaning. Paul (Stansifer) 03:06, 17 May 2010 (UTC)

If you read the intro and the "Fruits vs. Vegetables" section of our article "Vegetable", you'll see where the problem comes from. The conclusion is that if you are speaking scientifically, then all fruits are also vegetables - but not all vegetables are fruit. If you are speaking in a culinary sense - then fruits and vegetables are entirely separate categories - with no overlaps. However, some thing that are properly fruits are sometime mis-categorized by the general public, leading to considerable confusion (eg Tomatoes are fruit according to both scientific and formal culinary definitions - but the average person in the street considers them to be vegetables. They also consider Rhubarb to be 'fruit' when it is of course a stem - and therefore a vegetable. So now you have science, the culinary arts and common usage all coming up with wildly different definitions. Bottom line is: You decide! SteveBaker (talk) 03:27, 17 May 2010 (UTC)

According to Jewish law, a benediction praising the Almighty must be recited before eating any item of food or drink. The text of the benediction depends on the type of food eaten. For fruit it is בורא פרי העץ (who creates the produce of the tree); for vegetables it is בורא פרי האדמה (who creates the produce of the ground). Should one err and recite the benediction for vegetables over fruit, one has fulfilled one's obligation, the reason being that fruit is also produce of the ground. However should one err and recite the benediction for fruit over vegetables, one does not fulfil one's obligation since vegetables are not produce of the tree. Simonschaim (talk) 07:36, 17 May 2010 (UTC)

So why would anyone be stupid enough to risk the wrath of the almighty and eternal damnation? If I believed in all that nonsense, I'd be saying the vegetable benediction no matter what! No sense in risking a mistake! (Maybe you get extra points for using the more precise version?) SteveBaker (talk) 15:56, 17 May 2010 (UTC)

Can anyone provide one or more scriptural references about requiring those benedictions? -- Wavelength (talk) 16:11, 17 May 2010 (UTC)

Benedictions before eating food are Rabbinical injunctions. Scriptural support can be found at Psalms 24:1 and Psalms 115:16. Further details can be found in the Babylonian Talmud, tractate Berachot 35a-b. (An English translation of the Babylonian Talmud can be found on the internet.)Simonschaim (talk) 18:48, 17 May 2010 (UTC)

I found http://scripturetext.com/psalms/24-1.htm and http://scripturetext.com/psalms/115-16.htm, which can be harmonized by the earth having been entrusted to the human race, but I did not find any requirement of those benedictions. -- Wavelength (talk) 19:34, 17 May 2010 (UTC)

Wavelength: Please read the reference in the Babylonian Talmud, which I suggested. Yon can find it via Google - Soncino Talmud.Simonschaim (talk) 19:44, 17 May 2010 (UTC)

Caesar's last breath

I read in a Paul Zindel novel that every person in the world contains at least two atoms of Julius Caesar's last breath, or something along those lines. I find it hard to believe (and obviously impossible to prove), but do the nature of atoms allow for something like that to happen? Thanks!  ?EVAUNIT^{神になった人間} 02:44, 17 May 2010 (UTC)

If sufficient time for mixing has taken place, such that Caesar's breath were evenly mixed throughout the atmosphere - then I think this is believable. However, the issue of whether sufficient time has elapsed is a tricky one.

Let me crunch the numbers for you: According to Atmosphere of Earth, the atmosphere weighs in at about 5x10¹⁸ kg. According to [1], 65% of the mass of the human body is oxygen and only about 3% nitrogen, so let's ignore the nitrogen for the sake of simplicity. Therefore, an average 75kg human contains about 50kg of oxygen - which is therefore about one part in 10¹⁷ of the atmosphere. So if the mixing is even, we each have about 1/10¹⁷th of Caesar's last breath. A typical human lung capacity worth of air weighs about 5g - but only 20% of that is oxygen - so let's go with one gram. (oxygen says that a human consumes around 2g of oxygen per minute...so we're obviously in the right ballpark). That's 1/16th of a mole - or 1/16th of Avagadro's number worth of atoms (Avagadro's number is 6x10²³ and the atomic weight of oxygen is 16)...so his last breath comprised roughly 4x10²² atoms of oxygen . So 1/10¹⁷th of the that last breath would be around half a million oxygen atoms.

Conclusion: If atmospheric mixing is perfect and the majority of the oxygen in our bodies ultimately comes from the atmosphere (possibly via plants and animals that we eat - possibly by breathing) - then there are half a million oxygen atoms of Caesar's last breath in each of our bodies...plus some carbon, nitrogen and hydrogen atoms. I don't know where the number '2' comes from...but it all comes down to how good that mixing truly is...and I have no clue how to even guesstimate that! However, that's a pretty good sized ratio - there is a lot of room for uneven mixing. So I'd say that this claim is almost certainly true - perhaps even a fairly large underestimate.

SteveBaker (talk) 03:09, 17 May 2010 (UTC)

"Et least two, Brute!" ←Baseball Bugs ^{What's up, Doc?} carrots→ 03:37, 17 May 2010 (UTC)

The Mixing time of the atmosphere is relatively short. The oceans complicate matters considerably though. It has a much longer mixing time and the exchange of oxygen atoms between the ocean and the atmosphere interferes in the simple calculation presented above by Steve. Dauto (talk) 03:31, 17 May 2010 (UTC)

"Et least two, Brute!" ←Baseball Bugs ^{What's up, Doc?} carrots→ 03:37, 17 May 2010 (UTC)

I think the version I heard was about each breath you make, rather than your entire body. So we have 5x10¹⁸ kg of air in the atmosphere and 5g of air in each breath, so each breath makes up 10^-21 of the atmosphere. There are about 2x5/14x6x10²³ atoms in a breath, so each breath will have 400 atoms from Caesar's last breath. There is probably sufficient margin of error in this calculation for it to be consistent with the 2 atoms quoted by the OP. For the reasons given by BenRG, this calculation has no real world meaning, it is just a way of demonstrating how small atoms are. --Tango (talk) 16:19, 17 May 2010 (UTC)

The actual answer to this question is that it has no answer because atoms are quantum particles. If you shake a bottle of oxygen for 2000 years and then remove a few molecules, the molecules that you remove can't be identified with any molecules that went in. If anything, it would be better to say that each molecule that you pull out is an equal mixture of all the molecules that went in. (It does still make sense to talk about mixing efficiency, though—if the mixing is not perfect then you may be able to make more-or-less vague statements about the distribution of Caesar's last breath, though you can't say anything about individual molecules. But I think that 2000 years is enough to thoroughly scramble the atmosphere.) -- BenRG (talk) 08:15, 17 May 2010 (UTC)

Your response reminds me of this comic: [2]. While the individual atoms are indistinguishable from and identical to one another, the one that you put in the bottle is still unique. In a bottle of gas, the molecules will generally behave as largely classical entities – ping pong balls bouncing off one another – no need to treat them quantum mechanically. If I put one ping pong ball in a huge bottle of ping pong balls and then shake, I won't be able to identify the (apparently identical) ball I put in afterwards — but that doesn't mean that any sort of ping-pong-quantum entanglement has occurred; it just means that they all look the same. TenOfAllTrades(talk) 12:46, 17 May 2010 (UTC)

A similar question came up last year, here, regarding whether we have atoms in us that were once in our ancestors. It reached a similar conclusion to BenRG's answer. Oh wait it was BenRG who pointed it out last time too! 131.111.30.21 (talk) 09:20, 17 May 2010 (UTC)

By now we all get Baseball's joke. Cuddlyable3 (talk) 14:46, 17 May 2010 (UTC)

Can you supply a reference for this quantum mixing stuff? I've never heard this theory before, and I'm somewhat skeptical. Thanks! Aaadddaaammm (talk) 14:58, 17 May 2010 (UTC)

BenRG is right. The reference you are looking for is identical particles. Dauto (talk) 16:57, 17 May 2010 (UTC)

I don't agree with TenOfAllTrades criticism of BenRG's argument. The applicability of a classical model to compute physical observables is not relevant here. Of course, typical quantum effects like entanglement or interference cannot be observed, but quantum mechanics itself explains how that happens (via decoherence).

The point made by Ten is basically that if you put a molecule inside a container that already contains N identical molecules, you could follow in time where the molecule you put in moves to. So, if you take out a molecule from the container some time later, you can tell with certainty whether or not that was the same molecule that you put in earlier.

Let's first treat the molecules as non-interacting point particles and pretend that the container is perfectly isolated from the environment. To give Ten's argument as much room as possiblem, we assume that the intitial state after we put in the molecule is described by all the molecules being in some minimum uncertainty Gaussian state. Then, even though we have to explicitely symmetrize or anti-symmetrize the total wavefunction, you can still say that there is a Gaussian peak somewhere that wasn't there before the molecule was put in and that extra Gauusian peak will move due to the time evolution described by the Schrödinger equation. So, it looks like we can follow the extra molecule in time.

The problem here is that the width of the Gaussians will also become larger and larger over time. When there is substantial overlap between two Gaussians and you were to detect a molecule, you can no longer say which of the two Gaussians was responsible for that. Now, you can argue that the spreading of the wavefunction can be prevented by including interactions. However, including interactions between the molecules in the container alone won't work.

If you split the N+1 particle system as 1 plus environment consisting of the other N particles and argue that collisions lead to an effective collapse of the wavefunction, you end up with describing the state of the single particle as a statistical mixture of pure states. This is described by a density matrix. So, you then have an ensemble of narrow Gaussians, but they can then be more or less everywhere with equal probability after a short time. So, the information where your molecule is still gets lost. No surprise, of course, because the statistical description could never yield more information than the exact description.

But what if we include interactions with the environment outside the container? Even though that can work, this would amount to cheating as you then appeal to information that resides outside the system in this toy model, while in reality there is no "outside envirmonment". Count Iblis (talk) 15:32, 17 May 2010 (UTC)

TenOf AllTrades, unlike the pingpong balls which are only apparently identical, atoms are actually indistinguishable. BenRG is correct in his assesment. Dauto (talk) 16:57, 17 May 2010 (UTC)

The point is that you are all masking the reason for the little canard rather than explaining why it exists in the first place. While the "facts" about distinguishibility of individual molecules or atoms is somewhat debatable, that debate isn't really useful here. The purpose of the statement is to expose the sheer size of a number like Avogadro's number. Students have a hard time wrapping their heads around fantasticly large numbers. Additionally, since things like numbers of molecules and concentrations are often dealt with logarithmicly (using things like scientific notation and pH and other things like that) it can be very difficult to truly grasp how small molecules are, and how many of them there are in something. Considering that we all have over a billion oxygen atoms breathed by Caesar in his last breath in out lungs right now, and that such statement is mathematically verifiable via simple arithmetic is a useful demonstration of the scale of the numbers involved. The fact that quantum mechanics makes the statement incorrect doesn't matter, its supposed to be a heuristic exercise, not a scrupulously true statement. Yes, eventually, if a student goes far enough in their education, the more accurate models of molecular behavior should be taught. But a sixteen year old who is being exposed to chemistry for the first time is not likely intellectually prepared for such information when they are still trying to wrap their heads around the idea of the mole and stuff like that. The statement is mathematically accurate, and that is all that is needed here. --Jayron32 19:39, 17 May 2010 (UTC)

(ec) All of this quantum nit-picking is all very wonderful - but surely it doesn't matter whether the mixing is at the quantum level or at the classical ping-pong-ball-atom level. The average amount of atomic 'stuff' that was a part of Caesars' last breath is goind to be the same. I'll agree that one oxygen atom is pretty much as good as any other - and there is a philosophical question of whether two things that are quite utterly identical can be said to have separate identities - but the question isn't about whether we could measure the number of ex-Caesar atoms - merely if they are statistically likely to be present. One could (in principle) insert some fancy isotope of oxygen into the dying emperor's lungs and measure the amount in later generations...so it would be possible in principle to do some kind of physical measurement (if you knew in advance of course!). Mostly, this question is a wonderful thought experiment that really drives home just how ungodly huge Avagadro's number really is...and it does a great job of doing that, IMHO. SteveBaker (talk) 19:41, 17 May 2010 (UTC)

I generally agree with both of you (Jayron and Steve). But particle indistinguishability is interesting too. Saying that particles are "indistinguishable" in the QM sense is not the same as saying that they have identical properties. The theory allows for particles that have identical properties but aren't "indistinguishable". The three fermion generations would be an example if it weren't for the Higgs interaction giving them different masses. Indistinguishable particles have different statistics than merely identical particles, so this isn't a philosophical question. In undergraduate QM, indistinguishability has to be introduced by hand and seems arbitrary, but in quantum field theory it's automatic, therefore (getting philosophical here) I imagine that the "reason" for indistinguishability is the QFT reason. That reason is that particles are oscillations of a field, and two particles in the same state is an oscillation with twice the amplitude. How do you split that into two pieces? Well, you don't. The lesson from particle indistinguishability isn't that molecules are like ping-pong balls except for a quantum technicality; it's that gases have a wave nature as well as a particle nature. -- BenRG (talk) 05:50, 18 May 2010 (UTC)

But that still doesn't really affect the result. Asking whether the proportion of that wavey/particalish stuff that came from the wavey/particalish stuff that was in Caesars' lungs amounts to two-atoms' worth is still a perfectly reasonable question. All this quantum debate does is to change what it means for the air to be "adequately mixed". SteveBaker (talk) 14:30, 18 May 2010 (UTC)

It's worth using terms accurately. Steve Baker estimated above that a breath is about 5g, but that estimate is based on human lung capacity (about 5-6 L) when a "breath" (or "tidal volume") is about 1/10 of that. Thus, all of the estimates are about 10-fold too high. The numbers are still huge, but let's not confuse tidal volume with total lung capacity. One could argue that Caesar exhaled the contents of his lungs with his last breath, but one could also argue that his last inhalation was likely to have been at low volume - it seems better to stick with "breath" =~ "tidal volume" as a concept. -- Scray (talk) 01:59, 18 May 2010 (UTC)

That depends on how you define "last breath" - was it the air he breathed in just before he died? The air he breathed out just before he died? The total air in his lungs at the moment of death? Each of those things produce different results...I happened to pick the last of the three. The other two are tougher to estimate...did he <gasp> "Et Tu Brute" or did he merely whisper it? Did he yell "OW!" when the various knives went in? Did he die before breathing out or just after breathing out? However, all of this is irrelevant compared to the error bars in the "even mixing" assumption and the complications of oxygen exchange with the world's oceans, oxygen getting lost to corrosion of metals, etc, etc. Fortunately for this claim, the half million atoms calculation leaves plenty of room for "at least two" atoms to be present in each of us no matter which set of assumptions you choose. SteveBaker (talk) 14:24, 18 May 2010 (UTC)

Efficiency of digestion

I've heard that there is something in pineapple that helps break down certain proteins and thus makes it easier for your body to absorb? them. I'm wondering about something similar, but in a more general sense.

Is there anything that has a similar effect in a larger scale, e.g. a foodstuff that assists the bodies digestive system allowing for more efficient digestion, or a "habit", for example eating slower or more frequently. There are obviously a lot of people who would benefit from food having less impact on their bodies, but I'm asking about the opposite effect, for example if an athlete wanted to waste less by increasing the efficiency of their body when digesting foods/vitamins/minerals etc. Thanks! 219.102.220.188 (talk) 05:16, 17 May 2010 (UTC)

Why are you are assuming that exercising the digestive system by normal eating is a bad thing? If you ate baby food your whole life maybe your digestive system would atrophy and stop working properly all together… would you also assume wearing a pressurized gas mask that assists your lungs by helping you breathe would be a good thing? In general chewing your food properly isn't a bad idea, but I wouldn't systematically take "digestive aids" just to make the job easier. Just because it takes some effort doesn't mean it's not efficient. Vespine (talk) 06:18, 17 May 2010 (UTC)

I understand that there is a very fine line between "adding benefits to the body" and "doing something that the body isn't supposed to do", but I don't think I explained myself well. In the case of morbidly obese people, could it be said that the digestive system was incredibly "efficient"? Probably not, but in some way (in some cases, obviously this isn't the only cause of obesity, probably not even a major cause) the body transfers more of the food intake into fat/muscle/growth than an average person would. On the other hand, there are people that can eat ridiculous amounts of food with hardly any transfer into fat/muscle/growth. 210.165.30.169 (talk) 01:02, 18 May 2010 (UTC)

Pineapple contains Bromelain which helps break down protein (like meat). See Digestive enzyme for many others. But I'm not sure anyone actually eats those, they are usually made by the body. But I guess you could. Ariel. (talk) 08:00, 17 May 2010 (UTC)

They are sold as supplements, and people do use them. 210.165.30.169 (talk) 01:02, 18 May 2010 (UTC)

(slightly off-topic) The biggest aid to digestion ever invented by humans was cooking. A recent experiment where some volunteers ate only uncooked vegetarian food ended with them having to spend all day eating and still feeling hungry. Dbfirs 08:49, 17 May 2010 (UTC)

Papayas contain papain, another enzyme that breaks down protein. And you can buy a product called Bean-O that helps break down parts of beans that many people have difficulty digesting (so that they don't get processed by gut bacteria with notably gaseous results). Those are the only two I know of; there may be more. There are many more foods that have the opposite effect, by reducing the ability to absorb certain nutrients. Looie496 (talk) 16:21, 17 May 2010 (UTC)

Nutrition Stores sell digestive enzyme formulas that have almost every enzyme needed to breakdown all food. They are great for heartburn. But use them sparingly or your stomache will become dependent on them to digest your food. —Preceding unsigned comment added by 165.212.189.187 (talk) 16:30, 17 May 2010 (UTC)

I think the OP is still making a mistake in their assumptions: there are people that can eat ridiculous amounts of food with hardly any transfer into fat/muscle/growth. I doubt this has anything to do with efficiency of digestion. I thought Metabolism but that article isn't quite what I mean by that word either, maybe my idea of "metabolism" is a fallacy too. Vespine (talk) 05:40, 19 May 2010 (UTC)

Basal metabolic rate may be what you're abbreviating to metabolism, Vespine. --203.22.236.14 (talk) 12:56, 20 May 2010 (UTC)

is it possible to follow a diet that would lead to perfect digestion and no poop? 109.205.114.118 (talk) 11:26, 21 May 2010 (UTC)

Saltation velocity - appropriate factor of safety

I have used the rizk correlation to find the saltation velocity of reasonably fine particles (~100-200 micron) in dilute phase pneumatic conveying. what is an appropriate factor of safety to take? and is there some scientific backing to this factor or is it something just widely followed? —Preceding unsigned comment added by 125.17.148.2 (talk) 05:33, 17 May 2010 (UTC)

Our article, Saltation (geology), is not very detailed or numerical; but I found this textbook: Pneumatic conveying of solids: a theoretical and practical approach. You might be able to find it at a university library. With a title like that, I would bet it explains the practical and theoretical support for any of the equations it uses. Nimur (talk) 16:13, 17 May 2010 (UTC)

Thanks. I have that book - it certainly gives nice theoretical models for saltation velocity but doesnt mention the factor of safety sadly. —Preceding unsigned comment added by 122.172.37.82 (talk) 15:52, 18 May 2010 (UTC)

Is "dilute phase pneumatic conveying" solid or liquid? You should be able to follow methods used in aeolian sediment transport and/or in Leo van Rijn's (or other saltation-based) sediment transport formulas to know something about it. As for factor of safety: what are you concerned about? Particle impacts? Particle velocities? Awickert (talk) 02:15, 19 May 2010 (UTC)

I am not sure I follow. Pneumatic conveying is the transportation of fine particles through ducts by the use of flowing gases (gen air). Saltation velocity is the min air velocity req to keep the particles in suspension. Factor of safety is the factor by which SV is multiplied to ensure particles never settle down which may lead to choking of the ducts. I dont know about aeolian or LVR's formulae but they seem to b something else altogether right? —Preceding unsigned comment added by 125.17.148.2 (talk) 05:23, 19 May 2010 (UTC)

Ah, OK. I work in sediment transport often, in which "saltation velocity" is the particle velocity. I imagine that you could do a Rouse number-style balance on the grain between upwards-directed lift and downwards-directed settling and then make sure that this is well in favor of lift, but I can't give you any cut-and-dried answer. Awickert (talk) 06:04, 19 May 2010 (UTC)

Thanks. I will look into it. —Preceding unsigned comment added by 125.17.148.2 (talk) 12:25, 19 May 2010 (UTC)

Overclocking

Is overclocking my PC possible???What are the requirements??? And if it is possible....is it risky??And risky to what extent???Rohit.bastian (talk) 05:47, 17 May 2010 (UTC)

Apart from the fact that this would more suitably belong on the computer ref desk, you are probably much better off googling a beginner's guide to overclocking. It's not a subject that can be summarized in a paragraph and anything anyone writes here is probably written a hundred times before on websites all around the net. This might also be a good start Overclocking. Vespine (talk) 06:04, 17 May 2010 (UTC)

I should think that any chip manufacturer would be interested in seeing a nice high clock speed for their product, as high as they could possibly manage -- and still have a reliable product with minimal failure rates. So if you start overclocking you expose yourself to the possibility of problems. So ask yourself -- what good is a chip 20% faster than normal if it fails after 18 months, when you could have it for five years? Just food for thought -- these figures aren't based on any empirical research. I just would firmly expect Intel et el would set their chip speeds beneath the maximum for very good reasons. They've been doing this for a long time now. Vranak (talk) 08:26, 17 May 2010 (UTC)

Yes, most CPU's can be overclocked. But don't step them up too fast, or you will be left with a computer that won't start. Then you will have to move the CMOS reset jumper (at least that's how it is on my family's 9 going on 10 computers). --Chemicalinterest (talk) 10:42, 17 May 2010 (UTC)

Even if you overclock, unless you're doing something like overvolting or are allowing the CPU (or whatever) to get too hot & don't have sufficient cooling (which is always a problem) it's very rare the CPU will die in 18 months. In other words, 5 years+ is an entirely resonable expectation for a CPU lifespan even if overclocked. I'm not saying you won't reduce it, you may very well do so but in general if you know what you are doing the reduction in lifespan is not likely to be something the vast majority of people will notice.

You're wrong BTW about 'would be interested in seeing a nice high clock speed for their product, as high as they could possibly manage'. Why would they? That's not the way the market works in general be it CPUs or anything else. There's no point selling your CPUs at 3ghz if you can sell them at the same price but operating at 2.5ghz. You can sell some CPUs at a higher price for 3ghz, but it's unlikely you can expect to sell enough at 3ghz to make up for not selling any at 2.5ghz and if you just flood the market with 3ghz CPUs you create a glut. While these figures are completely made up, the general idea is sound. A manufacturer is going to want a decent range of CPUs for price differentiation depending on what the market demands and other things, if most of their CPUs can go close to the top speed, they just bin many of their CPUs lower then they have to be. They don't sell all their CPUs at the maximum possible clock speed and make a lower profit just to be kind.

Your point is well-taken, but a company whose processors are kept needlessly slow will not do well against a more honest and ambitious firm. Vranak (talk) 16:21, 17 May 2010 (UTC)

But Intel's only real competitor in the desktop and normal laptop space is AMD, and they do the same thing of course, as do competitors in most markets, and have problems for a variety of reasons achieving the high end that Intel does (at least when Intel isn't fooling around with stupid things like they were in the P4 era) nor have they ever been able to achieve the volume of Intel (one of the things that held them back in the P4/A64 era) so they aren't really a risk in that area anyway. Sure a competitor helps to keep prices down BUT they're not going to shot themselves in the foot by completing ignoring basic and sensible marketing practices. Note that if they did, it could easily be only several months before they're completely screwed themselves. Nil Einne (talk) 07:19, 20 May 2010 (UTC)

Also while process technology and other factors are always improving which enable you to achieve higher clock speeds or otherwise faster CPUs they won't always give you the advancement you desire at the rate you desire. Moore's law (from an engineering standpoint) generally holds over long timeframes, it may not always hold over shorter timeframes. If you flood the market with highspeed CPUs, then 6 months later you've only got 10% faster CPUs, then what? Far better to hold back your higher speeds so you can smooth out the speed increases.

Finally even if you can achieve higherspeeds, you also need to do it at whatever thermal design power you specify for your CPUs. If you can achieve a higher clockspeed but are above the TDP you specified, you either have to sell your CPU at a higher TDP or clock it lower. From an end user standpoint, if you know what you're doing, don't mind the additional noise and cooling requirements and have components to support such a TDP, then that's not a problem.

I'm not of course saying I recommend the OP overclocks. Actually if you have to ask at the RDS, I'm not sure if I would recommend it until you've done much more research. If things like Prime95 and stability testing (or if you think stability testing only means 'the computer doesn't crash') don't mean anything to you, if you don't know how to reset the CMOS etc, then personally I won't.

Nil Einne (talk) 12:35, 17 May 2010 (UTC)

Note that overclocking will probably not have any noticeable effects. CPU speed is one factor among many in the performance of modern desktop computers, and not the most important. Memory size and disk speed are more important for most things, and the graphics card is more important for games. A faster CPU probably won't have a noticeable effect, unless you're into protein folding or raytracing, which require little memory and lots and lots of computation. Paul (Stansifer) 13:56, 17 May 2010 (UTC)

The CPU chips that come off the production line probably have a spread of performance similar to a bell curve. The company quality controllers estimate values of the maximum and spread of the distribution. The sales department are concerned that equipment manufacturer customers should not reject chips so the clock rate they promise is far down the tail of the distribution. That means the majority of CPU's have much better potential clock rate than their specification. In other words a company with a reputation to protect supplying thousands of CPU's is more conservative about taking risk than an enthusiast interested in getting the most out of their PC needs to be. Cuddlyable3 (talk) 14:37, 17 May 2010 (UTC)

There are two reasons why (careful) overclocking works:

1. Manufacturers often run their manufactured chips through speed tests where they figure out which chips (by luck) happen to be able to run faster than others. They then sort the devices by speed range and sell them at different price points. However, if the speed selection is set at (say) 4GHz - they'll typically sell the ones that pass at 4GHz as 4GHz chips - and retest those that fail at (perhaps) 2GHz. Then they'll sell those that pass this second test as 2GHz chips and scrap the rest. Well, it might be that your 2GHz chip actually only just failed the 4GHz test and would actually run quite happily at (let's say) 3.5GHz. Of course, it might not and it could just as easily crap out at 2.1GHz.
2. Heat. What limits the speeds of many chips is heat. The faster you clock a chip, the hotter it gets. There comes a point where it'll overheat and die (possibly permenantly). If the chip was designed to work with a particular kind of heatsink and fan such as (let's say) DELL are prepared to use - then it will probably overheat if you just overclock it. However, if you can install substantially better cooling (eg using one of those fancy liquid cooling systems) then you can probably overclock the thing to the point where it would melt using the standard cooling system.

However, this is always a risky business - you may well end up with a computer that becomes flakey and crashes out or behaves weirdly - or doesn't work on warm days - do even dies an early death because of the extra heat it's experiencing. For that reason, I wouldn't recommend doing it unless this is an old computer that you don't much care about - and which you're only using for relatively unimportant jobs like playing games or surfing the web - where the occasional crash is worth the speedup. SteveBaker (talk) 15:48, 17 May 2010 (UTC)

Do people actually use computers for anything else? Matt Deres (talk) 16:21, 17 May 2010 (UTC)

I spent a few minutes trying to come up with a witty response to that, and couldn't think of one. Err...Instant messenging? Vimescarrot (talk) 19:08, 17 May 2010 (UTC)

Someone has to actually write the games...er...and I'd better stop writing RefDesk answers and get back to doing exactly that! SteveBaker (talk) 19:32, 17 May 2010 (UTC)

But playing games and surfing the web probably won't be improved by overclocking. Games tend to be GPU-bound and memory-bound, and surfing the Web is very IO-bound. I use a few CPU-bound programs (POV-Ray, Foldit) but making my test renders take 20 seconds instead of 25 is not so valuable that I'm willing to go to the trouble, let alone risk damaging my computer, or having it fail sporadically and not being able to track down the problem. Overclocking probably made sense back when CPU speeds weren't insanely much faster than memory and hard drives, and graphics acceleration didn't exist (Though apparently it is possible to overclock memory, also. That sounds more like it could be useful, if you're unusually risk-tolerant. But overclocking won't do anything about having too little RAM.), and you could play DOOM more smoothly with a 25% higher clock speed. Paul (Stansifer) 21:15, 17 May 2010 (UTC)

Yes, that's my impression as well. The people who most wanted to overclock were gamers and they've moved on to console games instead. I suppose there are still folks who find it fun or enlightening to see how much they cane tweak their hardware, but I'd guess that most people wouldn't want to risk it - even if the odds were actually pretty decent. If you're not in a position to upgrade or buy a new system, you're probably not in a great position to risk losing the old system, ya know? Matt Deres (talk) 23:53, 17 May 2010 (UTC)

It's also possible to overclock your GPU - so the argument that you shouldn't bother overclocking because you're typically GPU-bound (something I'm not sure I agree about) doesn't entirely hold water. SteveBaker (talk) 01:31, 18 May 2010 (UTC)

In fact, I don't really understand the memory comment either. In many cases, improving memory bandwidth by overclocking makes only a small difference in real world performance (timings sometimes help more although even then not necessarily that much). Of course if your CPU or app is significantly limited by memory bandwidth it will make a difference but this often isn't the case (well most of my knowledge is mostly 2-3 years out of date but I don't think things have changed that much, e.g. from a quick search [3]). Sure you can get great results with sythetic benchmarks, but who cares?

In addition, the risk of losing your computer is questionable for some overclockers. Yes there's a small risk, but how big if you know what you're doing? In particular unless you do something stupid, the chance anyone will know you overclocked your computer is slim if you don't tell them so while you may not be entitled to warranty support, it doesn't mean you won't get it and many people may not have ethical qualms about this for whatever reason (and obviously it's a complicated issue, some may argue it's difficult to know if their overclocking contributed to the failure), whatever the harm to non-overclockers and whatever others may think of them (and clearly the RD isn't the place to discuss such people, except to acknowledge they exist).

People seem to be forgetting that a slightly better processor can sometimes be double the price. In other words, you could get a cheapish processor and in some cases you could easily overclock it so it performs better then a processor which is close to double the cost (the more expensive processor would usually be more on average overclockable to a higher speed of course but it's irrelevant to the point). And even in the unlikely event it does die and either because your honest or the manufacturer can tell you did something dodgy it doesn't get replaced, you can buy a new one and you'd still spent the same as you would have if you got the more expensive one (and either decide to risk overclocking it again or you don't have the money to replace it don't). In such a case you're worse off then you would have been if you'd bought the more expensive processor but most people won't be such a case.

Even if you fiddle with the costs, considering the small risk of having to spend more, it may be worth it for many who choose to overclock. Do remember the price of the CPU is only going to be a part of the computer sometimes even a relatively small part. (I would note of course that the chance you'll kill all the components at the same time when overclocking one is slim to none, in fact in such a situation one may presume something, say the motherboard or PSU was inherently defective.) So even for an honest person the money equation may not work out in favour of buying the better processor and not overclocking. (If you throw in the cost of the time spent on overclocking perhaps.)

One final point. The question of stability and reliability is an interesting one. Anyone who's worked with computers enough knows that particularly when assembling your own with a variety of components, things can go wrong. (How many people really check out a mobo's memory compatibility chart?) If your lucky, these will be obvious. But what about errors that may not affect stability that greatly but may have some affect on reliability and are only likely to be easily detected by Intel Burn, Prime95, LinX or similar programs and of course memtest* for memory (note that if you think these sort of programs are simply supposed to make your computer crash, you may want to read up on them)?

While rare, you definitely can get these sort of errors with a new system and everything running at spec. How many people actually carry out such tests on their computer? Personally I'd probably trust an overclocked system, put together and well tested by someone who knows what they're doing then one running at spec but put together by someone who has little clue and only really browses the internet with the occasional solaitare on it. Yes the system run at spec, and put together and tested by someone who has a good idea of what they're doing would be even better but that's beside the point. (Of course there's also a question of component quality. E.g. without testing, I'd probably trust mildly overclocked good brand RAM then some junk noname brand someone bought off eBay running at spec. Or with testing if both come up fine perhaps even for a stronger overclock. Particularly if it was found the good brand can handle say 250 mhz but is run at 230mhz, the other brand craps out at even 205mhz from the stock 200mhz.)

I'm not saying that the stress test tools are perfect, the fact that particularly for CPUs you'll ideally want to use several different ones, and different ones are more likely to pick up errors in with different CPUs is good evidence for that, not that you really need any evidence. Clearly manufacturers have access to tools and far better knowledge of how to stress test what they manufacture but it's not a simple equation of overclocking=unreliable/untrustable, spec=reliable/trustable. In other words, I'm not suggesting I would recommend an overclocked system if the output really matters and you really can't afford the risk of stability problems, I wouldn't (although I would also probably recommend things like ECC memory and a bunch of other stuff in such circumstances).

(I should clarify in case there is any confusion that I use stability above to mean 'causes crashes and other such problems' and reliability to mean 'output is what it should be'. These aren't exactly standard distinctions and they are mostly one and the same, if the output isn't what it should be, this could easily cause a crash or related problem but I find it helpful to make the distinction because some reliability problems won't be noticed except in very rare or specific circumstances. E.g. if you happen to get a Prime95 error after 15 hours and then next time you run it for 30 hours no error and next time you get it in 8 hours, it's probably fair to say you won't often notice that problem.)

BTW, SB always point out the always GPU bound assertation for gaming is questionable but there are a number of other things e.g. video compression (with modern codecs e.g. x264) will often be CPU bound (GPU encoders are still fairly crap), finishing a job in 7 hours instead of 10 may very well be worth it for some people.

P.S. I'm obviously only talking about memory bandwidth with the CPU, not the GPU. I didn't really consider GPU overclocking at all.

P.P.S. I should add that while many lament the state of PC gaming, and there's hottest growth areas are probably the casual and MMO crowd which aren't generally that demanding it definitely isn't dead.

Nil Einne (talk) 01:07, 19 May 2010 (UTC)

Thank you mates..it was awesome info!!I never knew contacting wikipedia wud make things so awesome!!! —Preceding unsigned comment added by 117.204.3.236 (talk) 16:48, 18 May 2010 (UTC)

BTW, I suggest you clean your keyboard. It seems to have the annoying habit of repeating question marks and exclamation marks multiple times. StuRat (talk) 21:43, 19 May 2010 (UTC)

amtrak "service disruption"

does Amtrak "service disruption" mean a suicide

Not necessarily. When people suicide on the Caltrain, they often euphemize it for about an hour or so, but within a very short time, even the CalTrain officials are announcing what happened and where. In general, there are other reasons for service disruption, though suicides are real and frequent[4][5][6][7]. Counseling and a suicide prevention hotline is available now as a response to the issue. Trains around these parts are delayed for all kinds of other reasons - baseball games, unexpected ridership levels, weather, and so on; but suicides tend to be the longest disruptions to clear. Nimur (talk) 09:41, 17 May 2010 (UTC)

what do you mean by suicides tend to be the longest disruptions to clear.

It is likely that the suicide scene must be photographed before anything is moved, the train driver will need to be interviewed, counselled ("this wasn't in my job description!") and/or replaced, and diverse body parts may need to be collected and rejoined in a bag. All this takes time. Cuddlyable3 (talk) 14:16, 17 May 2010 (UTC)

Questioned? Replaced? Why? Can any train driver be responsible for any suicide when it is impossible to stop a train in such a short distance to avoid running over anyone who is stupid or bored with life enough to wander onto the tracks? --131.188.3.20 (talk) 14:43, 17 May 2010 (UTC)

The driver will need to be questioned just as any witness to any violent death would need to be questioned. As to replaced - the experience of someone doing a "one under" is very traumatic for the driver, and he will need time and support to deal with the feelings that result. I think I'm right in saying that in the UK drivers have to take time off and have counselling before returning to work after an incident like this - some never manage to return to driving trains. As to other reasons for the long time it takes to restore services - body parts can be scattered over considerable distance, or entangles with the underparts of the train. I used to work on the railways (not as a driver), and I can attest that "one unders" do take an age to clear up, as well as very seriously impacting on the emotional well-being of drivers and of those involved in clearing up the mess. As a result, my initial reaction to news of an event like this is one of anger towards the selfish so-and-so responsible, rather than one of sympathy for the emotional pain they must have been going through. DuncanHill (talk) 15:13, 17 May 2010 (UTC)

I can confirm that - I spent a little while making simulators for railroad locomotives. When the company I worked for first started to get into that business (we'd previously been making flight simulators), we put together a 'demo scenario' to show off to potential customers. Since we knew very little about the business, we did the thing where a school bus stalls out on a railway crossing - and just manages to get started and move out of the way as the locomotive gets there - the idea being to perhaps train drivers to be alert to the dangers and slam on the brakes. We showed it to our first customers and were told in no uncertain terms that we should NEVER show that kind of thing to their drivers! Firstly, under those circumstances, there is nothing they can do about it, the stopping distance of the train is such that if you can see a problem like this, it's already too late. Secondly drivers who had been though the stress of hitting something like a car or whatever would have those traumatic memories reawakened and that this is highly undesirable. There is no question that under such circumstances, the driver needs months of help and cannot return to work for an extended period after the event. (So that demo got flushed in pretty short order and replaced with something much less exciting!) SteveBaker (talk) 15:26, 17 May 2010 (UTC)

As a longtime Amtrak Capitol Corridor rider, I can tell you that service disruptions can be caused by mechanical breakdowns, floods, malfunctioning bridge damage sensors, track repairs, and police incidents -- not just suicides. Looie496 (talk) 16:15, 17 May 2010 (UTC)

And lots of causes for delay are euphemistically described obliquely. Delays on the MBTA are often described as due to "police department activity" or "fire department activity", which conjures up an image of some convention of uniformed people randomly deciding to have a party on the tracks. Probably, they don't want riders imagining that a giant inferno has consumed Park Street, when it's a minor fire. Another rail system might choose to call everything a "service disruption" because that's not a whole lot less informative. Paul (Stansifer) 00:58, 18 May 2010 (UTC)

Is the selfish gene shortsighted?

If a gene is successfully passed on to 2 children then it is traditionally considered successful. However what happens after that? The gene is only 50% likely to pass on to the next generation with there ultimately being diminishing returns. So why a gene fight to continue when it's inevitable it won't survive more than a few generations? The reasoning I can think of is in a small enough population a gene could gain dominance in every member of the population and then be immortal. But in large populations the odds of that seem exceedingly low. TheFutureAwaits (talk) 15:23, 17 May 2010 (UTC)

Genes don't fight. They get passed on randomly. What you are doing is anthropomorphising genetics by assigning things like "motive" and "purpose". Genes get passed on via a process known as "dumb luck". Sometimes, a gene provides an advantage over an alternative gene, and as such, it tends to cause its host to survive longer and have more kids. Sometimes, this occasionally results in the gene being spread to more and more organisms in successive generations. But its all a random process, and there is no "intent" on the part of a gene to "spread itself". It just happens randomly. Its like the snow on a busy street. The snow that gets thrown off the street onto the side of the street survives longer than the snow that gets thrown to the middle. This is just because it got thrown out of the path of the cars, not because it fought to get to the side of the road. Likewise, a gene that provides some benefit to its organism tends to get passed on not because it fought to survive, but because it happened to cause some benefit. But it had no agency in causing its own survival. --Jayron32 15:31, 17 May 2010 (UTC)

I'm not sure I agree with any of your assertions! I don't know that there is such a strict definition of what makes a gene "successful" - and you're wrong about the gene only having a 50% chance of being passed to the next generation. Remember that in each generation, the number of descendants of the animal/plant that has that gene increases - so even though the chances of it being passed on are only 50/50 in each offspring, the increase in the number of descendants (typically) more than doubles with each generation. For example, it has been shown that about one in two hundred men in the entire world have a gene on their Y chromosome that came from Genghis Khan. Certainly he had a lot of immediate offspring - but beyond that first generation, over 800 years, all of the subsequent spread has been fairly normal. So a truly successful gene can be present in 120 million offspring within about 40 generations. SteveBaker (talk) 15:36, 17 May 2010 (UTC)

(edit conflict):Why do you think that "the gene is only 50% likely to pass on to the next generation" ? If an individual carrying the gene mates with an individual not carrying the gene, then 50% of their offspring (on average) will carry the gene. And if carriers of the gene have a higher number of offspring (again, on average) than non-carriers then the proportion of gene carriers in the population will increase with each generation (up to a limit). Gandalf61 (talk) 15:41, 17 May 2010 (UTC)

I think Jayron's explanation is trying to point out that there is an element of randomness, even when a gene provides a clear advantage for an organism's survival or reproductive ability. The genes are not "motivated" to survive; but those genes which increase survival have a much better chance of proliferating. Nimur (talk) 16:03, 17 May 2010 (UTC)

Suffice to say, procreation has a long history of compelling people onwards and upwards. This whole business about genes competing is, as was said before, a little wrong-headed though. They are expressed, may become latent, may re-appear. Nothing is really gained or lost. It's all a flux, love it or hate it. Vranak (talk) 16:18, 17 May 2010 (UTC)

No, we haven't gained anything since diverging from our non-human ancestors. 67.243.7.245 (talk) 21:38, 17 May 2010 (UTC)

I disagree. I think the ability to eat extra mature farmhouse cheddar throughout our lives is a definite gain (you can keep your fashionable foreign rubbish - give me a decent cheddar any day!), and lactose tolerance in adulthood only evolved since we started domesticating animals a few thousand years ago. --Tango (talk) 23:31, 17 May 2010 (UTC)

It's also fairly unconvincing to dismiss the genius of Mozart, Handel, ABBA etc. I feel we may gained a thing or two since diverging from our simian ancestors, to be honest. The only question is whether you're in an appreciative mood or not. Vranak (talk) 01:11, 18 May 2010 (UTC)

Yeah - 67's claim is nuts. Of course we've evolved since then. We've evolved really noticable changes like a range of skin colors and hair types as we adapted to living in climates with differing amounts of sunlight. We've gotten taller and smarter. Lactose tolerance is a great example (and one I use a lot in discussions here) - because it seems to have evolved in just the last 4000 years in response to our ability to keep farm animals. Furthermore, that evolutionary change didn't overtake the entire population, which is why some people are still lactose intolerant in adulthood, just like all the other mammals. SteveBaker (talk) 01:27, 18 May 2010 (UTC)

As a general rule, people compete for their genetics to be passed on to their candidate of choice instead of someone else's genetics, the genes themselves do no competing. The genes present in people more adept at passing on their genetics are the genes most likely to proliferate. Other than that, I echo the answers above. Ks0stm ^(T•C•G) 19:13, 17 May 2010 (UTC)

Storage of vitamin D

A while ago, I bought some soft gel vitamin D supplements (5000 IU). The label says that it should be stored in a dry place. However, I recently read somewhere that vitamin D is very sensitive to UV radiation, so I put it in a dark place. But it still has been exposed to a small amount of UV radiation during the few weeks that it was on my table. I'm wondering by how much the vitamin D content of the pills has been degraded. The UV index is around 3 this time of year where I live during a few hours around afternoon. The windows of my room are often open. The vitamin D is inside a plastic bottle. The bottle is white colored, but it is still a bit transparent to light. I do think it does block UV radiation to some degree.

Another thing to consider here is that the soft gel pills themselves are tranparant, so the UV radiation can affect the whole pill and not just the surface. Count Iblis (talk) 17:21, 17 May 2010 (UTC)

I very much doubt that they would have stored pills that were susceptible to UV damage in a UV-transparent container. If the label had said "Keep in a DARK, dry place" then I'd be concerned - but since it doesn't, I'm pretty sure you're OK. Also, it's possible that the gel in the capsules themselves will be UV-opaque, even if they are transparent to visible light. (The glasses I wear are like that - perfectly transparent to the naked eye, but incredibly UV-opaque). If the capsules are blue-ish in color then they very likely are designed to be opaque to UV. SteveBaker (talk) 19:30, 17 May 2010 (UTC)

Plastic breaks up in sunlight, I think, so I'd keep any out of sunlight if I can. 67.243.7.245 (talk) 21:35, 17 May 2010 (UTC)

Well, it does - but it takes a long time. I guess if it's a really gigantic pot of pills and you took them at really infrequent intervals, the container might break down before you finished the supply - but I honestly don't think that's a real problem. SteveBaker (talk) 01:17, 18 May 2010 (UTC)

Some plastics break down (not up) in sunlight, but the vitamins shouldn't be affected too much. Vitamin D is formed in your body when sunlight strikes a particular molecule. It is formed by sunlight (under limited conditions, of course), not decomposed. Vitamin C may decompose in sunlight since it contains a reactive double C=C bond, making it a reducing agent that reacts with the oxygen in the air. --Chemicalinterest (talk) 11:55, 18 May 2010 (UTC)

I've read that vitamin D formed in the skin will actually break down due to the very UV radiation that formed it. This leads to a dynamical equilibrium when you spend a long time in the Sun. A fair skinned individual can get 10,000 to 20,000 IU of vitamin D from the Sun in an hour or so. But 100,000 IU per day for many weeks in a row is a potentially lethal overdose. But due to this dynamical equilibrium you'll never get such a dose from the Sun.

That's why I am a bit worried about the fact that I put the vitamin D bottle on my desk for a few weeks. So, I hope that SteveBaker is correct about the vitamin D pills plus container being made such that this is not an issue. Count Iblis (talk) 14:10, 18 May 2010 (UTC)

The various things Steve said about UV and pills may or may not be right, however the bit about the people that make the pills not being stupid enough to not say on the label that you need to keep them in the dark if that were the case is almost certainly right. --Tango (talk) 21:40, 18 May 2010 (UTC)

Doctors who do nearly everything when caring for patients

You know like you see on shows like House, where the same doctor, or small team of doctors are involved in every aspect of their patient's care - diagnosing, performing x-rays and scans, administering the medication, operating, taking blood/tissue/CSF samples, performing the lab tests themselves, etc.? I know that this doesn't really happen in American or UK hospitals any more but do any doctors still work like this in other parts of the world? Or is this just something from the 'olden days' that people still believe happens until they actually experience hospital treatment for themselves? As far as I know, veterinarians still work this way (don't know the proper name for it). --95.148.107.150 (talk) 20:47, 17 May 2010 (UTC)

It may well happen when there are very few medics around, for example in the field hospitals set up following natural disasters or wars. --Tango (talk) 23:57, 17 May 2010 (UTC)

The better physicians in Vienna, Austria do just that (apart from surgeries unless they also specialize in this specific kind of surgery). Surely there's supporting staff (like lab techs performing blood tests) but the patient rarely sees them (the professor takes blood sample himself). Starts at a thousand euro a visit. Most patients don't need their level of experise and are fed into the ordinary conveyor belt, which ultimately costs nearly the same. East of Borschov (talk) 05:54, 18 May 2010 (UTC)

Sex on high peaks

Is there any record of anyone ever having sex at the summit of any of the world's tallest mountains? Is there a definate record of the highest ever intercourse? 188.221.55.165 (talk) 21:59, 17 May 2010 (UTC)

See Sex in space, and Mile High Club. Comet Tuttle (talk) 22:48, 17 May 2010 (UTC)

Stella Wobble

How can a stella woble detect whether there is one or more than one planets around a star? Most of the show I watch assume there is only one planet. --Tyw7  (☎ Contact me! • Contributions)   Changing the world one edit at a time! 23:00, 17 May 2010 (UTC)

Have you read our article on methods of detecting extrasolar planets, and in particular the section on stellar wobble detection? Nimur (talk) 23:07, 17 May 2010 (UTC)

OK - so you understand that gravity works both ways - the star pulls on the planet - and the planet pulls on the star - so the position of the star is very slightly shiftedtowards the planet. So as the planet orbits, the star wobblesto stay a little bit closer to it.

That wobble is very small - but it's enough to slightly shift the frequency of the light coming from it - which you can measure with a suitably sensitive detector.

With just a single planet, you can figure out the frequency with which the star wobbles - which would be the period of the orbit of the planet (how long its' "year" is) - and therefore how far away it is from the star. From the amount of the wobble - and knowing the distance at which it orbits, you can deduce the mass of the planet. So with just one planet, the center of the star makes a little circle - once per orbit - and it's all very easy to understand.

Now, suppose there are two planets. The star is attracted to both of them. But they don't orbit the star at the same rate - one might take a year to go around the star (like the earth does) - the other might zip around every 3 months (like Mercury does). When both planets are on the same side of the star, it gets pulled in the same direction by both planets and it's much further off-center than it would be if there was just one. When the planets are on opposite sides of the star - their pulls tend to cancel each other out - and the star sits in the middle. At other times, they are pulling at right angles. If you plotted a map of the position of the star over a few years, it would look something like a spirograph picture. You can use a mathematical technique called Fourier analysis to figure out the separate frequencies and magnitudes of the shift due to each planet...and if your measurements are accurate enough - and done over a long enough period - you can figure out the contributions of many planets. Of course the smaller and more distant planets induce much smaller wobbles - which are tough to detect from such a long way off...so there is a practical limit to how much of this complex motion we can figure out - and to what degree planets are detectable at all. SteveBaker (talk) 01:13, 18 May 2010 (UTC)

Excellent summary. Just to clarify, however: Doppler spectroscopy is a separate technique from direct observation of the star's position. Both techniques have been used to identify or characterize extrasolar planets, but doppler (frequency-domain) methods are significantly more sensitive; direct observations can only even potentially work for our closest neighboring stars. Other methods include transit detection (careful and extremely precise measurement of the star's brightness to check for a planet crossing its path), and other advanced image and spectral processing techniques. Nimur (talk) 01:30, 18 May 2010 (UTC)

Right - so the 'transit detection' thing only works in the relatively unlikely case where the planet is orbiting in a plane such that it comes between the star and us here on earth - it's interesting because it offers the possibility that we could measure the diameter, albedo and atmospheric composition of the planet as well as it's mass and orbital radius. But because that orbital angle is just a lucky coincidence, you can't use it on every star that's out there. The doppler spectroscopy approach works better for getting mass and orbital radius - but still can't detect every extrasolar planet because when the planet's orbit is in a plane that's at right angles to our line of sight, the star's wobble is at right angles to our line of sight and doesn't produce a doppler shift. The direct measurement approach (where you literally just look at photos of the star and see if it moved from one month to the next) can solve that problem very well - but is only useful for "nearby" stars with gigantic planets because otherwise the wobble is too small to detect as a direct positional shift with our telescopes. However, you can use fourier analysis on the position, doppler and brightness variation of the star to discover multiple planets with all three techniques. SteveBaker (talk) 14:11, 18 May 2010 (UTC)

Just want to point out that the star wobbles away from the planet(s), not toward. The center of mass of the whole system doesn't wobble, and the star and planet(s) have to surround the center of mass on both/all sides. -- BenRG (talk) 06:19, 18 May 2010 (UTC)

Argh! Of course, you're right...my bad! SteveBaker (talk) 13:33, 18 May 2010 (UTC)

Also of interest is the detection of planets around radio pulsars. This is also a doppler-shift technique, but more precise than optical spectroscopy. The measurements of the PSR B1257+12 system are precise enough that the planet-to-planet gravitational interaction can be observed. -- Coneslayer (talk) 17:18, 18 May 2010 (UTC)

Purity of USP and food-grade glycerin

How do the two grades of glycerin compare, in terms of the allowable types and levels of impurities? --72.94.148.35 (talk) 23:36, 17 May 2010 (UTC)

USP glycerin just seems like it is approved for drug use, while food-grade is for food use. USP may be for food use, too. --Chemicalinterest (talk) 00:16, 18 May 2010 (UTC)

This document from the Soap and Detergent Association states that "The specifications for food grade glycerin given in the FCC are generally comparable to those given in the USP." If you need detailed information about contents, specifications, and required tests, you should probably directly compare the monographs in the FCC (Food Chemicals Codex) and the USP (United States Pharmacopeia). TenOfAllTrades(talk) 13:20, 18 May 2010 (UTC)

Thanks. I don't need very detailed information. The answer is good enough. --72.94.148.35 (talk) 04:39, 19 May 2010 (UTC)