Wikipedia:Reference desk/Archives/Science/2016 September 21

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.

September 21

Image noise

When set the ISO of a camera to a high setting, meaning the amount of light that's required to hit the sensor to achieve a tone is decreased, the image gets grainy. The same thing happens in your eye in the dark— things look grainy. Why is this happening? Is it a property of light, or a problem with all light sensors?

You may wish to read our article on the topic at Image noise. The problem can be due to noise in the sensor, or down to the photons of light that come in one at a time. Graeme Bartlett (talk) 00:30, 21 September 2016 (UTC)

Your eye does not have light sensors whose sentsitivity can be adjusted. It has separate sensors that work in dim light and bright light, called rods and cones respectively. Only the cones are sensitive to color, so in dim light you can't see colors much. The part of your eye with the greatest density of sensors, the fovea, contains only cones, so in dim light you see with other parts of your eye that do not give as good a resolution. Thus what's going on is completely different from effects in a camera, though it may produce similar results. --69.159.61.230 (talk) 05:16, 21 September 2016 (UTC)

Interesting that in the days of film, ISO was a measure of grain size. 400 ISO film was larger grain than 100 ISO film. --DHeyward (talk) 05:12, 22 September 2016 (UTC)

Well, it was a measure of film speed. The 400 film would have larger grains if it used the same chemistry as the 100 film.

A linguistic note: "ISO 400" is not actually the correct way to write the film speed; it's "ISO 400/27". It was previously called "ASA 400" or "DIN 27", and the ISO standard just combined the two numbers from the earlier ASA and DIN standards. The DIN standard used smaller numbers because it was logarithmic: ASA 200 was DIN 24, ASA 100 was DIN 21, etc. If you're just talking about 400, you might as well call it ASA. All three abbreviations, ISO, ASA, and DIN, are the abbreviated names, or former names, of the standards organizations that respectively promulgated these particular standards. Again, see film speed. --69.159.61.230 (talk) 06:13, 22 September 2016 (UTC)

Modern digital cameras report an "equivalent" ISO value because this is a parameter that consumer photographers wish to use. In actual fact, a digital camera's "ISO value" is a best-effort to configure the digital system (or to report its automatic configuration), distilling many complex and independent variables into one scalar number.

In some respects, a digital camera's ISO value represents the total gain applied to the image. In other respects, the ISO value reflects a model of the sensor and processor's signal to noise ratio in a fashion that is meant to correspond to photographic film.

In broad brush-strokes, we can say that using a higher ISO value will yield both a "brighter" image and a "noisier" image. When we compare two different images captured on two different digital cameras whose reported ISOs match, the intent is typically to make their exposure equivalent, and their signal-to-noise ratios will vary as a free parameter.

There are many problems with this simplified model. First, noise in digital cameras can be quite complex. Even defining the signal-to-noise-ratio is a challenge of both engineering and of basic optical science. Frequently, we just assume gaussian noise and let some sophisticated automatic mathematical system analyze and estimate noise for some ideal reference image.

Another problem is that gain in a digital camera is not simple, either. Even if we ignore the internal details - like the fact that gain can be applied many different times, or that different gain may be applied to different parts of the picture in various spatial-, frequency-, or color- tranform domains - all of which a film camera generally cannot do! - we still have the problem that gain need not be a scalar value either. Consider only the most obvious of these issues: the gamma equation tells you that your gain is non-linear. Well, here's some bad news for the enthusiasts who want to use their digital camera as a scientifically-accurate photon-to-electron- transducer: even the most elementary and archaic electronic circuits use automatic, non-linear application of gain. Your camera's image-sensor contains millions or billions of transistors. Each of these millions of independent circuits need not behave passively - let alone identically !

The image may experience gain at every single stage of its processing pipeline - from nonlinear optical attenuation, to nonlinear photoelectrical coupling, to nonlinear digital signal processing. From all these stages, one must estimate an effective, average gain. You're trying to determine which coefficient you use to multiply the input-photon count that will yield the 8-bit number in RGB color-space in your output picture. Well, ... that's an overly-simplified model!

ISO rating is a dramatically inapt way to represent the image that is produced by a digital camera. But, it's the value that people want, so a lot of computational effort goes into estimating it.

The ISO standard, ISO 12232:2006 publishes several reference algorithms to standardize this methodology... but it costs something like a hundred US dollars to buy their spec sheet, and who has that kind of money to burn on their digital camera hobby?

Nimur (talk) 15:58, 22 September 2016 (UTC)

Nimur has explained it, but even I as a photographer, had trouble following it ;¬) There must be a simpler way of explaining the signal to nose ratio and its connection with gamma compression which contributes to the graininess. Will have a stab at it, never the less. Light falling on each pixel represents 'information'. If no light from (say) the shadows on a moonless night are falling on the sensor, then the camera records mostly the sensor noise. Should the light level gets up-to (say) 254 bits in blazing sunshine, the noise (which is still there) is swamped by the extra information and is no longer perceivable, since the percentage of noise is a very small fraction, compared to the light signal. Or in other words, in blazing sunshine the light information is hitting the sensors faster than the 'noticeable' noise is being generated. Therefore if one set ones camera to a too higher ISO it will show grain because it has not been given enough time to gather light information. --Aspro (talk) 19:03, 23 September 2016 (UTC)

I apologize that my answer was not as clear as I had hoped.

Let me try to re-orient towards the original question.

The usual root-cause of "graininess" in a digital image comes from two places: shot noise and thermal noise. (There are many other noise sources with different physical sources, but we can ignore them for today's discussion).

When a digital camera raises its ISO value, it usually does so by boosting its gain. This does two things:

1. It permits the camera to collect a similarly-exposed image using a shorter exposure time. This reduces the total number of incident photons, and the inevitable and purely-mathematical consequence of that change is that the statistics of the photon shot noise are amplified in magnitude. For example, if we're collecting 100 photons and we get one single stray photon, we might say we have "1% noise;" compare this to a smaller photon-count: if we only expose for half as long, we expect to collect 50 photons - but now, one single stray photon is "2% noise." This is literally and actually quantum physics at play, creating a real, visible, observable effect inside a toy that the average person can play with! Just be careful with taking this simplification too far - it's harder to know which photon is the noise photon - which of the 51 identical particles was the one you didn't want to count?! So, you can see how the simplified explanation breaks down pretty quickly. In this example, we'll expose for half as long, so we get about half as many photons, but ...we're about to use "an amplifier" to multiply the count by two - and we should get an image that looks the same, even though the "shutter speed" was half as long!
2. It also boosts the gain, which may include analog gain. Every electrical circuit implementing an amplifier has both a gain and a non-zero noise figure. Therefore, thermal noise is added to the output signal, because the circuit is made of real devices that are made of real atoms and electrons with actual, non-absolute-zero temperatures. This process degrades the signal-to-noise ratio. In tandem with that additive process, all other noise that was already present in the signal - including the input signal's thermal noise and the photon shot noise - gets amplified by the gain (though this does not affect the ratio relative to the signal).

The end-result is that you see more noise, visually. The character of that noise is often called "graininess" because it looks very similar to the photochemical granules in old photographic film technology. Professionals who study digital image signal noise may prefer to use more specific or accurate terminology based on the mathematics/statistics/physics/engineering analysis of that noise. We care how gain got applied: what type of transistor, what type of circuit? Digital or analog? How hot was the device? How much of that heat is caused by inefficient photoconversion of incident-photons in the diode, and how much of that heat is static dissipation of the electronics around the photodiode? Details!

So, that's why the general rule of thumb is that high ISO yields both brighter pictures and more graininess. The rest of my discussion was intended to emphatically explain why ISO is a poor measure of this effect: there are so many additional confounding factors that make the details even messier - to the extent that I might go so far as to say the following: the rule of thumb does not apply equivalently to all digital cameras, and ISO is a terrible way to measure or characterize a digital image.

But, we use it anyway, because it's possibly the most universally-appreciated system we've got. Sometimes, a simple, one-parameter model is good enough for most people.

Nimur (talk) 19:52, 23 September 2016 (UTC)

Blood pressuer: white coat syndrome and its opposite

Hello. I understand that white coat hypertension is where a person's blood pressure is higher in a doctors surgery than it normally is in their daily life. In the UK the NHS gets around this by using home blood pressure monitors to get an accurate reading away from the surgery.

My question is. If blood pressure can be affected by mood and stress (ie. the stress of being in a surgery can make it go up) and made to look artificially higher than it really is, surely it can also be affected by relaxation and calm and made to look artificially lower than it really is. If a patient who panics around doctors takes their blood monitor home but then listens to calming music, closes their eyes, lies down for a bit, then does their reading... surely that's just as inaccurate/unrepresntative as their panicked reading at the surgery? — Preceding unsigned comment added by 213.120.143.245 (talk) 10:20, 21 September 2016 (UTC)

It's kind of funny; any time I go to see the doctor, the intake nurse will check my blood pressure, which will invariably be just over 140/90, even if I have been sitting comfortably for 5 minutes. Then the doctor will come in and say my blood pressure is a bit high, and will retake it, and it will have dropped 15 points to 125/75. μηδείς (talk) 18:04, 21 September 2016 (UTC)

The objective is to measure the patient's resting blood pressure. When the patient is at home, listening to calming music, closed eyes etc. that IS resting so the machine should achieve an accurate measurement of resting blood pressure. It isn't possible to observe a blood pressure significantly below resting blood pressure, unless the patient is using drugs or other medication for that specific purpose, and that would be unwise. Dolphin (t) 12:30, 21 September 2016 (UTC)

Off-topic discussion of one user's personal blood pressure experience. This is not an anecdote desk. SemanticMantis (talk) 13:50, 21 September 2016 (UTC)

Then there's the problem I have, that a blood pressure cuff constricting my arm makes me extremely uncomfortable and thus raises my BP, regardless of where I am or what music is playing. StuRat (talk) 13:14, 21 September 2016 (UTC) Maybe you could use a little self-psychology and pretend to like it? ←Baseball Bugs What's up, Doc? carrots→ 13:27, 21 September 2016 (UTC) It's not a mental issue, it's a physical one. I can't wear anything tight on my skin, as my skin then breaks down. I don't wear a watch or any jewelry. Before I switched to contact lenses, my glasses damaged the skin on my ears where they rested, creating open sores. StuRat (talk) 13:45, 21 September 2016 (UTC) I think you should take this to your doctor. ←Baseball Bugs What's up, Doc? carrots→ 15:09, 21 September 2016 (UTC) No need for that, I have the cure already, I just don't wear anything constricting (incidentally, this includes shoes, so I take them off while driving and as soon as I go inside). I do need an alternate way to take my BP, though. StuRat (talk) 17:24, 21 September 2016 (UTC) That's not a cure, it's a workaround or coping mechanism. ←Baseball Bugs What's up, Doc? carrots→ 19:04, 21 September 2016 (UTC) Life is all about coping. Cures are not well defined in general, e.g. the fact that we're fur-less apes makes us wear clothes. That's a coping mechanism, but few would want a cure in the sense of a drug that would give us back our lost fur. Count Iblis (talk) 19:32, 21 September 2016 (UTC) Reminds me of W.C. Fields' cure for insomnia: "Get plenty of sleep!" ←Baseball Bugs What's up, Doc? carrots→ 19:36, 21 September 2016 (UTC) Let them spend some time with the Tsimane people "Though the San, Tsimane, and Hadza often average less than seven hours of sleep, they seem to be getting enough sleep. They seldom nap, and they don’t have trouble dozing off. The San and Tsimane languages have no word for insomnia, and when researchers tried to explain it to them, “they still don’t seem to quite understand,” Siegel says." Count Iblis (talk) 20:04, 21 September 2016 (UTC) There are people who have a much more serious situation along this line (Epidermolysis bullosa) due to weakness of skin proteins. (I'm thinking anything like this will have a long tail of people with minor mutations who aren't affected enough to count as having the disease, but have weaker function than the average person) There's an article about someone imaging the lesion caused by some idiot putting a blood pressure cuff on them here but I didn't access the text. Apparently there is a "noninvasive blood pressure monitoring" scheme, involving direct pressure only, that can be used on them but in this paper even that was avoided - I think they just didn't take it. But a lot of other papers talk about putting wool padding under the cuff and then taking the pressure almost normally, which seems uninventive to me. Surely there is some bulk property of fluid that changes with pressure and could be measured remotely! Wnt (talk) 03:20, 22 September 2016 (UTC) Isn't there a finger cuff ? That would at least limit the discomfort, although I understand that they aren't particularly accurate. Fortunately I know when I have high BP, as I feel my own pulse in my ears (feel, not hear). High sodium intake does this to me, as I also have sodium sensitive hypertension. I have lots of health issues, but fortunately they all seem quite minor. But if I started taking meds for them all I would soon have major medical problems, due to all the side effects. This happened to my Dad, who took blood pressure meds, that caused kidney failure. (Even sitting causes me discomfort, so I am going to make a standing desk, right now.) StuRat (talk) 14:42, 22 September 2016 (UTC) Done ! I pulled a drawer out on my desk, put a wooden plank on the drawer, put a printer box on that, and my keyboard and mouse on top. I then put the monitor on top of my desktop PC, which is sitting upright on my desk, and now I have a standing desk. MacGyver has nothing on me ! :-) StuRat (talk) 14:58, 22 September 2016 (UTC)

"To measure your resting heart rate and blood pressure, pick a time when you’re feeling relaxed, Dr. Faulx advises. Randomly sampling both measures throughout the day can also help you reach an average." - From the Cleveland Clinic, here [1].

There isn't just one blood pressure, as indicated in my link above, it changes during each heart beat, let alone other factors. Resting, ambulatory, and mean blood pressure are all used in modern medicine. Here [2] is a paper that discusses a bit about resting vs. ambulatory blood pressure, and how they are measured. Here [3] is an article about 24 hour mean blood pressure, and here [4] is an article that uses resting blood pressure. SemanticMantis (talk) 13:50, 21 September 2016 (UTC)

If your arteries are in bad shape, they will be stiff and then a small amount of stress will lead to a significant rise in BP. Healthy arteries will expand a lot more in response to stress or exertion, so the blood pressure will actually not rise all that much. The same is true for your heart rate and this is then also affected a lot by your physical fitness. While your resting heart rate will be quite a bit higher if you are not physically fit, what also happens is that it will stay elevated due to recent minor exertions for lot a lot longer. But I can then get the opposite effect where the hart rate goes down so fast after a short exertion that it will undershoot the normal resting value and then climb upward to the resting value. E.g. I have measured values as low as 34 bpm just after a short exertion like running up stairs while my resting heart rate would typically be around 38 bpm. Count Iblis (talk) 17:40, 21 September 2016 (UTC)

Can cheese be made with unsaturated fat only ?

To start with, if you use skim milk to make cheese, does that work ? Of course, not having any fat would make it rather unappealing, so perhaps some vegetable oil with little or no saturated fat could be added back in. Keeping it from separating might require a treatment like they give Miracle Whip.

Or, perhaps fake processed cheese, made entirely from vegetable oil, could be made from all unsaturated oil ? I would expect that you might get a liquid cheese, but that could be put in a bottle and sold that way.

So, using any of these techniques, could you make a cheese, whether a solid bar, semi-soft cheese in a tub, or liquid in a squeeze bottle ? Is there any such product out there now ? (NOTE: I do not want to discuss the health effects of saturated fats, if you want to do that, please open a new Q.) StuRat (talk) 13:43, 21 September 2016 (UTC)

Yes, you can make cheese with skim milk. For example, here [5] are instructions for how to make ricotta using skim milk. Wikihow also [6] has some cheeses that can be made with skim milk. SemanticMantis (talk) 13:53, 21 September 2016 (UTC)

Good, then can we add in unsaturated oils so it has a better texture ? StuRat (talk) 14:39, 21 September 2016 (UTC)

That's a good question, and one I don't really know how to answer. Or rather, I do know how, but I can't do it for you. Experiment! The ingredients are fairly cheap, and only you are qualified to say when you think the texture is better. Another one to try might be paneer. It's incredibly easy to make, I've used wikihow instructions for that too. If you want more discussion before you dive in, I'd recommend you chat for a few days with the good folks at CheeseForum.org - they seem to have a lot of users with a great deal of cheesemaking experience for a variety of methods and cheeses. SemanticMantis (talk) 15:37, 21 September 2016 (UTC)

Vegan cheese exists, you could check the ingredients and processing to see if that meets your requirements. --Jayron32 14:10, 21 September 2016 (UTC)

I tried those, and didn't like them. StuRat (talk) 14:34, 21 September 2016 (UTC)

Not to harsh your vibes, but there's no such thing as a vegetable oil with absolutely no sat fats. Something like high-oleic sunflower will get you in the single digits, but it's still not zero. shoy (reactions) 15:13, 21 September 2016 (UTC)

Is there no way to separate the saturated from the unsaturated portions ? Even if not, we could still hopefully reduce cheddar cheese, from saturated fat content being 2/3 of the total fat, to under 1/3, using something like grapeseed oil. StuRat (talk) 15:34, 21 September 2016 (UTC)

Can saturated fats be removed from vegetable oil ?

...in order to make an entirely unsaturated oil, which is still edible ? StuRat (talk) 01:09, 22 September 2016 (UTC)

Removing all traces would be hard, but industrial and laboratory procedures exist for separating saturated and unsaturated oils, typically by exploiting differences in their melting point, e.g. [7][8]. Assuming you used a process that doesn't introduce inedible contaminants, I don't see any reason the separated oil wouldn't be edible, though the taste and cooking characteristics would be somewhat different. The only commercial cooking oil to be processed in a way that significantly reduces saturated fat that I know of is diacylglycerol oil which results in ~97% unsaturated fats (though removing saturated fat is not actually the immediate intent of the processing). I would suspect that the cost of processing to remove saturated fat is seen as too high in most cases to justify making such a product, especially since there are many cooking oils that are already ~90% unsaturated. Dragons flight (talk) 11:36, 22 September 2016 (UTC)

What kind of price increases would it take ? Considering how expensive some oils are, like quality olive oil, hopefully you could make an unsaturated oil less expensively than that, if you start with a cheaper oil, like grape seed oil, that's already low in saturated fat. StuRat (talk) 13:41, 22 September 2016 (UTC)

Assuming the process is fractional distillation based on melting point, and thinking about similar processed foods, my wild ass guess is that one would add somewhere between $5 and $30 to the cost of a 1 L of cooking oil. Dragons flight (talk) 14:10, 22 September 2016 (UTC)

That sounds like a reasonable price for a healthy vegetable oil. StuRat (talk) 02:13, 23 September 2016 (UTC)

Also, have to ask, to what end would one want to remove ALL saturated fats? --OuroborosCobra (talk) 11:41, 22 September 2016 (UTC)

One obvious reason is for nutritional studies. Feed one group of people food made with the normal oils (the control group) and feed another group the same food made with the oils with all the saturated fats removed, then measure if they have any significant differences in weight, blood chemistry, blood pressure, etc., after the trial period. Then, if the unsaturated oil is found to be healthier, you may have a marketable product. StuRat (talk) 13:38, 22 September 2016 (UTC)

Is saturated fat a risk factor for cardiovascular disease ?

(I broke this off as a separate Q, since, as I said in my Q, I'm not interested in discussing this now. StuRat (talk) 15:53, 21 September 2016 (UTC))

Whether saturated fat is a risk factor for cardiovascular disease is a contentious question:

Systematic review	Relationship between cardiovascular disease and saturated fatty acids (SFA)
Harcome, 2016	"The current available evidence found no significant difference in all-cause mortality or CHD mortality, resulting from the dietary fat interventions. RCT evidence currently available does not support the current dietary fat guidelines."
Ramsden, 2016	"Available evidence from randomized controlled trials shows that replacement of saturated fat with linoleic acid effectively lowers serum cholesterol but does not support the hypothesis that this translates to a lower risk of death from coronary heart disease or all causes."
de Souza, 2015	"Saturated fats are not associated with all cause mortality, CVD, CHD, ischemic stroke, or type 2 diabetes, but the evidence is heterogeneous with methodological limitations."
Schwab, 2014	"there was convincing evidence that partial replacement of SFA with PUFA [polyunsaturated fat] decreases the risk of CVD, especially in men."
Chowdhury, 2014	"Current evidence does not clearly support cardiovascular guidelines that encourage high consumption of polyunsaturated fatty acids and low consumption of total saturated fats.".

References

See Inuit diet, Maasai people#Diet, French paradox, Israeli paradox. --Guy Macon (talk) 15:50, 21 September 2016 (UTC)

In a society where everyone smokes, let's do studies where we compare smoking heavy cigars to lighter cigarettes. The results will be conflicting, as cigar smokers tend to smoke less than cigarette smokers. RCTs could be done but in the end you're comparing one bad thing to another bad thing but they cause problems in slightly different ways. Also, the results ending up not being not so clear will be totally distracting from the fact that smoking in general is bad. But no such studies would be done in a hypothetical society were everyone smokes, where not smoking is totally out of the question.

Now compare smoking to eating fat. Not eating any fat is "out of the question" as we all eat fat, we only ask if saturated fat causes problems relative to unsaturated fat. But let's now consider if fat in general is healthy or if it causes adverse health effects. Note that just like smoking is not necessary for health, there is no nutritional reason why you need fat in your diet at all apart from a few grams of Omega-3 and Omega-6 (which you can get from vegetables and who whole grains). So, let's see how the people who live in primitive society where no cooking oil at all is used fare w.r.t. people who live in societies where cooking oil is used. If you e.g. compare people in rural Uganda to people living in the West you find results like this:

"Maybe the Africans were just dying early of other diseases and so never lived long enough to get heart disease? No. In the video One in a Thousand: Ending the Heart Disease Epidemic, you can see the age-matched heart attack rates in Uganda versus St. Louis. Out of 632 autopsies in Uganda, only one myocardial infarction. Out of 632 Missourians—with the same age and gender distribution—there were 136 myocardial infarctions. More than 100 times the rate of our number one killer. In fact, researchers were so blown away that they decided to do another 800 autopsies in Uganda. Still, just that one small healed infarct (meaning it wasn’t even the cause of death) out of 1,427 patients. Less than one in a thousand, whereas in the U.S., it’s an epidemic."

The study on which this was based was one of the earliest papers in which the idea that cholesterol levels were linked to heart disease was put forward. But at that time this idea was not taken serious, it was decades later until this idea gained traction and today everyone knows about it. What happens is that if you stick to a plant based diet eating only small amounts of meet, don't use cooking oils and as a result eat a lot more (whole grain carbs) your cholesterol level becomes a lot lower.

But the most damning evidence against oils comes from studies that show that even olive oil is bad for heart health in an absolute sense (when you don't compare different Western style diets to each other but only look at the effect of the olive oil), see e.g. here. Now, one may look at other studies that suggest that polyphenol rich extra virgin olive oil seems to be not so bad or perhaps even good for the heart. But then the bigger picture here must surely be that if the olive oil itself is bad, then the polyphenols are just compensating for that, it doesn't make the oil good.

Intervention studies are i.m.o. best done in indigenous populations who still stick to eating a plant based diet low in oils. Take 100 such people and put half of them on a diet with a lot of olive oil and then see which group develops more cardiovascular disease. Such tests should be very sensitive to small effects because we already know that the control group is free of cardiovascular disease. Such studies have never been done but we do know that the zero heart disease rate in thee populations is not due to genetic factor. E.g. in this article where the focus is on salt consumption, it is pointed out that the Yanomamo Indians don't get heart disease despite their very stressful lives. But when they move away from the jungle to live in modern civilization they tend to drop dead from heart attacks and strokes just as everyone else there.

While one can fault such arguments as not being 100% rigorous, it should be clear that had we started out eating plant based diet in past centuries then the FDA would never have approved the use of any oils, salt and many other of the common food products we use today. The burden of proof would be on people claiming that oils can be safely used to prove that it doesn't lead to more heart attacks relative to a plant based diet. Count Iblis (talk) 18:44, 21 September 2016 (UTC)

Some dietary fat does seem to be necessary for life. See rabbit death. StuRat (talk) 21:07, 21 September 2016 (UTC)

Indeed. In fact, the claim that "the human body is designed to burn carbs, not fat" is unsupported by science and should not be given as an answer on the refdesk. Cut your protein intake to 0% and you die. Cut your fat intake to 0% and you die. Cut your carbohydrate intake to 0% and your body produces carbohydrates, allowing you to live normally. That being said, adding a multivitamin, non-digestible dietary fiber and Resistant starch is probably a good idea for anyone on a low carbohydrate diet. --Guy Macon (talk) 22:00, 21 September 2016 (UTC)

Yes, but the essential fatty acids that we cannot make ourselves are Omega-3 and Omega-6 fatty acids and you can get these from vegetables and whole grains. If you don't eat any fats, you'll need to eat quite lot of carbs so, it's then quite easy to get enough of the essential fatty acids. Count Iblis (talk) 21:56, 21 September 2016 (UTC)

Did you actually read our article on Rabbit starvation? You need fat to survive. You need a lot of fat to survive. It doesn't have to be animal fat, of course; millions of people are living healthy vegan lifestyles. They do need to take care that their diets have enough fat and protein, but that isn't hard to accomplish. --Guy Macon (talk) 22:33, 21 September 2016 (UTC)

Think this fear of saturated fats can be traced back to a sales and marketing ploy that was promoting Canola. They cherry picked research. Guy Macon gave a reference to the Inuit (think Inuit is the correct plural – if it isn't I am going to have lot of Inuits correcting me) . They had a very low CVD problem despite their high saturated fat consumption until they adopted 'healthy' southern diets. Think the researcher that has be most able to lay it out clear and explain the importance of a balanced fat diet is Udo Erasmus, who spent 6 years researching the nutritional and industry literature to properly understand the effects of oils and fats on human health. [9] His book provides a comprehensive bibliography that shows the broader view. In more resent years the sales and marketing lie has been exposed. For instance: The science of saturated fat: A big fat surprise about nutrition?. This is one of many. Don't think any budding nutritionist at college today would swallow (pun not intended) the saturated fat fallacy.--Aspro (talk) 18:53, 21 September 2016 (UTC)

See here, and also consider that it's quite difficult to get all your energy needs from fats. The human body is designed to burn carbs, not fat (diabetics can get Diabetic ketoacidosis where the body poisons itself due to burning fat when insulin levels are too low). If you put people on a low carb diet they develop insulin resistance, and they perform a lot worse on fitness tests. I know quite a lot of obese people and they all eat unhealthy fatty diets. When they see me eat (plate after plate of potatoes or whole grain pasta), they wonder how on Earth I can eat that much and be so thin. But that's only because they've been indoctrinated about the dangers of eating a lot, particularly carbs. They are eating foods that don't give them enough energy to be physically active. Thing is that there aren't all that many people who are obese and stick to eating a high calorie low fat, high carb diet with plenty of vegetables. Almost all obese people eat a junk food diet that actually isn't all that high in calories. Count Iblis (talk) 19:58, 21 September 2016 (UTC)

I do not believe that your claim that "If you put people on a low carb diet they develop insulin resistance" is supported by the science. See , which says " Reductions in dietary carbohydrate should be used as a strategy to treat insulin resistance." --Guy Macon (talk) 21:33, 21 September 2016 (UTC)

References

Yes, but then these people already have insulin resistance, and then it's well known that a low carb diet works well. In more recent work it has been shown that type 2 diabetes can be reversed using a low carb diet (at least some of the people could stop using all medications). But this does not mean that such a diet is optimal for healthy people who don't have insulin resistance or diabetes to begin with. Note here what the risk factors are to get type 2 diabetes, this involves eating an unhealthy diet that's completely the opposite of what I've been writing about above. So, in these studies we're studying primarily people who have done a lot of damage to their bodies for quite some time by eating a bad diet.

What we need instead is a rigorous test with only healthy fit people to see if more or less oil in the context of a super healthy diet is better (healthy = only whole grains, large amounts of vegetables and small amounts of meat). What I've read is that when you measure insulin resistance of healthy people who you put on an extreme low carb diet, you get worse outcomes, but I admit that such tests are not so rigorous, because it's not a natural diet that they could keep on eating permanently and it was also a small study. For me the difference between what I eat and what the average person eats and how I exercise and what the average person physical activity level is, is just too big to take studies done on the average person serious (just consider that I get 50 grams of fibers per day while the average person gets just 15 grams). Count Iblis (talk) 22:45, 21 September 2016 (UTC)

As for the Inuit, they have been isolated long enough that they may have developed the ability to thrive on a diet that would kill the rest of us (and heart disease does appear to have a significant genetic component). They exhibit many other specializations for their environment, such as the ability to withstand lower temperatures, so this would not be surprising. StuRat (talk) 21:12, 21 September 2016 (UTC)

That theory was first refuted in 1928. See Vilhjalmur Stefansson#Low-carbohydrate diet of meat and fish and No-carbohydrate diet#Research on effects of no-carbohydrate diet. --Guy Macon (talk) 21:47, 21 September 2016 (UTC)

I don't see anything in those studies that disproves the idea that the Inuit have adapted to their diet. As for Europeans eating the same diet, those were short-term studies, and cardiovascular disease only shows up after decades, so they aren't relevant there, either. StuRat (talk) 21:57, 21 September 2016 (UTC)

Steffansson followed the extreme low-carb diet for the rest of his adult life, remaining in good health until he died at the age of eighty two. Besides, the burden of proof is on the person making the claim. Where is the science behind the theory that saturated fat is unhealthy? Where is the science showing that inuuit have such an adaptation? In researching this, you might start with Seven Countries Study#Contemporary criticism and Seven Countries Study#Debate since 2000. --Guy Macon (talk) 22:21, 21 September 2016 (UTC)

Regarding the burden of proof and the Inuit, if we had the lone statement that "Inuit's are more able to cope with an all meat, high fat, low carb diet than non-Inuits", then I would agree. However, when you start with the assumption that "Inuits have an all meat, high fat, low carb diet, so non-Inuits can do well on that diet, too", then I feel the burden of proof shifts to whoever makes such a assumption to establish that Inuits and non-Inuits do, in fact, have the same dietary requirements. This is especially true in view of the fact that such prolonged isolation in a radically different environment has led to other obvious adaptations (for example, Inuits tend to have a lower surface-area-to-volume ratio, by having an endomorph somatotype, to limit heat loss).

As for his experiment on himself, a sample size of one is of no statistical value. StuRat (talk) 13:49, 22 September 2016 (UTC)

Freezing, Triple, and Boiling Points of Water

Is it a simple coincidence that the triple and boiling points of water are ⁠1/100⁠ and 100 °C respectively, or is there something deeper behind it ? — 79.118.172.77 (talk) 21:34, 21 September 2016 (UTC)

Celsius, Kelvin, Farenheit. The bit you're talking about only happens in the scale that's designed explicitly to peg the boiling point of water at 100. Would you be fascinated by the fact that water boils at 212°F and has a triple point at 32.01°F? SemanticMantis (talk) 21:47, 21 September 2016 (UTC)

The fact that g ≈ π² m/s² is not a complete coincidence. I am asking if something similar might not be at play here as well with regards to the degree Celsius. — 79.118.172.77 (talk) 22:27, 21 September 2016 (UTC)

The Celsius scale being pegged to water is a thing of purely human convenience and has no bearing on any greater mystery of the universe. As already linked to, you can read about the history of the Celsius scale. Since the trick only works in the Celsius scale, and does NOT work in any other scale, it's plainly obvious it is a coincidental artifact of the Celsius scale; which itself is a scale which has no real connection to the actual physical definition of temperature, and was arbitrarily set up merely to be convenient for people to use. --Jayron32 22:41, 21 September 2016 (UTC)

(edit conflict)In what sense is the approximate value of g not a complete coincidence? The triple point of water is marginally more than a hundredth of a degree above the freezing point that was supposed to be fixed at zero, so I can see no possible "deeper meaning" behind your ⁠1/100⁠. The 100 degrees boiling point was deliberately fixed at a nice round number, of course. Dbfirs 22:45, 21 September 2016 (UTC)

The meter was originally conceived as the length of a pendulum with a half-period of one second. (Later, it was redefined as the forty-millionth part of a terrestrial meridian). Since T ≈ 2π · (L / g)^1/2 , a choice of T = L = 1 and g = π² is the most natural. — 79.118.172.77 (talk) 01:04, 22 September 2016 (UTC)

This New Scientist article reports that the boiling point of water is subject to variables including how quickly it's heated, the shape and nature of the container, and the presence of dissolved solids, including air. 18th-century Swiss geologist and meteorologist Jean-André Deluc apparently carried around a flask of air for about four weeks in order to eliminate dissolved air, via shaking. He then recorded a BP for the air-free water of 112 °C. But the article says nothing about the shape or substance of the container he used to heat the air-free water, nor how quickly he heated it. Which prompts me to ask what is the international standard for determining the BP of water, and what is the relevant international body?

See Vienna Standard Mean Ocean Water. Smurrayinchester 08:19, 22 September 2016 (UTC)

Is there really much truth to this notion of a "pendular metre" though? AIUI, Newton thought of this as a combination of his gravity work and Huygens pendulum work, along with the beginnings of an interest in standards-based metrology. So $g\approx pi^{2}$ is certainly more than coincidence.

Yet this was "Newton's pendular yard", not any precursor to the metre. AIUI it had no influence on ideas for how to define the metre and the pendular concept didn't take off either in France, or in the century after Newton.

Both the yard and metre are of course approximately sized on an arm's length, because that's the human-scaled length unit that's convenient for measuring cloth to clothe that arm, or timber to make houses for them to live in. One might even, after Pennycuick, claim that seconds are the length they are as that too is a human-convenient interval, similar to breathing and the pace of walking. Andy Dingley (talk) 11:36, 22 September 2016 (UTC)

I can't see 60 bpm (once per second) as a walking speed, personally. If I wanted a human-scale time unit, that would be possible as a heartbeat or an average walking pace, I think 76 bpm would be more suitable (even then, it is a slow walk). Double sharp (talk) 15:36, 23 September 2016 (UTC)

30 bpm. Andy Dingley (talk) 15:51, 23 September 2016 (UTC)

Once every two seconds seems even worse... Double sharp (talk) 15:54, 23 September 2016 (UTC)

Yes, fascinated, because Fahrenheit used an arbitrary scale too. He just put 180 degrees between his fixed points, not 100 (not too weird, pre-metrication). But why 32°F? The fascinating part is that he started out with a totally different scale, sensibly beginning at 0°F. The other points were sensibly picked at 32°F (a nice power of two) and 96°F (blood temperature, and a simple three times). Then, he found out that his first measurements weren't terribly accurate. So he had to redefine the scale and its definition, but without changing it. By picking some new fixed points, and the only slightly awkward 32 and 212, he managed to do so.

A similar thing has now happened with the standard kilogram. Rather than the old measure of an arbitrary-sized lump of metal, carefully looked after and yet still shedding atoms, we've now redefined it to a more fundamentally-based standard using crystal structure and other existing standards of length. Yet this has managed to be done without changing anything, by finding another fixed-point value that was within the previous margin of error. Andy Dingley (talk) 11:09, 22 September 2016 (UTC)

No, we haven't. This is planned to happen, but not until 2018. --69.159.61.230 (talk) 06:00, 23 September 2016 (UTC)

2nd postulate

If the second postulate is true then why moving observer with constant velocity / speed notices other moving observers (with constant velocities / speed) with newtonian relativity.

Let we have, stationary observer on an asteroid: OA, astronaut in space ship A = AA and astronaut in space ship B = BB

All observers are at the same location on asteroid initially. Two space ships A and B are launched with speed v1 (say 0.6c) and v2 (0.5c) from an asteroid at time t1 and t2 in the same direction. There is a noticeable distance between the space ships. A pulse is also emitted by OA with velocity “c” in the direction of space ships at time t3. OA can see a pulse, A and B due to their elevation difference. After sometime,

OA observes that a pulse is travelling at “c” ; d = ct

AA observes that a pulse is also traveling at “c” ; da = ct’ Experienced time dilation

BB also sees pulse traveling at “c” ; db = ct" Experienced time dilation

A pulse , A and B are in motion (relative to asteroid e.g.) but BB can observe a pulse is travelling at “c”like OA while the same BB can't observe the v1 of A like OA. Shouldn’t BB observe A is also moving with v1 instead of 0.1c if the second postulate is true for all observers? Can somebody explain why special relativity for “c” and classical physics for speed of A when observed by BB. — Preceding unsigned comment added by 2001:56A:739C:6D00:852F:60EA:A53E:C483 (talk) 22:38, 21 September 2016 (UTC)

Your question seems exceedingly convoluted and hard to follow. Are you basically asking why does an astronaut in a space ship not behave like photons? It seems a bit disingenuous. The 2nd postulate applies to the speed of light in empty space, not "anything going really fast", it's a property of light not a property of "the speed". Vespine (talk) 01:19, 22 September 2016 (UTC)

This isn't strictly true. The Lorentz transformation establishes the invariant velocity c, and light happens to travel at this speed because photons have no rest mass. c in the equations is a property of space (or the universe, if you prefer), not a property of light specifically. Tevildo (talk) 12:33, 24 September 2016 (UTC)

Can't really follow the question. A "pulse" of what? Also, many of the "paradoxes" are solved by observing the acceleration of your 2 observers are different. --DHeyward (talk) 05:07, 22 September 2016 (UTC)

See Lorentz transformation and Velocity-addition formula. Someguy1221 (talk) 05:11, 22 September 2016 (UTC)

I mean a pulse of light, Ok

Try this way

Let a second pulse of light is fired from another asteroid towards the aforesaid asteroid. The speed of this pulse is also the same for all observers OA, AA and BB. Photon doesn't experience time but from the second pulse (if it can feel) prospective (frame of ref) which observer would hit second pulse first - or - all observers would hit the second pulse of light at the same time. — Preceding unsigned comment added by 162.157.217.155 (talk) 21:38, 22 September 2016 (UTC)

Although photon can't observe anything but it has frequency therefore its life span can be figured out from the # of cycle between the emission and first absorption — Preceding unsigned comment added by 162.157.217.155 (talk) 02:58, 23 September 2016 (UTC)