Miscellaneous desk
< August 14	<< Jul \| August \| Sep >>	August 16 >

Welcome to the Wikipedia Miscellaneous 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.

August 15

Citation using Chicago Referencing

How do you:

1. Cite a powerpoint presentation? It was given during a university lecture and my tutor seems to want me to include it in my assignment, but I have no idea how. 2. Cite a work that has been quoted in a different work? I would just include the entire book but it was written in German and it would be somewhat obvious that I hadn't actually read it.

Thanks.

Sarah —Preceding unsigned comment added by 86.53.104.14 (talk) 01:51, 15 August 2010 (UTC)[reply]

For 1, if I were citing this for publication, I would cite it as a "personal communication", and possibly add a note clarifying the origin. For 2, I would cite the original and add a "quoted by" note pointing to the source that I actually got it from. In my experience, though, it's pretty common to just cite the original in cases like that. But for a class, I would play it safe and add a note. Looie496 (talk) 02:02, 15 August 2010 (UTC)[reply]

On #1: I would just cite it like a lecture. See here. That is what my students have done for me in the past, anyway. I would not cite it as a "personal communication." In any case, whatever you do, just be consistent, and give as much information as is necessary for the reader to understand what it is you are trying to cite, in a general Chicago-like appearance, and it will probably be fine. Make a good-faith effort but don't spend all day worrying about it; your teacher will probably not have any idea how it should be cited either, as it is not a common thing to cite. Citation, in general, should be taken seriously, but not at the detriment of the substance!

On #2: If it was translated I would use "Quoted in <source with the translated quote>." I have read some places where they very much want you to cite the original (e.g. the German) and then say "quoted in," because it would then keep you from having to track down the intermediary source, I suppose. I have always thought that to be a little silly unless you have yourself tracked down the original and confirmed it to be exactly what they said it was. In any case, you really must cite the source that you are getting the translated quote from. At the very least, because as you say, it will look suspicious if you don't. I had a student who plagiarized quite a bit once and the really key giveaway (it was otherwise a pretty good essay) was that he cited lots of sources that I knew he really could not have possibly gotten his hands on (because they were not only in a language which I was fairly sure he didn't know, but were probably available only in obscure archives in countries I was fairly sure he hadn't gone to), which prompted me to start running those quotes through Google, which revealed that he had just paraphrased another book entirely without citing it. Anyway, you are obviously not doing that, but it will stick out to a close reader for the reason you'd mentioned. --Mr.98 (talk) 03:16, 15 August 2010 (UTC)[reply]

On #1, if the lecturer was kind enough to put the powerpoint files up in hir personal area of your university's webserver, the you might be able to cite the URL. CS Miller (talk) 12:17, 15 August 2010 (UTC)[reply]

Script/antiqua typeface

Does anyone know of a typeface that is script (i.e. looks like handwriting) but isn't as curly as Brush Script? I guess I want something a bit like an antiqua typeface but a bit more like handwriting. Yaris678 (talk) 07:33, 15 August 2010 (UTC)[reply]

The Lucida group of typefaces contains my favourites: Handwriting and Calligraphy.--TammyMoet (talk) 08:52, 15 August 2010 (UTC)[reply]

We also have a collection under samples of script typefaces. Maybe you can spot one there. ---Sluzzelin talk 14:25, 15 August 2010 (UTC)[reply]

Thanks guys. Lucida Calligraphy seems to be what I am after. Yaris678 (talk) 18:14, 15 August 2010 (UTC)[reply]

Kiddies' rhyme

Can anybody help with the text (and variations?) of a rhyme that starts something like <One two three, mother caught a flea, put in the teapot and made a cup of tea>?Froggie34 (talk) 07:35, 15 August 2010 (UTC)[reply]

One, two, three, Mother caught a flea,

Put it in a teapot to make a pot of tea.

The flea jumped out! Mother gave a shout!

In came Father with his shirt hanging out![1] or

Father came in with his shirt hanging out![2] Clarityfiend (talk) 08:12, 15 August 2010 (UTC)[reply]

Hyde Park / Kensington Gardens Border?

Where is the exact border between Hyde Park and Kensington Gardens in London? —Preceding unsigned comment added by 188.220.46.47 (talk) 13:19, 15 August 2010 (UTC)[reply]

The official website[3] isn't much help. It does say that the Gardens were enclosed during the 18th Century and the map posted by Cuddlyable3 seems to show that. There is no boundary fence now. I seem to remember that Queen Caroline of Ansbach (maybe?) once asked how much it would cost to enclose the whole of Hyde Park to keep out the public and was told "a couple of crowns". Alansplodge (talk) 14:51, 15 August 2010 (UTC)[reply]

A lot of current maps, including official Royal Parks ones, don't mark the boundary on. But several modern maps (e.g. [4]), using unstated evidence, continue to mark West Carriage Drive as the boundary, not the Serpentine. Warofdreams talk 16:50, 15 August 2010 (UTC)[reply]

That's very similar to the 1833 map. Alansplodge (talk) 17:14, 16 August 2010 (UTC)[reply]

The placing of the "H" of "HYDE PARK" on this map seems to identify all the land east of the Serpentine. Cuddlyable3 (talk) 19:06, 15 August 2010 (UTC)[reply]

Pokémon probability

How come the probability of an attack hitting is inversely proportional to how important it is that the attack hits? --70.134.48.188 (talk) 14:42, 15 August 2010 (UTC)[reply]

My guess is that it makes the game more exciting. If the reverse were the case then you'd almost always succeed when it mattered most and only have a serious chance of failing when it didn't matter at all. The game would be very unexciting because all of the 'drama' would have been sucked out of it! Game designers spend a lot of time thinking about things like that - they even plot graphs of player progress of stress versus reward and so forth. There is quite a bit of science to it. Making a game be both 'balanced' and 'fun' is rather tricky. SteveBaker (talk) 16:17, 15 August 2010 (UTC)[reply]

It isn't. It's confirmation bias. You don't notice all the times when the important strike hits because it's non-notable, but you pay attention every time that important strike misses and place more importance on them. That said, as the battle wears on it's more likely that you will, at some point, have been hit with an accuracy-reducing move.

If it still really bothers you, have your Pokémon use Faint Attack, Aerial Ace, Shadow Punch, Magnet Bomb, Swift, Magical Leaf, or Shock Wave. They all hit without fail...with a few rare exceptions. Vimescarrot (talk) 16:23, 15 August 2010 (UTC)[reply]

The random elements of games frequently appear unfair to users. In Battle for Wesnoth, this problem was apparently so severe that the developers made the game keep detailed statistics (accessible at any time through the menus) so that the user can see the overall behavior and trust that the "dice" weren't loaded. Paul (Stansifer) 21:56, 15 August 2010 (UTC)[reply]

Will that explain why my rivals Magikarp can use splash 68 times when the move only comes with 40 power points? ^Avic _ennasis @ 16:28, 6 Elul 5770 / 16 August 2010 (UTC)

That's because the computer is a cheating bastard. 86.164.66.83 (talk) 00:48, 17 August 2010 (UTC)[reply]

(Confession: In real life I am actually a computer game programmer.)

Well, yes, sometimes the AI's cheat a bit...but bear in mind that without cheating the AI's would fire 17 milliseconds after seeing you - no matter how subtly you hide or sneak - and with zero network lag. They would always get off the first shot. They would fire at the maximum rate their weapon will cycle and every single shot would be 100% perfectly on target - at maximum range, and even if both you and they are running flat out at the time.

To make the game even slightly playable, game programmers have to "cheat" by slowing down the AI's, by making them die from practically the first wound. The player always survives lots of hits without so much as slowing him down and has magical curative 'health packs' scattered around which the AI's are somehow careless enough to leave lying around their secure hideout...yet are stupid enough never to use themselves! We also make the AI's wait before they shoot - and shoot deliberately wildly.

So, (on occasions) we have to give them a helping hand ("cheat") the other way to make up for the deficiencies of them being a mere thousand lines of software running on a fairly cruddy computer - rather than being as smart as an actual, live human. There is no way that game AI's are ever going to be as smart as the player (or as smart as...say...a cockroach) - the art is to make use of the things an AI has got going for it - to cover for it's extreme stupidity. That means there is some compromise of reality. The art is to make the (inevitable) cheating look as unobvious as possible...and that's the tough part.

If you take the example of a car racing game. It would be the easiest thing in the world to make the AI's drive a perfect race - taking the perfect racing line, doing everything 100% right. You'd never beat them. So you have to dumb them down. But if you just make them perform at (let's say) 105% of the performance of an average player - then average players will almost never win a race...FRUSTRATING! If we make them perform at 95% of the performance of an average player - then average players will win about the right amount of the time - and they'll have fun. But the problem is that when someone first picks up the game, they won't be playing at average skill level. They'll spend the first 10 hours of game play getting frustrated - then they'll have a few hours of fun as they just start beating the AI's...then they'll get a little better and win 100% of the time which is BORING.

So the way to make the game challenging is to have the AI's adjust their skill level to yours. To keep themselves always close to your skills. That's easy enough to program - but it's still not enough to keep the excitement going.

In most real car races, if you spin out or crash - you're out of the race...game over. But game players can't stand to play games that are that realistic - nobody but the most hard-core would buy racing games where any crash ended the race. So we have to cheat and let the player carry on racing after a crash - with just a few seconds of penalty to make them try to avoid doing it. The trouble is that if the AI's are racing at close to the player's skill level, they'll be so far away after a few seconds of delay that you'll be racing alone through empty track for a long time...and THAT is BORING. So we cheat some more...when you crash, the other drivers slow down a bit to give you a chance to catch up and to rescue the game experience.

It works the other way too: if you get too far ahead, you are again racing on an empty track - which in the real world is great news! But in a game, it takes all the fun out of it. Where is the challenge? So when you get way ahead, we may have the AI's speed up some to get closer and keep the pressure on you. Having gotten ahead of the pack, you can't build up a massive lead and relax - because the AI's are always right there behind you.

Of course if that's all we did, it wouldn't matter whether you drove well or badly, the outcome of every race would be more or less random...and if anyone ever realized that's what we were doing, it would also be a disaster! So the trick (which is sometimes done well - and sometimes completely screwed up) is to make it so that the player wins races by doing better than he did the last time he played. You always feel like you're under pressure to get ahead or to keep your lead - and that pressure makes it fun. Even if you aren't a great race driver, the AI programming can make you feel like you are - which is what makes the game fun. In the end, you are really racing against yourself...which is automatically a challenge, no matter your actual skill level.

It's a common view that game programmers are cheating because they want the AI's to win...but that's ridiculous. The objective is sell more games - and to do that, we need to give the player the most exciting experience we possibly can...and sometime we...erm...cheat a bit...to make it more exciting. SteveBaker (talk) 03:39, 17 August 2010 (UTC)[reply]

Wow! Thank you for that informative reply! It's a real eye opener and really helps to explain so much of my gaming life! Really, awesome info, thanks! ^Avic _ennasis @ 07:20, 7 Elul 5770 / 17 August 2010 (UTC)

How can I make my rice stick togther

I really like sushi and occasionally like to try to make it at home. I have lots of recipie books but usually follow the instructions on the back of the pack of sushi rice I buy from my local supermarket here in the UK. The problem appears to be, the rice has a huge preference for sticking to everything except itself. I end up with rice stuck to the pan, the utensils, my hands and anything else it touches. I really struggle to form the rice into the oblong shapes because each movement of my hand pulls it apart again. If I wet my hand with vinegared water - the usual advice in my recipie books - the rice doesn't stick to my hands but it still won't stick to itself; and if I do get somthing of a vaguely acceptable shape and size, it readily falls apart again when I go to eat it. So, how can I make my rice stick togther? Astronaut (talk) 15:15, 15 August 2010 (UTC)[reply]

Hard to know for sure what the problem is without observing you in action, but two things that occur to me are (1) it's easier if the rice is cold before you try to shape it, and (2) even with the vinegar water you have to keep your hands rigorously clean -- as soon as they get starchy you need to clean them before continuing to work. Looie496 (talk) 17:06, 15 August 2010 (UTC)[reply]

try adding a little more water and/or turning down the heat when you cook the rice. in my experience, the more liquid the rice absorbs, the stickier it is. just don't add so much water that the rice turns into a paste. --Ludwigs2 18:30, 15 August 2010 (UTC)[reply]

If you stir the rice a bit during cooking it will make it stickier. Don't do it too much or you'll get something like rice pudding! Yaris678 (talk) 20:24, 15 August 2010 (UTC)[reply]

This might sound like a stupid question, but are you cooking the rice correctly? The best way to cook sushi rice is in a (preferably Japanese) rice cooker - considering the sheer number of Japanese people in the UK, these should be available in any major kitchen appliances store. If you don't have one, you can cook it in a regular pot, but it takes a little extra work - the lid must be perfectly closed (possibly weighed down as well) so that no steam escapes - the rice should effectively be steamed, rather than cooked. Lately special little plastic dishes for steaming rice in a microwave oven have become available, too - possibly you can get those in a store as well, but I wouldn't know. TomorrowTime (talk) 21:01, 15 August 2010 (UTC)[reply]

Thanks. I thought maybe I wasn't cooking the rice properly. Maybe too much water is escaping as steam - certainly the lid of the pot I use is not very tight. The instructions I use simply say put the lid on the pot, simmer for 10 minutes, then leave the pot to cool for 20 minutes without removing the lid. Today, I left the rice to cool for considerably longer (nearly an hour) before emptying the rice into a shallow dish and adding the sushi seasoning. I still got rice that stuck to everything but itself (it tasted good though). I had previously looked into buying a rice cooker, but I thought the typical 2+ litre capacity was much larger than I would need and they were quite expensive for an occasional use item. However, looking again just now, I can see Amazon has a few mini rice cookers at much more reasonable prices. Any hints on what I should look for (or avoid) in mini rice cookers? Astronaut (talk) 00:23, 16 August 2010 (UTC)[reply]

My wife says that you're probably using the wrong kind of rice. Sadly, she went to bed without elaborating on that thought. SteveBaker (talk) 03:30, 16 August 2010 (UTC)[reply]

You know, that was my first instinctive thought when I first saw the title of the question, but the rice seems to be OK, if Astronaut is using the rice on the JPG in his first post - it's an export from Japan and is supposed to be sushi rice. Steve, what your wife had in mind is that there are basically two broad types of rice, long corn and short corn, and here in Europe we use almost exclusively long corn, while in Asia it's short corn. The difference between these two is that long corn doesn't get nearly as sticky as short corn. TomorrowTime (talk) 08:28, 16 August 2010 (UTC)[reply]

While botanically rice is often divided into indica and japonica ([5]), commercially there are far more then two types of rice, even most simple divisions usually divide rice into long grain, medium grain and short grain. Note that the more then two types thing is important in a number of ways, for example there are glutinous varieties of rice (often called sticky rice) which are used in some Chinese dishes as well as SEA, South Korean and Japanese ones but these aren't used for Sushi.

Also it isn't true short grain is exclusively or even I would argue predominantly used in Asia, Jasmine rice and Basmati are well known long grain aromatic varieties that originate in and are commonly used in parts of Asia. (Although medium grain varietes of rice are popular in India too.) In fact my impression is long grain varieties are fairly popular in China as well. Medium grain is generally preferred in Japan and South Korea however. [6] says something similar and makes a distinction between Southern (preferring long grain) and Northern China (preferring medium grain). BTW I'm not entirely sure about the European division either. Arborio for example is a short grain variety. (I know medium grain varieties are quite common in NZ, probably partially because Australia produces a lot of that.)

P.S. Further research suggests that japonica derived varieties are generally more suited to more temperate area and indica varieties generally more suited to tropical area, hence tropical areas of Asia do actually tend to use longer grains of rice more. This supports my belief, which I didn't mentioned earlier because I wasn't able to find any refs, that longer grained varieties are common in SEA. [7] suggests the split is 70 indica-30 japonica in China. To state the obvious, with significant proportion of China's rice longish grain, probably something similar for India and seemingly much of SEA (although glutinous rice appears to be popular in some parts of that particularly Laos) it's probably more accurate to say longish grains predominate in Asia then shortish grain varieties.

In terms of the OP, since it's apparent they are using a sushi variety of rice I would recommend what others have mentioned, try adjusting the amount of water and the cooking equipment.

Nil Einne (talk) 13:56, 16 August 2010 (UTC)[reply]

It seems to me that you are using the right rice. Looking at the image of the pack, your rice is 無洗米. So you don't have to wash it. Just rinse it and put the rice in the pan and add water. Be sure to use soft water. Your water might be too hard. As for the water ratio, see Japanese rice#Preparation. If you have dried kombu, cut it about 5x3 cm and put it in the pan for umami. Put the lid on the pan and wait for at least 30 minutes. Then cook rice with the strongest heat. When the water is boiled and bubbles come out between the lid and the pan, turn the heat down to the lowest and cook 15 minutes. Turn off the heat and wait for another 15 minutes. Now you can open the lid. Mix sushi vinegar with rice on a tray or in a hangiri while the rice is hot. Oda Mari (talk) 16:16, 16 August 2010 (UTC)[reply]

Rather than installing a (very expensive) filter in my water pipes, would a filter jug like this one make my very hard water soft enough? Thanks for the instructions. I particularly like the idea of waiting half an hour before turning on the heat - I might give that a go and I could use 2 rings on my electric hob to simulate an immediate turning the heat down to the lowest (it takes a while to reduce the heat with electric hobs). Astronaut (talk) 09:15, 17 August 2010 (UTC)[reply]

Why don't you buy a bottled soft water like this one? Of course the still one. Is it difficult to find any at your local stores? It is OK to wait half an hour before cooking. I didn't know you use an electric stove! I only cook with gas. Mmm....I cannot assure you my way of cooking rice would work well with an electric stove. Well, as I don't know how low the lowest of your stove, listen to the sound of the pot. When the bubbling sound of the rice stopped or if you smell something burning, turn off the heat, even if it's before 15 minutes. I think using the rings is a good idea. As for sushi vinegar, please use Japanese rice vinegar. Oda Mari (talk) 16:24, 17 August 2010 (UTC)[reply]

I have a general purpose steamer like this which I've used to cook sushi rice effectively. You may find it cheaper and more flexible than a specific rice steamer. --Frumpo (talk) 09:11, 18 August 2010 (UTC)[reply]

Odometers vs Sat Nav Speed Indicators.

I guess the Subject line will alert the intelligentsia here to the basis of my query, but for the benefit of the less informed (including myself) and wider Wiki audience, please permit me to explain. I bought a new Suzuki car in March this year and am always careful to observe statutory UK speed limits. But I also always use a TomTom SatNav device so as not to get lost, but also, to advise me when I exceed chosen speed limits such as 30 mph or 70 mph. But I have begun to notice that whilst the Suzuki Odometer says, say, 70 mph, the Tom Tom SatNav says 67 mph, and other SatNav values are always lower by about 10% in favour of the SatNav reading against the Odometer. THIS IS NOT A REQUEST FOR LEGAL ADVICE. But I am interested to know which device is likely to be the most reliable? And as an afterthought, what reading will the UK Traffic Police believe? 92.30.101.74 (talk) 20:42, 15 August 2010 (UTC)[reply]

Don't know about the UK police, but here's some general thoughts. Maybe the car maker winds the odometer to show slightly more than your actual speed to keep you in the safety margin. Maybe the odometer is functioning slightly off - my last car would show as much as 10 km/h over the actual speed I was making on regular roads, and even more on highways - but it was an old car so maybe that can be expected. If it troubles you, maybe you could call your local Suzuki dealership and inquire. Lastly, the police is going to measure your speed with their own equipment, so how much your odometer shows is really not that important in that regard, IMO. TomorrowTime (talk) 20:53, 15 August 2010 (UTC)[reply]

(edit conflict) :I've noticed the same discrepancy (confirmed using mile markers on motorways before I purchased a satnav) on several cars of different makes. I assume that manufacturers set their odometers to read about 3 mph "fast" so that they will not be liable if you are fined for exceeding the limit, and so that the dial will not read too ~~high~~ low if different tyres are fitted. I would expect the discrepancy to increase as the tyres wear, but I haven't checked this. Dbfirs 21:03, 15 August 2010 (UTC)[reply]

I see the same thing, with TomTom vs a Mazda's ODO. I seems that the discrepancy grows with speed, leading me to believe the car's odometer shows about 92% of the speed the gps does. Note that, as a driver is legally responsible for operating a safe vehicle, telling a policeman that your odometer is defective will either fall on deaf ears or result on him issuing you a "fixit" notice (a "vehicle defect rectification notice"), which means you need to get that aspect of the vehicle tested (generally by an MOT tester) and then show a certificate to that effect at a police station. If the vehicle is new, or has passed the MOT, then it almost certainly isn't defective, so telling the policeman it is defective is daft. -- Finlay McWalter • Talk 21:36, 15 August 2010 (UTC)[reply]

I think I saw a story on TV before about a popular car manufacturer taking advantage of the allowed tolerances in odometers. If I remember it right, for years, some of the models from the manufacturer would have odometers that ran too fast but within legal limits. A curious thing was that the model(s) that had the problem would changed from one year to the next. Supposedly the manufacturer benefited from the fast odometers because warranties would run out faster too. --173.49.81.80 (talk) 01:08, 16 August 2010 (UTC)[reply]

The article Speedometer#United Kingdom has info about tolerances of inaccuracies. ny156uk (talk) 21:37, 15 August 2010 (UTC)[reply]

An odometer measures distance travelled. A speedometer measures speed. Apart from that pedantic point, you might want to check your tire pressures. A speedometer works by measuring how many rotations your wheels do every second and multiply that by the circumference of the wheel to get the speed. If the circumference of the wheel changes, the speedometer will be wrong. The most common thing that can change wheel circumference is a change in tire pressure. That said, I would expect your tire pressures to be low, if anything, which ought to make the speedometer give a lower than actual value, rather than the higher value you are reporting. As for the police - they don't care what your speedometer or your sat nav say. They measure your speed with a laser and if that says you are speeding, then they'll give you a ticket. --Tango (talk) 01:27, 16 August 2010 (UTC)[reply]

The two meters are almost always supplied as one unit with one feed, of course. I think Finlay McWalter's Mazda must have had oversize wheels fitted if the speedo read low (which is illegal in the UK). Tango, lower tyre pressure will make the speedo read even higher! Dbfirs 02:11, 16 August 2010 (UTC)[reply]

You are right, I managed to get it back-to-front in my mind. The OP reports a higher than actual reading on the speedo, which is consistent with low tire pressure. --Tango (talk) 14:25, 16 August 2010 (UTC)[reply]

Cars in the US (at least) almost all have speedometers that read high. This topic comes up on a lot of car forums. I believe your sat nav system is telling the honest truth and your speedometer is lying. The reason for this is that no speedometer can be 100% accurate (because of tyre pressure, tyre wear, descrepancies due to going around corners, etc) - but if for any reason it read LOW, then it's possible that you could get a speeding ticket while you were under the impression that you were driving a little under the speed limit. Because you'd probably have car owners suing car companies as a result (sigh), the manufacturers make sure that the error is always on the high side. Numbers between 2.5 and 5mph have been reported - I've heard that BMW's read high by 5% of your speed or 2.5mph whichever is larger.

I'd bet that the exact same thing was true in other countries too.

The cops don't give a damn what your speedometer reads - they are measuring your speed using radar or lidar or by measuring the time it takes you to go between two marks on the road. SteveBaker (talk) 03:24, 16 August 2010 (UTC)[reply]

I think it's a little subtler than that. Comparing with my GPS, I find that my Toyota's speedometer is pretty much dead on up to about 70 mph or so (I'd guess a standard error of measurement of 2 mph or so, but it doesn't seem to be biased high). However once you start getting the speedometer needle up around 80, the GPS will read at least a few miles lower. I think the discrepancy goes up faster than linear — I haven't had the nerve to really run systematic experiments.

So the speed-limit thing wouldn't really explain this (when the car was made the speed limit was still 55! I'm not big on buying cars, or really, shopping in general....). It seems more like they want to reduce the harm when drivers intentionally go fast just to see the needle go up. --Trovatore (talk) 08:38, 16 August 2010 (UTC)[reply]

Your GPS (the Tomtom) is incredibly accurate as a speedometer. Even in the days of selective availability, when we only had 100m accuracy on GPS. Conveniently the science has been done, and the accuracy of consumer GPS devices was tested and found to be around 0.5km/h in most cases and 1.5km/h in the worst cases, and remained accurate even with poor satelite signal see this for example. —Preceding unsigned comment added by 203.202.43.53 (talk) 04:21, 16 August 2010 (UTC)[reply]

That is not my experience at low speed. For example when I am straining up a hill on my bicycle at a speed I know by experience to be roughly 4 mi/hr, my GPS will show maybe anywhere from 2 to 5 (that's miles, not kilometers), and occasionally even 0. (I think it biases low.) This is using a high-sensitivity model in a hilly area where some satellites may be intermittently visible and ghosts are certainly not out of the question. I do have the impression that the reading is more stable at higher speed (even in terms of absolute difference, not just as a proportion). --Trovatore (talk) 08:28, 16 August 2010 (UTC)[reply]

A point raised from you uphill cycling: does the GPSr register the horizontal component of speed (change in 2-D co-ords divided by time), or does it take elevation into account and calculate the true speed along the slope? Dbfirs 11:24, 16 August 2010 (UTC)[reply]

I don't actually know, but it doesn't make much difference. A 10% grade is relatively steep; at a 10% grade the difference between the two speed measures is only about 0.5%. --Trovatore (talk) 18:13, 16 August 2010 (UTC)[reply]

It makes a difference on a 30% slope (common where I live). I would test it out but these slopes are usually combined with dangerous hairpin bends! Dbfirs 18:29, 19 August 2010 (UTC)[reply]

Here in Auckland, they sometimes have a sign which tells your your speed, to encourage you to slow down if you're going too fast. I presume it uses LIDAR or whatever like a police speed gun or a speed camera. I've found on a friends red Mazda these show a speed about 10 km lower then what the odometer say (well at least at around 40-60 km/h. As I've hinted I've seen this on multiple different signs so I presume it isn't a calibration issue. And I don't see any reason why the government agency who put up the sign would want to make you think you're going slower then you actually are (no I doubt revenue collection comes in to it). In other words, the sign is probably resonably accurate and for this (fairly old) car the odometer shows a reading about 10km/hr faster then you're actually going. Nil Einne (talk) 13:40, 16 August 2010 (UTC)[reply]

I have nothing but personal experience to go on, but I do believe speedometers run a little fast. Fun test if you are really curious and have time to kill: Find a relativity straight stretch of highway. Get up to any safe speed, say, 55 or 60 miles per hour, (use cruise control if your car has it), reset your trip odometer and pay attention to mile markers, and drive for an hour. :) If you have driven 60 miles per hour for one hour and your trip odometer shows something like 56 miles, it's fast. If the trip odometer also shows 60, check the mile markers to see if that matches. (It's possible that your odometer is off, too.) ^Avic _ennasis @ 16:42, 6 Elul 5770 / 16 August 2010 (UTC)

I've worked as an engineering student at a Honda manufacturing plant in Canada and I can tell you with certainty that the speedometer is always calibrated to have a positive to zero tolerance. This means that your speedometer will always display a speed higher than your currently going or exactly your current speed. This ensures that your never speeding because of a bad calibration on the speedometer.