Ambient temp display granularity

[maxtog]

Way low on the list of things to be concerned about....  Over the years, I have noticed a very curious thing about the ambient temperature display.  Once I check tire pressures, I leave the computer display on the ambient temp- seems to be the most useful info available for general riding use.  Anyway, I have noticed that it tends to almost always jump in 2 degree (F) increments.  I can be on a two hour ride, and the entire time, not see a single even-numbered temperature.  Sometimes it will jump back and forth between two odd numbers, skipping the even one.  On other rides, it will do the same thing, but only with even numbers.

Very rarely will I see an odd AND and even number on the same trip.

Has anyone else ever noticed this?  What kind of explanation could account for this type of behavior?
--------------
UPDATE:  It is simply due to converting from single/native degrees C to F... duh

[gPink]

http://youtu.be/8CtjhWhw2I8
.

[lather]

Fascinating. I did not even know the newer models had ambient temperature display. Now I want a new Concours!

[B.D.F.]

I don't know Bob, I hear they will not display odd temperature numbers. That might be a deal- killer for some.....

 

There are over seven billion humans on the planet. More than half of them wonder what, or if, they will eat in the next week. Then there is 'us', wondering why the ambient temperature gauge on our ambient temperature gauges does not have 'correct' resolution. Now ya' gotta' admit, that right there is funny. Pathetic too, but with enough funny to give it merit IMO.

Wonder what Marie Antoinette (of "Let them eat cake" fame) had for breakfast before her head fell into that basket?

Brian

Fascinating. I did not even know the newer models had ambient temperature display. Now I want a new Concours!

[khager01]

Funny but I have noticed the same thing too!  Following this thread.

[maxtog]

I forgot to post a follow-up after thinking about it more.  I am pretty sure I know what they are doing and it is exactly as I previously described.  To prevent the meter from jumping constantly from one reading to another, they are only allow the screen to update if it is 2 degrees difference.  It is a type of rounding.  The odd/even thing is just random- it depends on the very first reading when the ignition is turned on.  If it was odd, it will remain odd until the bike ignition is power cycled.  If was is even, it will remain even until the bike ignition is power cycled.  I find it a little annoying, but I am about 99% sure it was intentionally designed this way to make the reading more "user friendly."

They probably should have said this in the user manual.  But we all know how user manuals are...

[chap]

Maybe it calculates in deg C and converts

[maxtog]

Maybe it calculates in deg C and converts

I rejected that thought a while ago because it wouldn't explain the odd/even thing.  Looking at a table, you can see either truncated or rounded Celcius conversions would still result in odd and even numbers in F.

http://www.temperatureworld.com/ctable1.htm

[Rhino]

Maybe it calculates in deg C and converts

Just a guess but I think this is correct. I have observed the exact same thing as you Max. For awhile I thought it only displayed in odd temps. But then I would see it even a few times. I have noticed on my Ducati CC when you bump the accel button sometimes it does not read 1 mph higher but sometimes it does. I think it is bumping by 1 kph but I have it is displaying in mph. I think something similar may be going on with the outside temp display on the C14.

[Rhino]

I rejected that thought a while ago because it wouldn't explain the odd/even thing.  Looking at a table, you can see either truncated or rounded Celcius conversions would still result in odd and even numbers in F.

http://www.temperatureworld.com/ctable1.htm

Maybe a floating point error. I see that a lot in some languages.

[maxtog]

Just a guess but I think this is correct. I have observed the exact same thing as you Max. For awhile I thought it only displayed in odd temps. But then I would see it even a few times. I have noticed on my Ducati CC when you bump the accel button sometimes it does not read 1 mph higher but sometimes it does. I think it is bumping by 1 kph but I have it is displaying in mph. I think something similar may be going on with the outside temp display on the C14.

It is possible that with certain rounding methods, truncation, or even floating point anomalies, it might keep the numbers even or odd throughout a certain RANGE that is big enough to fool us most of the time.  It it is quite possible that on any one trip, one could find themselves only in that range and see only odd or only even numbers.

The "temp at startup" with 2 degree granularity VS. the strange C to F conversion could be easily proven/disproven if someone can verify they have, indeed, seen BOTH odd and even temps without restarting the engine.  I can't be 100% sure I haven't seen both, even if it is really rare.  I will keep an eye out for it, though

I suppose another good test would be for people to report what it does when the display is put into Celsius units.  I have never done so (and really don't wish to, since I can't think in C), but if it moves in single degree C numbers (both odd and even), it lends even more credibility to it being due to strange C to F conversion...

First world problems!  https://en.wikipedia.org/wiki/First_World_problem

[Rhino]

It is possible that with certain rounding methods, truncation, or even floating point anomalies, it might keep the numbers even or odd throughout a certain RANGE that is big enough to fool us most of the time.  It it is quite possible that on any one trip, one could find themselves only in that range and see only odd or only even numbers.

The "temp at startup" with 2 degree granularity VS. the strange C to F conversion could be easily proven/disproven if someone can verify they have, indeed, seen BOTH odd and even temps without restarting the engine.  I can't be 100% sure I haven't seen both, even if it is really rare.  I will keep an eye out for it, though

I suppose another good test would be for people to report what it does when the display is put into Celsius units.  I have never done so (and really don't wish to, since I can't think in C), but if it moves in single degree C numbers (both odd and even), it lends even more credibility to it being due to strange C to F conversion...

First world problems!  https://en.wikipedia.org/wiki/First_World_problem

I didn't notice the startup state much but the running values were certainly odd numbers MUCH more often than even numbers. If I had to guess 10 to 1 odd to even.

[B.D.F.]

Yeah, that is the most reasonable idea I think. Because the <basic> ratio between Celsius and Fahrenheit is 1.8, which is very close to 2, it is quite likely that the software simply rounds temp. data to the nearest degree C, then converts to F, which would take place two numbers at a time for great ranges of the conversion.

If this is correct, it would also mean that small 'batches' of numbers would be displayed exclusively in even, and then odd numerals only. So for example, the numbers might go something like this in Fahrenheit:

50
52
54
56
57
59
61
62
64
66

and so on.

Brian

Maybe it calculates in deg C and converts

[B.D.F.]

Maybe C-14's are made in batches or lots of even / odd displays? You know, the cosmic balance and all.....  like Yin / Yang, Right / Left, Clockwise / Counter-clockwise, Heter..... wait, wait, can't go there anymore.

 

Personally, I have no idea or concern with such things; as I have an '08, all I get for ambient temp. readings is {null} / {null} anyway. But I do have an external temp. gauge that displays in 0.1 degree F increments. I found this very, very helpful when riding through South Dakota for what seemed like 50 hours one day and the temp. settled in at 102.6 F. That 0.6 pushed the situation over the proverbial edge and made the ride unpleasant. Later, when my heart attacked me in PA, and that poor, poor nurse took my riding boots off my feet, that 0.6F extra temp. came back to bite her too; my socks were kinda' stuck to my feet. Not a pretty story. Although I did tell 'The Cat Joke' well enough that the nurse I originally told it to went and got other staff for an encore performance.

Brian

I didn't notice the startup state much but the running values were certainly odd numbers MUCH more often than even numbers. If I had to guess 10 to 1 odd to even.

[chap]

It is possible that with certain rounding methods, truncation, or even floating point anomalies, it might keep the numbers even or odd throughout a certain RANGE that is big enough to fool us most of the time.  It it is quite possible that on any one trip, one could find themselves only in that range and see only odd or only even numbers.

The "temp at startup" with 2 degree granularity VS. the strange C to F conversion could be easily proven/disproven if someone can verify they have, indeed, seen BOTH odd and even temps without restarting the engine.  I can't be 100% sure I haven't seen both, even if it is really rare.  I will keep an eye out for it, though

I suppose another good test would be for people to report what it does when the display is put into Celsius units.  I have never done so (and really don't wish to, since I can't think in C), but if it moves in single degree C numbers (both odd and even), it lends even more credibility to it being due to strange C to F conversion...

First world problems!  https://en.wikipedia.org/wiki/First_World_problem

When in dec C mode the temp goes up and down in 1 deg C increments

[maxtog]

When in dec C mode the temp goes up and down in 1 deg C increments

Well, there ya have it, then!  Mystery mostly solved.  It is a conversion issue PLUS some creative truncation and rounding strangeness (because if were done "correctly", the spans of odd or even numbers would not be so broad).

Thanks for the info.

[datsaxman@hotmail.com]

You guys sure want to make up some complicated stuff to explain something quite simple.

Kawasaki - and the whole world except for a few places - uses Celsius temperature.

The Concours 14 is happy to display Celsius temps.  OR to display the equivalent Fahrenheit temps.

This are correctly rounded.  Not some crackpot granular or odd and even thing.  Max ... really!
Not truncated, but properly rounded. 

Notice that the pattern in the decimals repeats.  That is because 180 Fahrenheit degrees is EXACTLY the same size as 100 Celsius degrees.  So every fifth Celsius temp.  is an exact integer Fahrenheit temp. also.  This was a deliberate choice at the time, BTW.

Actual temps seen in the last two days by me (it is getting chilly in the mountains, and I had to go to LA yesterday, so got a great range.  Confirmed and filled in some gaps today.)  I usually have the display in Celsius.  It makes my tortured brain happy to see even increments.  This week I have it in Fahrenheit just to gather the data.  Yer welcome.  I think.  It gave me a headache. 

C       = F     Display
5       41.0     41
6       42.8     43
7       44.6     45
8       46.4     46
9       48.2     48
10     50.0     50
11     51.8     52
12     53.6     54
13     55.4     55
14     57.2     57
15     59.0     59
16     60.8     61
17     62.6     63
18     64.4     64
19     66.2     66
20     68.0     68
21     69.8     70
22     71.6     72
23     73.4     73
24     75.2     75
25     77.0     77
26     78.8     79
27     80.6     81
28     82.4     82
29     84.2     84

NO intermediate temps. were seen.  And I looked and looked.  Watched it bounce back and forth a whole lot of times.  77,79,77,79,77,79 on the way to LA.  61,63,61,63,64,63,64,66,64,63,61 going up the mountain.  46,45,43,45,43,45,43,41,43,41,43,45 almost at home (6000'). 

Max, I betcha a dollar you did NOT see 42, 44, 49, or 53 degrees Fahrenheit.  EVAH. 
$1 US in the mail for each of those temps.  Get a screen shot, post here on the forum.

Didn't think so.  Luv ya, man. 


dat
sax
man 

[maxtog]

Max, I betcha a dollar you did NOT see 42, 44, 49, or 53 degrees Fahrenheit.  EVAH. 
$1 US in the mail for each of those temps. 

LOL!  You are "da bomb."  I could swear the ranges I see of odd numbers I often encountered was wider than that.  But perhaps I was riding too much in the narrow odd ranges (and maybe with enough variation to magically skip the even number that separates two odd series).  I can't swear that wasn't an even thrown in every now and then that I didn't see, but I also was not in a position to write down all the numbers (or memorize them, which would be impossible for me), or stare at the screen enough.  Kudos to you, indeed!

Oh, I updated the first post to quickly answer for those later finding this thread.

Others that should then be impossible include (since we are in that temp range now): 29, 31, 33, 35, 38, 40

[mikeyw64]

wonder if we will ever switch to using Kelvin in daily use?

[B.D.F.]

Yeah, we be stupid. [need emoticon for drooling, stammering idiot BTW]

Except 'chap' said that very thing, on 7 December, in post number 6. And I agreed with the idea and offered a method on how the display would skip some individual readings in Farenghight but not all. But hey, that was one page before and 9 days ago so you certainly introduced the idea on THIS page and THIS day. Thanks!

 

Brian

P.S. Next week I plan on posting this same thing again, after you 'introduce' us to this same idea.... again. 

You guys sure want to make up some complicated stuff to explain something quite simple.

<snip>