Has anyone had an accurate temperature measuring device on a bicycle computer?
On club rides, if there are 10 riders, there will be 10 different readings. Most of the readings are different by 4 to 15 degrees. My computer is a pretty good one. Almost everything else reads fairly close to other riders computers, so it seems that accuracy is in other parts, but not temps.
Before I make this sound like a rant, I am pleased with the overall level of most computers. It's just the anomoly of the temperature sensors in all computers that has me amused.

Brendon The Animal,"Wow, it seems warmer than 45 F".
Larry the Energizer Bunny,"That's because it's really 60".
Jim The Diesel,"I've got 56".
Tarantula,"It's 52".
Ironhead,"Who cares. We still have 14 miles to the coffee shop".

well, i've never tested mine against a real quality thermometer, but...

* my Cateye seems to often be quite wrong - the temperature just seems way off - like it'll say +5C when it's snowing and the predicted high for the day is -1C.
* my Specialized pro i used to have (fell off and got lost on a steep descent Fall 2001) seemed to be pretty much right
* my Ciclo (414M?) seems to always be right on

so i think there is some major variance in the quality of the temperature readings which makes sense since it's an "extra" feature often just added on to increase the list of functions...

also on the elevation and the elevation gained measurements, my Cateye is just worthless (for instance i haven't calibrated it in weeks and it currently shoes -300m when it should be 510m) while my Ciclo is so unbelievable accurate and reliable i have been amazed -- the values always seem to jive with map values AND when you compare 4 different computers readings for the same trip you get within a few meters (maybe some guy rode back a little bit or someone took a detour to relieve himself or check out a view...) --- it often seems to be much more accurate than my expensive mountaineering Sunto altimeter - especially in counting vertical ascent/descent.

needless to say, i am QUITE satisfied with the Ciclo... although i'm not sure how common they are out of Europe. in Germany it is pretty much the TOP computer - they make HRMs too but i've no experience with them. the major complaint about Ciclos seems to be that they have SO many functions that they can be hard to use. my Ciclo doesn't have as many functions as others (some have ski-speed and HR and many more averages, maxes than the basic ones i have) although it has all i need (2 wheel settings, min/max/average for most values, plus computer interface to download all my data)

if i think about it, sometime i'll run a temperature reading test between my Ciclo and my Sunto (top mountaineering altimeter with temperature) on my next tour...

I would think color of the computer casing would have an effect. If it is black you will probably get a much higher reading. I had a watch one time that was always off, however, if I kept it in the shade and off of my wrist, it was fairly close. Direct sunlight will throw just about any thermometer off.

You should also check the instruction booklets that come with the computers. I bought a specialized computer (the pro I think). It clearly stated that it was only rated for temps between 45F and 100F. And it was true. I would go out on days that were 32F and after 2 hours outside, it would finally start reading 40F. I moved onto another computer, the Planet Bike Protege 9.0 at it is very accurate to... at least to the temps I see when I open my Weather Bug and it's reporting the temperature at the local elementry school. Plus I have an old Trek Radar computer that is also very accurate. Both the Trek and the Bike Planet read almost exactly the same thing all the time. The other two I've tried with temperature (the specialized and a vetta) were usually off and gave me wild readings. Plus both of those were also wireless. I have swore off wireless computers.

PBW