Certainly agree with your reasoning and conclusions. There is a lot of computation involved in just getting a gps position fix. All the derived stuff the gamins provide (speed, distance, vertical rate, grade, feet climbed, etc.) involves more computing. Plenty of room for errors. Could be bugs, poor numerical methods, or just limits of accuracy in the system. In any case if adding the speed-cadence thing eliminates the jumps, that clearly implies the GPS only system is deficient.
Roughly what size jumps you are you seeing? Do they show up in the time history plots? Can you say what sort of duration and frequency the jumps have? I'm thinking I should try to look for them with my 510. Perhaps I just haven't noticed them even though they are there, perhaps the GLONASS make a difference. I'm kinda curious. On Monday, September 15, 2014 6:53:26 PM UTC-7, Benz, Sunnyvale, CA wrote: > > I'm not sure if it is specific to my 500 or includes other units. > Certainly, the jumps are small but they are noticeable and consistent, even > when under clear skies and constant speed on a straight road. They do > disappear abruptly when the Garmin sensor is installed which gave some a > clue. Nevertheless, the speed that GPS-based units display must be > calculated instead of measured like with the Cateyes. If GPS signal isn't > optimal, it is conceivable (doesn't mean it's the real reason) that the > calculation reveals the lag in positioning. > > > On Monday, September 15, 2014 9:35:14 PM UTC-4, ted wrote: >> >> I have never noticed the speed reading jumping around on my Garmin 510, >> exepct when I hooked up a speed cadence sensor and didn't set the wheel >> size. >> I this erratic speed reading you speak of a feature of all Garmin units? >> Is it limited to those that don't use GLONASS and GPS? Is it all the time >> every where or limited to specific situations? >> >> On Monday, September 15, 2014 5:45:53 PM UTC-7, Benz, Sunnyvale, CA wrote: >>> >>> Although it is true that the Garmins do not require anything more than a >>> quick charge, the speed reading without the optional speed/cadence sensor >>> tends to jump around. While I don't believe this impact average speed much, >>> I can't stand the lack of precision from such a measuring device (without >>> its optional speed/cadence sensor that further optionally can be >>> calibrated). >>> >>> The Cateye Strada wireless units are (relatively cheap), have big main >>> numbers (speed), only need a battery change once every year or so, and is >>> better than 1% accurate even if you follow the lookup table on the >>> instruction sheet. You don't have to break out your measuring tape and >>> inscribe markings on the garage floor if you don't want to, although doing >>> so will give you better than ppt (parts per thousand) accuracy. >>> >>> The Knog meters are more or less a joke because of their mounting >>> system. They use the same silicone straps that are quite useful on lights >>> but are dreadful on both the display/head and sensor units. The straps slip >>> easily. I have two sets that I got cheap but I don't use anymore. >>> >>> >>> On Monday, September 15, 2014 12:40:19 AM UTC-4, ted wrote: >>>> >>>> though you can probably get a nice basic unit for 1/100 the $$, the >>>> garmin 510 is nice. Being gps it needs no wires or wheel magnets or >>>> pickups. It can be mounted on your bars or just thrown in a pocket or bag. >>>> You can load routes to it and have it beep at you when turns are coming >>>> up. >>>> you can pick what and how much info is displayed on the screen at one >>>> time, >>>> and you can set up several screens that you can page through or just stick >>>> with one. >>>> >>>> On Sunday, September 14, 2014 8:39:48 PM UTC-7, Neil wrote: >>>>> >>>>> Despite historical resistance to the idea, I am contemplating a bike >>>>> computer for my Sam, the better with which to follow cue sheets and the >>>>> like. Any suggestions from the Bunch? I suppose I would prefer wireless, >>>>> and a small, modest screen. >>>>> >>>>> Cheers, >>>>> >>>>> Neil >>>>> >>>> -- You received this message because you are subscribed to the Google Groups "RBW Owners Bunch" group. To unsubscribe from this group and stop receiving emails from it, send an email to rbw-owners-bunch+unsubscr...@googlegroups.com. To post to this group, send email to rbw-owners-bunch@googlegroups.com. Visit this group at http://groups.google.com/group/rbw-owners-bunch. For more options, visit https://groups.google.com/d/optout.
