I’ve read the posts of people describing the IM Louisville bike course as rolling hills… not too hard… not too easy… etc and since that leaves room for interpretation, I figured I’d put myself at ease and hopefully a few others with a GPS report of climbing… I traveled to Louisville and rode the course with GPS and I hope this helps some of you…
There is 6909 ft of climbing over the entire length of the whole course according to my Garmin. The “loop” we do twice has 2000 ft by itself. Less than 7000 ft of climbing spread out over 112 miles… should be fun!
Thanks for your effort. I’m also sorrry to rain on your parade but would hate to think that many would take your numbers as “accurate” without realizing just how inaccurate they are.
Measuring altitude variation by GPS is inherently inaccurate. Have you ever ridden a pancake flat route and checked how much climbing you did? When your GPS unit indicates your accumulated elevation change was over 1000ft how do you respond?
Don’t despair! Use your GPS points to plot your route with topographic software and you can get some accurate and meaningful numbers. You’ll also probably be surprised by how much difference there is between the two.
Civilian GPS is extremely accurate for measuring linear distances. But the “float” or variation for measuring vertical distances makes GPS unsuitable for accurate measurements of this type. Even a barometric altimeter is more accurate than GPS for measuring vertical distances.
For approximately two years I beta tested software for Garmin’s Forerunners. On rolling bike courses the variation between GPS measurement of accumulated altitude changes and topographic measurement of altitude changes over the same routes was typically in the 26-44% range, FYI. On which end of this spectrum do your numbers fall?
If others have more accurate means of providing data… knock yourselves out. I had no parade to rain on and was simply sharing GPS data for those who could benefit from what I recorded on my ride.
In relative terms… I read on another thread that IM MOO showed less than 50 feet seperating the climbing on that course from what my Garmin came up with at Louisville. After riding both I thought they were fairly similar as far as amount of climbing, although I thought the layout and course design made the Louisville course seem a little easier. (Less turns, more gradual/fewer steep climbs, etc) . So if the GPS data is flawed, it’s probably flawed in both cases and the info might still be helpful to others and that’s why I posted it.
Dispute the accuracy of GPS if you want, but after riding both, I don’t believe there is any way you can convince me that the Louisville course is 26-44% more or less climbing than IM MOO if that’s the margin of error you wish to associate with GPS data.
“Dispute the accuracy of GPS if you want, but after riding both, I don’t believe there is any way you can convince me that the Louisville course is 26-44% more or less climbing than IM MOO if that’s the margin of error you wish to associate with GPS data.”
Time out my friend, I wasn’t raining on anything either, just stating an indisputable fact about GPS elevations. I have a lovely graph in front of me that looks like uncle Louie’s EKG right before we lost him. It indicates 634 feet of “climbing” on a 3.5 mile run course that is DEAD FLAT with one 60 foot hill. I also have a graph of the same course made with my Polar 625 which shows the same course having just over 100 feet of “climbing.”
So I’m not trying to convince you of anything regarding the relative hilliness of the two courses. I’m just telling you that your Garmin elevation data isn’t worth the paper it isn’t printed on.
I’m just telling you that your Garmin elevation data isn’t worth the paper it isn’t printed on.
So, then, what’s the use of having an elevation feature on a Garmin? (Not being difficult, just wanna know. I don’t have one, but seems like everybody else does).
“So, then, what’s the use of having an elevation feature on a Garmin? (Not being difficult, just wanna know. I don’t have one, but seems like everybody else does).”
In hopes that someday they’ll perfect something in the soft or firmware to give elevation data that has value? Good question for the folks at Garmin.
As it is, Polar gives good HR data, good user interface and a nice compact watch. Weak points are you have to send it back once a year for a batter and pacing data (625SX) is about as useless as warts on a pickle…
So Garmin gives you nice mapping data, great pacing data, and a nice selection of screens. Weak points are that the battery won’t last through an IM unless you’re really fast, wrist unit is about the size of a Mini Cooper, questionable water resistance and ludicrous elevation data.
SOLUTION: Garmin on the right wrist, Polar on the left?
Actually the comment that the elevation on gps is not worth much is not a true statement…
What you guys are talking about are total ascent or descent over a time or distance… That has nothing much to do with the gps itself… It has more to do with how you calculate the total climbing given the gps track data.
Would like to know resolution of topo maps to compare.
With gps you gonna get blips either loss of one or more sattelites for short periods or other factors.
Any instantanious track point(every 1 second or 5 depending on your hardware) can be off. But if you average things over time (just look at a profile from a gps) it’s pretty good. If you look at your garmin while climbing and look at it’s current altitude you will see it pretty dead on. I’ve been using gps in training since 1998 I think before they made them for that purpose.
I used to use a garmin gps III on my biked and even run with it had maps and everything… I have forerunner 201,301, foretrex and now a 305.
My garmin plus if you were standing still the elevation would move up steadily. I don’t have that problem with the new devices…
Topofusion software if I remember right will calculate
total ascent using multiple algorigthms…
If you take out the blips(much like polar lets you edit
heart rate info to take out those loose lead electrical interference blips like 220 bpm that happen from time to time) and decide what interval you want to smooth things it will very accurate. Now this you do after the fact download the tracks let the software do it.
I believe topofusion gives you option to decide what the minimum values to ignore up and down like 1,2…
whatever feet ups and downs to better average things.
Like I said I’ve been logging my runs and rides for almost 10 years on gps and for most part the speed,elevation I get during the ride are pretty dead on(at least with road signs for elevation on mountain climbs) and they are consistant over time each time I go over the same route.
The total climbing yes thats skewed and depends more on the software and algorithm you use to calculate it.
I’ve posted this here before. http://forum.delorme.com/viewtopic.php?t=6812 good discussion on the accuracies of varies elevation measurements. GPS is considerably less acurate than some others.
That being said in this case the original data posted appears to be fairly close. I laid the course out in TOPO, which is most likely more accurate than GPS, and although I doubt I have the map perfect here it is with stats…Sorry it’s so freakin’ big…
Cool thanks I’m doing louisville and have been looking for a bit more detail on the elevations of the route.
Like I said in my post it’s not the accuracy of the gps(especially with the new ones) it’s how they calculate
over all elevation gain(climbing,ascending,descending)
that can be off depending on how you filter it.
If you do the course pop your gps profile you will get pretty much same as your delorme. Now any specific location may be off depending by small amounts and you will have to filter out the blips which come and go but overall it will be right on.
I would like to know(guess I can research but if someone knows fell free to post) how the topo values are obtained. I assume from physical traverse,radar mapping,aerial and sattelite ranging surveying etc etc etc… But at what resolution both horizontal and in elevation… What I mean is do they get the contours every 5 feet,10,feet,500 feet or mixture along the way.
I guess for the roads if they were surveyed while being built and that information was saved probably that way.
Would be interesting to know.
I assume they know each grade down and up along the road and some major surveying was done both for property lines and construction. Hell If I know but would like to know.
My understanding is that TOPO gets there info from the USGS’s. The version I hae is less accurate than others based on the density of teh mess. If you go to the link I gave they go over some of this.
As far as GPS and accuracy you have two problems, one is the calculation, the second is simply the ability of the the satelites. Think of it this way. You have a block 100ft away from you. Is it easier to see when the block moves left or right or away and towards you. You have the same problem with GPS. Although the satelites are seperated by a distance greater than your eyes are apart and likely the overall ratio is larger, it’s a problem of “Depth perception”.
The accuracy of the GPS is simply far worse, even with the best calculations, in teh vertical than in the horizontal. Certainly TOPO is not perfect, but I’d trust it more so than GPS as far as vertical goes.
Highest elevation? Or do you mean most feet of climbing? If so then no. IMWI is similar around 6200-6500 and IMLP is significantly more at around 7500-8000.
I meant most feet of climbing. And sorry cos I’m used to dealing in Metres. But Nice is 2100 metres (6800ft) and UK is 2200 metres (7150 ft). Both are pretty tough courses, but I found UK easier than Nice because it was more rolling.
"Actually the comment that the elevation on gps is not worth much is not a true statement… "
OK, then let me say it again, so we can figure out what part of what I said wasn’t clear. I did the Lake Merrit 5K in Oakland, a course with an elevation change of less than ten feet and total climbing of less than thirty. Garmin indicated over 580 feet of total gain, which is an error of nearly 2000%. There is no term for that margin of error other than useless Last week at the Angel Island 25K, the topo stats were about 2000’ and Garmin recorded 4761 - much better, but still in excess of 100%.
“What you guys are talking about are total ascent or descent over a time or distance… That has nothing much to do with the gps itself… It has more to do with how you calculate the total climbing given the gps track data.”
Maybe the confusion is that you’re talking about the satelite-based gps system, and I’m talking about the Garmin unit on my wrist. To me, the Garmin on my wrist is the “gps itself,” because it’s all the data I’m going to get from all that expensive satelite equipment. The Garmin on my wrist DOES NOT provide elevation data that has any value whatsoever.
“Would like to know resolution of topo maps to compare.”
Topo has some pretty significant data as well, as anyone who has ever used DeLorme Topo (rereafter refered to as Topo) to create a CompuTrainer course knows. Topo maps are graded in increments, usually 10 feet, and the ridge lines (for the purpose of Topo) are treated as if there is a ten foot rise or gain right at the line. I could be wrong and that could be five feet or twenty feet, but that doesn’t matter - the transition on Topo is immediate andnot gradual. This is further complicated when dealing with roads because in many instances (PCH is a great example) the road is cut into rolling hills, so that at the top of the hill, you are riding in a groove between two rather tall bits of land. Topo doesn’t know about this, so it gives you a course showing what the road would do had it been laid directly over the existing land.
If you’ve ever ridden an Expresso Spark bike at your gym, this is the flaw with these. As you’re riding (on hills mapped from rides in the surrounding areas of Northern CA. If you watch the gradient as displayed on screen, you can see that Topo has indicated some gradients of 40% and more. Since such gradients are not even allowed, obviously these are incorrect.
I mapped the Santa Cruz Sentinel for the Computrainer several years ago, and it was ludicrous. I started riding it (and knowing the course like the back of my hand) visualizing where I was. Suddenly, from nowhere, I’m going up a 31% grade, standing on the pedals in my lowest gear at a point on the course where I’d be in the big ring…
“With gps you gonna get blips either loss of one or more sattelites for short periods or other factors.”
Considering that my Garmin shows that it’s got a grip on 15-25 satelites at almost all times, I’ll go with the “other factors,” most notably that it’s pretty hard to judge distances by triangulation from 100 miles up.
“if you average things over time (just look at a profile from a gps) it’s pretty good.”
No, it is beyond awful… It’s absolutely useless.
“I’ve been using gps in training since 1998 I think before they made them for that purpose.”
No, they made them for “global positioning,” which is to find something in TWO dimensions. The whole satelite deal is some time away from being able to reliably give you the third. If your particular GPS does well on altitude then it is equipped with a barometric altimiter in addition to the GPS. Garmin will tell you this if you e-mail them, and Motion Based has reference to this info on their site.
“Topofusion software if I remember right will calculate total ascent using multiple algorigthms… If you take out the blips(much like polar lets you edit heart rate info to take out those loose lead electrical interference blips like 220 bpm that happen from time to time) and decide what interval you want to smooth things it will very accurate. Now this you do after the fact download the tracks let the software do it.”
I’m interested in any means available to edit Garmin 305 files. I can’t even find a discrete file for any of my Garmin activities. I would love to be able to edit them as I’ve always done with Polar files, not only for the “glitches” but also because it messes up my training log when I either forget to turn it off at the end of a workout or forget to turn in back on after a stop.
“Like I said I’ve been logging my runs and rides for almost 10 years on gps and for most part the speed,elevation I get during the ride are pretty dead on(at least with road signs for elevation on mountain climbs) and they are consistant over time each time I go over the same route.”
So that makes Louisville the IM bike course with the highest elevation. More than Nice, Wisonsin, Malaysia and Lanzarotte? Can this be true?
Has anyone done LOU with a barometric altimeter. That I would trust.
don’t know if it is. BUT your not talking about elevation or elevation gain(in fact a loop course elevation gain is 0 or should be.
your talking about how much climbing you do.
So take for example(doubt any courses like this but lets use it)
Lets say you have this loop course it’s 2 miles long.
Each loop you climb to 200 feet half way in a mile then back down 200 feet straight up I believe that would be 3.7% grade (please correct me if I’m wrong thats 100 times 200/5280.
So if you do that 2 mile loop 56 times to get 112 miles
you have 11,200 of climbing or total ascent.
So now take a mountain course you start sealevel go straight up for 56 miles and back down you’d have to go 11,200 feet to get the same thing again no rollers just stright up. while thats the same grade 3.87 percent. overall climbing is identical.
Heres an extreme how about 1 feet up every 40 feet
then down 1 feet about 5% grade do this for 112 mile that would be 2.5% grade but over 112 miles of this you’d climb 7392 feet and you never got over a foot of in elevation. Thats why a supposedly flat course like florida and others will still show alot of climbing but there are no big climbs.
Now I threw this together real quick trying to get out to ride so forgive the math and please feel free to correct but thats how you can have more climbing on a what looks like flat course over a mountainous course.
In reality probably doesn’t exist unless you find some loop to do over and over.
I’m talking about the garmin on your wrist as well.
The garmin on your wrist is trying to calculate the over all climbing(thats what you are talking about) They just chose a bad algorithm for adding up all the climbing and averaging things out I guess because of lack of space for the extra code. thats whay you get…BUT if I look at my gps on my wrist while riding or driving whatever and have it showing elevation mine is almost always(if I have a clear sky good signals right on any road sign that tells me elevation… So on average they are accurate enough…
To get a more accurate measurement of over all climbing which most definitions means if I climbed 5 foot went down 10 foot then up 20 foot over a distance thats 25 feet climbing 10 feet descent. thats what gamins is trying to tell you on your wrist(total climbing that is) But it doesn’t filter out the momentary glitches…
But if you download the data use different algorithms
and different windows to smooth the gain per distance you will get pretty close to what other methods will show.
It’s just math thats all the unit itself is doing bad math. yes your instantanous accuracy is going to fluctuate right now it might be off 100 feet then later lower by 100 feet then seconds later 50 feet it fluctuates depending on how far you move and the sampling rate they all affect the average but over all over a longer course it will be pretty accurate
(not talking about quater mile track won’t work espsecially for the bends maybe if it could dsample 100 times a second and average over multiple laps but not very good there)
Anyway my experience (I’ve got 100’s of gps tracks from repeat courses over the years and with exception of when I run trails with lot’s of trees and ups and downs and turns which will with the current 5 second and down to 1 second(which is better) sample rate is terrible but on the open road it’s consitant over multiple times over the same courese…
I have couple imusa tracks,imwisconsin,lot’s of imfl,ironman hawaii and a number of smaller races over the years with various types of gps units and they seem to be pretty consitant…
So that makes Louisville the IM bike course with the highest elevation. More than Nice, Wisonsin, Malaysia and Lanzarotte? Can this be true?
Has anyone done LOU with a barometric altimeter. That I would trust.
Well, I’ve ridden the IM Lou course and have done IMLP and been on the IMMoo course. The IMLou course was MUCH hillier than I anticipated, but no way is the total elevation/climbing more than Lake Placid for example. I don’t trust the GPS altimeters at all either. My seat-of-the-pants meter is not very accurate either, but it is better than those things!
“With gps you gonna get blips either loss of one or more sattelites for short periods or other factors.”
Considering that my Garmin shows that it’s got a grip on 15-25 satelites at almost all times, I’ll go with the “other factors,” most notably that it’s pretty hard to judge distances by triangulation from 100 miles up.
NO this is not always the case there are 15-25 but because they are moving and some signals are lower than others it’s not always using them all. You will loose the signal say under a clump of trees or building etc from time to time thats what I meant by blips also blips for other reasons technical and just the way they operate…
“if you average things over time (just look at a profile from a gps) it’s pretty good.”
No, it is beyond awful… It’s absolutely useless.
NO IT’s not useless I have 100’s of profiles they look pretty much like those posted by races and consitantly same for multiple times over the same route.
“I’ve been using gps in training since 1998 I think before they made them for that purpose.”
No, they made them for “global positioning,” which is to find something in TWO dimensions. The whole satelite deal is some time away from being able to reliably give you the third. If your particular GPS does well on altitude then it is equipped with a barometric altimiter in addition to the GPS. Garmin will tell you this if you e-mail them, and Motion Based has reference to this info on their site.
NO THEY Don’t have barametric altimiter Garmin GPSIII does not forerunner 201 and 301 do not… Don’t think the foretrex has either could be wrong about the 305… If it did i’d get reading for altitude without sattelites. I’ll check the reference anyway.
My polar watch has barometric and barometric is accurate to a point(not sure how much resolution but same is true as gps)
Barometric over a long day will not be accurate if say a low preasure or high preasure system comes through (storm front cold front etc) but it’s pretty good if you calibrate it to know elevation.
“Topofusion software if I remember right will calculate total ascent using multiple algorigthms… If you take out the blips(much like polar lets you edit heart rate info to take out those loose lead electrical interference blips like 220 bpm that happen from time to time) and decide what interval you want to smooth things it will very accurate. Now this you do after the fact download the tracks let the software do it.”
I’m interested in any means available to edit Garmin 305 files. I can’t even find a discrete file for any of my Garmin activities. I would love to be able to edit them as I’ve always done with Polar files, not only for the “glitches” but also because it messes up my training log when I either forget to turn it off at the end of a workout or forget to turn in back on after a stop.
Theres a program out there I downloaded long time ago called get301tracks.exe it will download the tracks for most garmin including 305… Theres software out there for editing for sure. Mainly I have topofusion I’ve used for a long time it has section to do climbing analysis and lets you see different methods for doing this and some settings as to select say anything over 3 feet difference between tracks ignore or whatever value you choose and to filter out small fluctuations…
OR you can just download or search for the garmin protocol and write your own software but theres plenty out there(see yahoogroups garmin or gps forum plenty of stuff there)
“Like I said I’ve been logging my runs and rides for almost 10 years on gps and for most part the speed,elevation I get during the ride are pretty dead on(at least with road signs for elevation on mountain climbs) and they are consistant over time each time I go over the same route.”
Then your device has a barometric altimiter.
NO IT DOES NOT HAVE an ALTIMETER… At least the ones before 305(I don’t know about it I just got one but I’ll check)
But I haven’t use it more than a week but my other gps units, forerunner 201,301,foretrex no altimeter…
there are some handheld versions garmin makes and made just regular gps units(not geared for cycling,running ) that
do I know that… Foretrex is one of those but it has no altimeter…
“NO this is not always the case there are 15-25 but because they are moving and some signals are lower than others it’s not always using them all. You will loose the signal say under a clump of trees or building etc from time to time thats what I meant by blips also blips for other reasons technical and just the way they operate.”
yeah, some HUGE blips. Actually, what Garmin does (don’t know about your GPS) is that if it loses signal, it retains the data and assumes nothing has changed until signal is regained.
“NO IT’s not useless I have 100’s of profiles they look pretty much like those posted by races and consitantly same for multiple times over the same route.”
Yeah, like I said, useless. Have you never noticed the disparity from the map elevation on the website to what you actually experience on the bike and run? The majority of these are from Topo, however.
“NO THEY Don’t have barametric altimiter Garmin GPSIII does not forerunner 201 and 301 do not… Don’t think the foretrex has either could be wrong about the 305… If it did i’d get reading for altitude without sattelites. I’ll check the reference anyway.”
Garmin says the 305 has barometric, but you couldn’t prove it by what I’ve seen from mine.
“Barometric over a long day will not be accurate if say a low preasure or high preasure system comes through (storm front cold front etc) but it’s pretty good if you calibrate it to know elevation.”
Barometric will vary with heat also, but it’s still WAY more accurate than GPS elevation.
“Theres a program out there I downloaded long time ago called get301tracks.exe it will download the tracks for most garmin including 305… Theres software out there for editing for sure. Mainly I have topofusion I’ve used for a long time it has section to do climbing analysis and lets you see different methods for doing this and some settings as to select say anything over 3 feet difference between tracks ignore or whatever value you choose and to filter out small fluctuations…”
