I've had an opposite experience with realspeed. My max display on my foretrex said 31.3 but once it was put into realspeed it ended up as 30.4. How come I didn't get an extra 4 knots like Troppo and Firie???
Ahhh... the mysterious case of the missing knots! :-)
The case you describe is quite normal and not to do with the analysis software. The trackpoint speeds calculate from a different set of data (location)than the GPS readout speed (Doppler). The GPS units that can Log the NMEA data can save the Doppler speeds and when these are used for the calculations there is usually very close agreement between the peak speeds from the software and the GPS display.
The trackpoint max speeds tend to be a bit unrealistically optimistic. :-)
can anyone let me know what lap racig is in the new version of realspeed.?
cheers
Gestalt I think you can set Waypoints to make a coarse that you have too sail within a certain distance from each waypoint to complete the lap. mmm sound interesting.
ta vando
It's good to see Mal has stepped up to the plate and started posting here. Who better to answer questions about the software and how it works other than the author. Hopefully Mal can explain all the parameters that need to be set in options to measure our speeds correctly and filter out the spikes in the data. It seems every update of the software has more parameters for filtering data but there is no guide to let us know how and what to set them at. I'm sure with the formal release due in a month or two a users guide will be included with the purchase price.
quote:
Originally posted by decrepit
Thanks guys, hopefully I'll get to try that tomorrow.
Just tried File > Properties, and RealSpeed crashed every time.
But as Elmo says total distance is at the bottom of the track data in meters
for me.
Sorry that should have read
"So .PTL.gpx and .PTL.NMEA files work OK," and not as I put in previous post 
"So .PTL.ptx and .PTL.NMEA files work OK"
Lap Racing allows very easy analysis of your performance around fixed points or along a track. Taking a course race as an example, you might do a number of laps around two or three markers. To anlyse your performance between the markers, you can open the trackfile, click on the map view to create the two or three waypoints, create a new division and add the waypoints to the division.
RealSpeed will then show you all the performances between each waypoint so you can compare all your upwind legs and downwind legs for example. If you race around the same waypoints regularly then this is very convenient because you wont have to add the waypoints or division again. Just open the new file and the results will automatically be there.
Another example is if you want to do a cycle or a run along a known course. You can create waypoints for major points of interest like the bottom and top of hills and look at your performance on the hill climb. Adding these Lap Racing divisions do not affect the calculation speed very much and wont clutter your results if they dont apply, so you can create a Lap Race division for all the courses you want to sail, ride, run or paddle!
RealSpeed 1.9 will also allow multiple files to be opened at one time. This gives you a great way to compare your performance between races or on differrent days.
And one more... RealSpeed will analyse results from multiple GPS units recording over the same time period. The calculations will then be based on the merged tracks and provide even more accurate results. Because the GPS units typically record out of sych, it can add a lot of detail to the tracks, filling in some of the gaps particularly on turns and around the points of maximum speed.
The issue with File/Properties crashing seems to depend on the file. It looks like an invalid lat or long is corrupting the bounds. I'll try to solve this once and for all in 1.9.
If you have a file that crashes in RealSpeed then send it to me so I can work on a fix. Thanks.
Kean, I'll do my best on a user guide and FAQ over the coming weeks.
Hi Mineral.
If you are working with files with those endings you must be working with the files downloaded from the internal memory via Navilink.
Those files are not as good as the ones saved to the SD card. The SD card files contain more accuracy and a lot more information. They are the ones you need to use. You just copy them from the card with a card reader and the file extention should be .txt. (Well, it is for me.)
If you open it in a text program (like notepad) it should look like this small sample:
This is the data from 1 second from an SD card:
$GPRMC,034158.682,A,3850.7095,S,14609.5444,E,26.05,182.88,130407,,*2C
$GPVTG,182.88,T,,M,26.05,N,48.2,K*64
$GPGGA,034159.682,3850.7169,S,14609.5441,E,1,08,0.8,1.9,M,-4.7,M,0.0,0000*76
$GPGLL,3850.7169,S,14609.5441,E,034159.682,A*26
$GPGSA,A,3,28,09,04,08,29,17,26,11,,,,,1.9,0.8,1.7*32
$GPGSV,2,1,08,17,75,128,45,28,36,107,46,26,35,295,41,09,30,226,33*72
$GPGSV,2,2,08,29,29,313,43,04,23,023,38,08,19,043,41,11,10,137,41*7B
Note that there are 6 NMEA 'sentences'.
Here is what I just got from the internal memory:
$GPRMC,015045.000,A,3851.1783,S,14609.9631,E,24.84,112.00,110607,,*27
$GPGGA,015045,3851.1783,S,14609.9631,E,1,,,4.5,M,,,,*23
In this sample there is only 2 NMEA sentences, and it looks like the Doppler speed value is rounded off in some way in the files I just ran because the doppler speed graph is too blocky. The trackpoint data looks the same though.
The speed figure is in the RMC line. The seventh item (comma deliniated) In the above examples it is after the letter E and in the first it is 26.05 (knots) and in the second it is 24.84. (The two samples are not from the same run or even the same day).
The normal and encouraged proceedure is to use an SD card and analyse the data from that.
If I understand it correctly, there is firmware being tested now that can write the Sirf binary data to the internal memory as well as the SD card which should be quite an advantage because we may be able to have the same data twice. One set on the SD card and another duplicate set on the internal memory. There will be a backup built in. Since the binary data takes up much less space we should be able to get many more hours of data on the internal memory than we can now.
The gist of all this is: Use the SD card data.
sailquick, intend sticking too the SD card. I was mucking about and trying to get the NAVi on board to record, eventually did with a lot of coaching from forum members. Big tip, when all else fails RTFM
sent some files to Mal for him to play with. Buggered if I know what the issue is. I do know I used to get this similar program crash problem due to spikes when I ran Application Severity Analysis program with Caterpillar in early days working with large mining gear data logger files. Cat I.T crew worked it out so reckon Mal will be onto it in a flash with the new version.
mineral
I have fixed the File/Properties crash. It was an error in logic when processing the altitude. It only occured when the max altitude was less than zero.
This fix has been posted up in version 1.905 on my website with many other fixes (its been another big weekend of coding to stabilise RealSpeed...).
Also available on my website is a pair of example files (actually my best runs in March) so you can see how the multi-file calculation is working. Take a look at the merged doppler option in the speed graph by right clicking in the speed graph and select Graph Fields... to choose which speeds you want to see.
While you are there, take special note of how well the two doppler tracks (green lines) agree.http://www.intellimass.com/RealSpeed/realspeed_beta.htm
Loaded the 1.905 version today, certainly fixed the File>properties crash. but doppler file drops my speeds by .5kn. guess that's the accuracy increase, caus it's more in line with the unit readout.
But something's wrong with the alpha calculations, seems to only do one direction.
Here's the results from 1.804 and 1.905, from a small file made on Saturday.
1.804
Competitor,Division,Series,"Start Time","End Time",Seconds,"Distance (m)",Speed(knots)
,"Alpha Racing 500",20070623_112325,"23/06/2007 3:33:52 AM","23/06/2007 3:34:47 AM",55,498.888,17.632
,"Alpha Racing 500",20070623_112325,"23/06/2007 3:35:51 AM","23/06/2007 3:36:56 AM",65,445.751,13.33
,"Alpha Racing 500",20070623_112325,"23/06/2007 3:41:51 AM","23/06/2007 3:42:03 AM",12,70.709,11.454
,"Alpha Racing 500",20070623_112325,"23/06/2007 3:38:16 AM","23/06/2007 3:38:29 AM",13,74.294,11.109
,"Alpha Racing 500",20070623_112325,"23/06/2007 3:30:47 AM","23/06/2007 3:31:07 AM",20,93.301,9.068
,"2 Second",20070623_112325,"23/06/2007 3:29:16 AM","23/06/2007 3:29:18 AM",2,31.292,30.413
,"10 Second (Average 5x 10sec)",20070623_112325,"23/06/2007 3:29:15 AM","23/06/2007 3:40:39 AM",50,118.609,23.056
,"1852m (Nautical Mile)",20070623_112325,"23/06/2007 3:31:57 AM","23/06/2007 3:34:58 AM",181,1856.364,19.936
1.905
Competitor,Division,WayPoint,Series,"Start Time","End Time",Seconds,"Distance (m)",Speed(knots),Claimed(knots)
,"Alpha Racing 500",,20070623_112325,03:35:51.954,03:36:56.950,64.996,445.751,13.331,0
,"Alpha Racing 500",,20070623_112325,03:41:27.933,03:42:58.928,90.995,487.447,10.413,0
,"Alpha Racing 500",,20070623_112325,03:30:41.973,03:31:47.969,65.996,303.141,8.929,0
,"2 Second",,20070623_112325,03:29:16.978,03:29:18.978,2,31.292,30.413,0
,"10 Second (Average 5x 10sec)",,20070623_112325,03:29:15.978,03:40:39.936,49.997,118.609,23.057,0
,"1852m (Nautical Mile)",,20070623_112325,03:27:14.986,03:30:50.972,215.986,1860.034,16.74,0
The alpha here doesn't have as many results, on the diagram, they're all on the way out. the 1.804 results have both directions.
Sorry Mal looks like more coding needed!!!
In 1.901 the Nautical Mile and 1 hour calculations don't show up. Is this fixed in the latest update?
quote:
Originally posted by nebbian
In 1.901 the Nautical Mile and 1 hour calculations don't show up. Is this fixed in the latest update?
Hi Nebbian, the issue was that I tightened up the handling of invalid points when implementing all the new features in 1.900. Long runs like the nautical mile and one hour (and virtually any distance cycle or run) with a few invalid points will not produce a result so I need to fix this. My solution is to provide an option in the divisions to use a "Strict" method for short course speed competitions like on GPS-Speedsurfing, and a less strict option for all other divisions. I should have this done by the end of the week.
This will again raise the complexity of setting up divisions, so I will probably separate advanced settings from standard settings and default to not strict. For GPSSS we can produce a standard rsp file with all the correct settings so that should not be a problem either.
If you have any other suggestions then let me know.
Hardy and I went for a 1hr today, wind was light and gusty, my navi registered 27.something max speed. when I loaded the sd card .txt files into realspeed 1.804 and deleting invalid, it had a 32.something 2sec max. It was obviously wrong, as it was at the very beginning of the session while I was still in the carpark!!!! possibly caused by going to the car for something and getting in a satellite shadow. For some reason "delete invalid" didn't remove it.
However when I used 1.905 with the doppler file, that came up with a similar 27.something 2sc max.
Be nice if everybody used the doppler version for max speed, as it does seem much more accurate. But don't rely on it for alpha's at the moment, seems to miss one direction out of the calculations.
Aplogies Decrepit, I missed your big post...
Could you send me the file at mwright@intellimass.com so I can find out what is going wrong? Thanks!
quote:
Originally posted by TimeMachine
Aplogies Decrepit, I missed your big post...
Could you send me the file at mwright@intellimass.com so I can find out what is going wrong? Thanks!
I may well be wrong, but doesn't the display use doppler speed, instead of 1 second distance between points?
So if you set up RealSpeed to use doppler, then it will always agree with the display.
Or am I missing something?
quote:
Originally posted by nebbian
I may well be wrong, but doesn't the display use doppler speed, instead of 1 second distance between points?
So if you set up RealSpeed to use doppler, then it will always agree with the display.
Or am I missing something?
I suggest using Trackpoints for Alpha runs, not Doppler. This is one area where it seems to be better for some reason I don't know. All the current posted alpha speeds are Trackpoint speeds and if you contimue to use this it keeps things comparable.
Trackpoints would need to be involved anyway, to calculate the distances wouldn't they?
The problem I experienced with 1.905 happened with the old alpha.rsp as well as the new Doppler.rsp.
Mal is working on the problem and will post a fix.
Hi Decrepit, I have posted a revised version of RealSpeed to you that fixes the Alpha problem on the files you sent and some other files I have tested. Thanks for the tip on this one. Its also up on my website now.
Regarding the VMax on the GT11, it is very close to the one second max in the doppler data. I think teh VMax is an unfiltered doppler value which you can choose to see in RealSpeed in the Speed Graph. Note that it is so close to the normal doppler you will hardly pick the difference.
G'day Mal,
I sent the request with my PC number last week and have not heard back from you as yet.
I am trying to do the firmware upgrade from the cd but it is asking me for a password, I am wondering if this is what I need from you???
Cheers mate,
Regards GT.
Hey GT,
I have searched high and low and no unanswered emails except a couple from the USA and Israel. Even checked my spam folders...
Send it to me again tonight and I'll send a reg Id post-haste!
Tested 1.906 today and alpha's are now fine.
But I still have a problem saving to the gpx format, well it seems to save ok, but the subsequent file won't load back into realspeed and crashes googleearth.
quote:
Originally posted by TimeMachine
Hey GT,
I have searched high and low and no unanswered emails except a couple from the USA and Israel. Even checked my spam folders...
Send it to me again tonight and I'll send a reg Id post-haste!
quote:
Originally posted by MintoxGTquote:
Originally posted by TimeMachine
Hey GT,
I have searched high and low and no unanswered emails except a couple from the USA and Israel. Even checked my spam folders...
Send it to me again tonight and I'll send a reg Id post-haste!
Hi Mal,
Weird, I sent the email to mwright@intellimass.com this is correct?
No worries, I will resend tonight.
Thanks heaps, GT.
