Looking to get a datalogger

[Prodnose]

The NMEA standard was originally written for marine use. The "M" stands for "marine". It pre-dates GPS and was produced to interconnect radars, echo sounders, speed logs etc. For GPS, it contains information on number of satellites, estimated error, whether WASS/EGNOS corrections are being received as well as lat, long time etc. That sentences repeat information is, I guess, is due to different manufacturers wanting a particular sentence to meet their needs.
Some uses may require time down to milliseconds. The addition of ms is relatively recent. In the SG-289 example, I assume that the .000's are added merely to comform to the standard. The 5 second interval is the fastest set by the maker, ZyCraft, not by NMEA.
As to whether space used vs convenience and simplicity is most importance it look as if the NMEA committee came down on the side of the latter.
I am quite certain that my (and all other) hand held GPS only stores what it wants and in a economical form.
The convenience of the data logger storing a couple of the NMEA sentences is that I can feed it into software written for NMEA. In practical terms, if I save both (optional) at the maximum frequency and run it continuously, the SD card will store more than a year's data.

[mfrebuilt]

Well, I would have done it differently, I really can't stand inefficiency. As for simplicity, it's really not much harder to read numbers binary data than text data. But I suppose when you program in C you actually have to understand a bit about what's going on, unlike other languages where you can just happily ignore it.



Edit:
I've somehow overlooked this GPS which, as far as I can see, fits my specifications.
http://www.pocketgpsworld.com/qstarz-travel-recorder.php
User manual: http://www.qstarz.com/download/BT-Q1000UserManual-V2.rar

I've found others capable of 1Hz data logging but one didn't have a rechargeable battery and the other didn't record altitude. The Qstarz has a rechargeable battery and there's stuff in the user manual about recording height, so I assume it can log altitude. Would someone mind checking the specs for me? I want to make sure I'm not missing anything obvious (the no altitude GPS nearly fooled me..) The only odd thing I see is the MTK chipset, most of the others I've looked at seem to be sirf III.. no idea which is better

[Prodnose]

Well I'm sorry that you can't stand inefficiency. Life must be very difficult.

Other languages are quite capable of reading and writing binary code. I have limited ability in programming but quite capable, for example, in reading NASA topographical data, swapping byte order and employing the data in programmes written for my interest and amusement. More gifted amateurs and the professionals won't be phased by it.

[robertn]

The data you are collecting and processing is not likely to be as good as you are expecting. Have a look at a previous post (In the sports forum I think) from a few years back read using GPS for skiing.

Problems are :
GPS Altitute - no good, It's OK if you want to know if you are on top of a mountain or on the valley floor, (provided you know the error in the elipsoid vs the mean sea level). It's inaccurate for anything else, as the constallation of satalites is all above you. If you could pick up the ones on the other side of the earth, it would be great, but the earth kinda blocks the signals (and if they were strong enough to get through, everyone on the other side would be fried).

Speed / Distance etc
this is reported as if you are level (traveling on the surface of the elipsiod), so the error when skiing down a 45 degree slope is about 50%

Changes in velocity (turns and stops)
You tend do this quite a lot skiing, Unfortunately the GPS will usually smooth these sudden changes to motion. Try this, Cruise along at 60mph in you car, stop suddenly (not on the M1). Watch the GPS speed report, it will usually take a few more seconds to get to 0. SiRF GPS's have a number of options that affect this, but most consumer devices do not allow it to be changed, and usually default it to automotive modes, and cars just can't turn and stop as quick as skiers.

So what can you get from the data for a days skiing - number of runs, rough altitude of runs, very rough average speed (40 - 80% of actual would be my guess -40% for a gun skier, 80% for a beginner), - lots of really cool stuff to show you mates, who won't understand any of this....

Plot them against a skifeild map, and you can get some more detail.

[robertn]

Oh - and for you discussion on inefficency (I am also a C programmer of too many years experience) - I hate it as well, but much much less so than closed, propritory systems.

Without inefficent, easy to understand, human readable NMEA the entire GPS industry would be 10 years or more behind where it is now, if there was such a thing as consumer GPS.

Without HTML where would the internet be?

[mfrebuilt]

Thanks for the info robertn. I have heard that GPS altitude should never be trusted, but I was hoping that the change in altitude (which is what would be more important to me) is more accurate. Do you have any idea how accurate the change in altitude might be? The device I found isn't even $100, so I think I'll just purchase it and see how it goes. Worst case scenario I'll be able to track my latitude and longitude on the mountain and display it on Google Earth.

[robertn]

I have played around with altitude a very small amount - I found that with the GPs stationary, it varied by around 10 meters over a period of a few minutes.

You may find that the rate of change is not too bad (particularly over a short space of time when the satalite constalation and atmospheric distortion is near enough to constant ).

I think you should have a crack at it and see what happens, after all, whats the worst case scenario - I bet less spent on toys that sit in the bottom draw than me .

[Duddy]

I use several GPS Dataloggers at work to record journey times. I currently use Globalsat DG100 a good unit (google search for photo's etc) I use 2 second intervals but it will go down to 1 sec. Uses 2x AA recharg batteries 2500 ma claim to last 24 hours but I get 32 hours max
Downloads in various file formats (1) Exploring to KML ( Interface with Google Earth)
(2) Show recorded data on Google Maps
(3) Text format
(4) Excel file format
(5) RMC Compatible format
(6) GPX format
Above is paste from manual.
I,m out and about in Manchester UK tomorrow so I can record a short route for you to play with if you want. Let me know which file formats you want
_________________
HUAWEI P30 Pro (new edition) with Speedtrap Alert & alcatel1 for SatNav
CoPilot 10 with CamerAlert
RoadHawk in-car video
Reading glasses getting thicker as is my waist
Retired but want to go back to work for a rest.

[NickG]

mfrebuilt,
Please note that they've just released an updated version of the Qstarz Q1000 recorder which has better software and twice as much RAM. It's called the "Q1000P Platinum". It has only JUST been released in the last week or two so it's difficult to find.

Press release
http://www.qstarz.com/news/news-20080205.htm

I bought mine from here, but you might not be in the UK?
http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=150220675640

I've only had it a couple of days but so far it seems to work very well. I've tested it in the car and the tracks I've got are stunningly accurate (I can email you samples if you like).

I too intend to use it to log my skiing and I'm off next week to the French Alps so I can give it a good test run. I need to figure out which settings will record in good detail while enabling me to record a week's worth of skiing. I've already decided that a 5 second interval is not enough, but 1 second will mean I run out of RAM, so I'm thinking that 2-3 seconds might be optimal with recording set to trigger above 3mph (to cut out standing around at the top of the blacks thinking "oh my god, I can't do down there!).

I'm also a software developer myself (.NET platform mainly, but I understand C and C++). What software are you intending to write? There is an enormous amount of GPS related software already out there that might save you some work. If you give me an idea of what you're going to write, I might be able to point you in the direction of some helpful code samples etc. I think for example, there are examples/libraries available for decoding NMEA on codeproject.com.

I realise NMEA is inefficient and I'm not really sure why it's used internally in some loggers. It's a shame that a more compact binary format isn't used instead. I also don't get why loggers can't log to a microSD card. If we had 1GB of ram (£10 off ebay) we could set it to 1 second (or even more often) and not have to worry about running out of RAM.

Nick...

[philpugh]

[NickG]

The 15 second interval is way too long for skiing though. I could have done about 5 turns in that time and it would render them as a straight line if you log at 15 seconds. This would mean distances are very inaccurate as it would only take the mean path rather than the true distance skied. I think for skiing and snowboarding you need no more than 2 seconds at the max and preferably 1 second or greater. I saw a track recorded at 4Hz (4 times a second) on the web and it looked excellent - you could see each turn turns individually and as a smooth curve. No chance of that with the Q1000 unless you have the ability to offload the log files each day to a laptop.

I've just done my calculations again and the 200,000 track points at 1 second should mean that it has 55 hours of logging. More than enough for a week's skiing as that equates to about 9 hours per day (assuming a 6 day ski week). However I think that figure drops to 135,000 points if you choose to log altitude information. I realise it's inaccurate, but it's better than nothing.

Apparently there's a way you can insert altitude information afterwards using data from NASA but I'm not sure how you can do this. Basically, NASA have logged the terrain height for the whole earth at high resolution and this data is public. That means, if you know you're on the ground and you know where you were, you can calculate the altitude fairly accurately. Does anybody know any websites or programs that let you do this? I've seen Geotagging applications that do this, but nothing that can add this data to an entire log file.

I have a feeling that www.sportsdo.net does it.

Nick...

[philpugh]

You are talking about the NASA Shuttle radar elevation data (SRTM).

This is used by the Scottish Mountaineering Club to generate GARMIN compatible contour 'maps' for the UK (see http://www.smc.org.uk/ContourMaps.htm) for details.

The resolution of the data is 3 arc seconds for most of the world (about 90 metres resolution). Continental USA has 30 metre resolution.

Do a search for SRTM on the web.

This site ([url]http://srtm.csi.cgiar.org/ [/url]) shows what can be done.

Its not a straight forward process as you get a matrix of spot heights (at 90 metre intervals) and you then have interpolate to get the estimated height at your position.
_________________
Phil

[mfrebuilt]

Thanks for the update Nick, I was rather surprised to see activity in this thread, I was just looking over it again before I was going to buy the Q1000.. I might have to see if I can track down the Q1000P now. Regarding the libraries that might be available, I think I'd like to give it a shot doing it by myself first as NMEA seems to be dead easy, I know there are already applications out there, but I'm a computer science student so it should be good practise for me. I haven't actually done any 3D graphics yet (and I won't be doing any at university, academically I'm going to focus on algorithms and computation) so this is the project I needed to encourage me to learn graphics in my own time. It will also allow me to program in some features which only a skier would find interesting

I'll definitely look into the height matrix thing, although if the resolution is 90m interpolating would be a bit sketchy where you've got sharp changes in gradient (lots of those on a mountain). I could use it to verify the accuracy of the GPS altitude recording if I happen to ski close to one of the points that have been accurately recorded, you could also see whether the error drifts or if it stays relatively constant.

[NickG]

[NickG]

I've had more of a chance to play with my Q1000 Platinum now and I'm quite impressed with the performance of the MTK chipset. When the logging is set to 1 second, the track hugs the roads on Google Earth with amazing precision - almost to the level where you can see which lane you were driving in. Like any GPS, it has good days and bad days - all depending on how many sats are in view, where they are (eg you don't want them all directly overhead or mostly to the east etc). But the results I've got while testing this week have been very good. I think skiiing is going to require the log interval to be set to 1 second. 3 seconds and 5 seconds is definitely not enough. I think I will get about

It's a shame it doesn't even ZIP the data as NMEA data compresses REALLY well.

We see you’re using an ad-blocker. We’re fine with that and won’t stop you visiting the site.
Have you considered making a donation towards website running costs?.
Or you could disable your ad-blocker for this site. We think you’ll find our adverts are not overbearing!