bikedaft
Well-Known Member
here is a far better explanation than mine. from Tom Cunliffe... (2016 article, raster and vector)
How to use vector charts safely
How to use vector charts safely
Keep off the rocks!
- Thread starter RichMac
- Start date
GHA
Well-Known Member
He's missed a lot though.....
Sat images, absolute gold dust cruising where charts might not be as accurate as one would like. Tem Vestas had as SOP - download google earth images for the entire route and check route planning & route changes with sat images, they didn't do either before the passage when they ran aground, a few minutes downloading & sticking to their own rules would have saved the boat. SasPlanet lets you jump with keystrokes between various sat images & a coupe of marine chart sources, again gold dust to check datums & chart accuracy.
OpenCPN 5.0 / mbtiles - PANAMA POSSE – Central American Rally
http://gdayii.ca/Downloads/Sasplanet.zip
dunedin
Well-Known Member
Read my post more carefully ..... “I suspect all chart source data is now vector”.
Clearly all paper charts and some electronic paper charts are converted to a fixed image for presentation (ie for printing paper charts or for creating the electronic images distributed as raster charts). But I believe most of these are now generated from chart data increasingly held by the chart owners in sophisticated vector formats. So the difference is whether the conversion from vector to pixels is done by the chart supplier’s software or the plotter software.
dunedin
Well-Known Member
There is a lot of useful stuff in there. But in the detail of the article it is still consistent. The issue that caused some boats to hit reefs when not zoomed in, related to how the software handles zooming. Not an inherent issue of vector charts.
But Tom refers loosely to the issue being the “vector charts”, not the “software presentation of vector charts when zooming”, which would be more accurate but less catchy to a journalist.
If you have a square mile of sea, with depths ranging from say 50m to drying 2m (ie the reef), when zooming in any sensible software designer should look at the highest risk part of the area being summarised and represent that as the summary. Hence show the 2m drying as the summary, not the 50m clear water. However, for reasons I cannot understand the dangerous opposite approach was adopted by some. That is the root issue. And could be fixed very quickly, without changing the vector charts.
[70521]
Well-Known Member
May we enquire on what basis you are making that argument?
Many of the posters here have vast experience and in any industry "best practice" is the distillation of the thoughts of many. RicMac has one opinion, many here have a different opinion and a public forum is really not the place if you want to change things. I suggested to RicMac in a post that he contact all the organisations who develop training material and start discussing the issue with them. He appears to have got the hump and left the building.
lw395
Well-Known Member
Until this is fixed, vector charts don't really seem fit for purpose.
Back when chart plotters started coming on the market, the whole point of vector charts was said to be that symbols indicating hazards etc would stay legible at any level of zoom, whereas if you zoom out on a raster chart, the hazards would become just one pixel and unnoticeable. Raster charts get over that by changing to a new raster at a different scale.
The trouble is, people want all this stuff to be cheap, a set of paper charts is a lot of cash. Electronic charts used to be very expensive.
AntarcticPilot
Well-Known Member
This is entirely accurate, and it's even simpler than that - all HOs use the S57 vector format (or possibly a successor - it is a while since I was in direct contact with charting), which stores vastly more information about the provenance and history of features than leisure chart-plotters do (ECDIS systems on ships also use S57, I believe). There may be some areas of the world where charts are still based on engraved copper plates or scribed films, but these are few and far between. They are the only places where raster data are the source for electronic charts. Otherwise, ALL charts start off as vector data; raster data are second-generation products from the vector data. Back in the 1990s the debate between raster and vector formats for map and chart data was sensible; vector won and is the preferred format for all mapping and charting these days; I personally have NEVER (since 1979!) used a raster format to create a map, though I have to admit that this was moot until the late 1990s when printing houses started to be able to use vector data directly! Raster data is still important when using image data such as satellite images or aerial photos, but for mapping purposes you embed an image in a vector representation of the non-image data.
While checking one's route on Google Earth is a wise precaution, is does suffer from drawbacks - first of all, it assumes that Google have acquired ALL the images that may be useful, and it also assumes that they are correctly positioned. They are likely to be better than pre-GPS survey data, but still potentially hundreds of metres off. There's also the issue of searching along a track hundreds of miles long, the vast majority of which will be bland blue colours! It is surprisingly difficult to (for example) zoom in on Ascension Island even if you know where it is - of course, named features can be searched for, but Google doesn't usually name underwater features!
Finally, when off the beaten track it is unwise to assume that ANY source will necessarily have captured all the potential hazards. I have no doubt whatsoever that there are uncharted reefs and islands in less travelled areas, and there are also parts of the world where volcanic action can change things literally overnight! The Polar regions have very large areas where the marine side of the mapping might just as well have "Here be Dragons!" as anything else on it!
Just as an aside, Francis Drake found shelter for his ships in the vicinity of Cape Horn at a place he called Elizabeth Island. He even produced a sketch map of it. But there is no island that matches his description in the area. There are many theories about it, but there's at least a possibility that it was a volcanic islet that has now eroded away. Even quite substantial places can come and go!
Dave100456
Well-Known Member
Thanks for posting the video, I enjoyed watching it and your presentation style. It’s not perfect but who/what is? For me, it refreshed some of the good and bad aspects of digital navigation. I hope you’ve not been put off creating and posting.
cheers
cheers
GHA
Well-Known Member
They can? Haven't heard of that before, the satellite images I've used have always been bang on to the GPS. Bang on ehough for a boat anyway. Though that's not often been in the middle of nowhere. SasPlanet shines here as well as you can keyboard shortcut between satellite sources, google won't always have the best resolution, sometimes bing is better, or yandex. Can you think of a spot where google earth is out? Might be interesting to see how the different sources match up.
Ascentian actually shows up on all the SasPlanet sources, just. Pale brown dot....but enough to get you to wake up and take notice. Or should be if you're out there.
AntarcticPilot
Well-Known Member
Yes, they can, depending on the satellite used. If it's old Landsat without any ground truth, they can easily be more than 100m off. They don't use data that bad in many places, but they do use it - especially off the beaten track. Google will also only have data for places known to exist!
Most of Antarctica is mapped with a precision of about 15m but an accuracy of about 100m.
GHA
Well-Known Member
Can you think of a lat and long somewhere a boat can go? Done quite a few evening virtual cruises on Sasplanet . Seems charting datums have got much better to not that long ago, sailed up a few main streets before
AntarcticPilot
Well-Known Member
Well, boats do go to Antarctica! But I wouldn't rely on absolute positioning better than 100m anywhere remote; Polynesia for example.Can you think of a lat and long somewhere a boat can go? Done quite a few evening virtual cruises on Sasplanet . Seems charting datums have got much better to not that long ago, sailed up a few main streets before
It isn't about datums, though that has been a problem which, as you say, is decreasing with time as mapping agencies replace older maps. It's about the accuracy of satellite orbits, which have also improved with time, but not as fast as the resolution has improved. It depends on several factors such as the accuracy of the orbital model. It's also about the pointing accuracy of the imaging device - again, improving all the time.
However, especially in remoter areas, much of Google's cover is from low resolution Landsat series satellites, which has the accuracy I stated unless corrected using ground control points. I don't know the source you quote, but it will use data from public sources and must have the same issues.
bikedaft
Well-Known Member
Sorry see what you mean.
The issue is how they are displayed, not the source data.
Will ask my programmer friend why they seem unable to display the shallowest depth when zoomed out, when display is vector chart. Decluttering menu choices have a part to play as well?
laika
Well-Known Member
This must be something of a cause celebre in Perth WA if someone else from the OP's hometown feels passionate enough to join the forum just to comment.
The OP has added to his youtube site:
No-one minds someone making a video. We wouldn't be posting on PBO if we didn't appreciate the desire to share knowledge. However some of us are puzzled by the claim that the RYA don't teach electronic nav as they've been doing it for years (though certainly not 25). He seems to have latched on to an earlier post saying they were currently updating the electronic nav bits and misinterpreted this as a statement that they did nothing before.
In an earlier post I noted that the OP says his certificate is an IYT one. The IYT are a separate body from the RYA and I don't think too many of us in the UK are familiar with their course content so maybe the OP's ire is directed at the wrong organisation?
longjohnsilver
Well-Known Member
that was boring, stopped watching/listening after about 60 seconds.
GHA
Well-Known Member
Still can't find any areas where the different sources don't match up to within a few M, could they all be out? Cmap, navionics, bing, google & yandrex. Though chart data are a bit sparse.
Anyone know of somewhere where the charts are way off position? Or sat images?
So what would you rely on less than 100m? Not charts , definitely guilty til proven otherwise, from experience sat images IMHO as good as anything with a jolly good eyeball when moving
dom
Well-Known Member
Still can't find any areas where the different sources don't match up to within a few M, could they all be out? Cmap, navionics, bing, google & yandrex. Though chart data are a bit sparse.
Anyone know of somewhere where the charts are way off position? Or sat images?
So what would you rely on less than 100m? Not charts , definitely guilty til proven otherwise, from experience sat images IMHO as good as anything with a jolly good eyeball when moving
Purely out of curiosity, where on earth are you sailing and what are you up to for such fine tolerances in such far-flung places to be so relevant?
doug748
Well-Known Member
This sailor reckons the San Blas area off Panama is dodgy, see 8 mins in:
Says his charts are Open CPN.
Later in the piece he expands on this, though whether his inaccurate charts are current, old or home brewed by fellow mariners is not clear.
Says his charts are Open CPN.
Later in the piece he expands on this, though whether his inaccurate charts are current, old or home brewed by fellow mariners is not clear.
Graham376
Well-Known Member
Still can't find any areas where the different sources don't match up to within a few M, could they all be out? Cmap, navionics, bing, google & yandrex. Though chart data are a bit sparse.
Anyone know of somewhere where the charts are way off position? Or sat images?
So what would you rely on less than 100m? Not charts , definitely guilty til proven otherwise, from experience sat images IMHO as good as anything with a jolly good eyeball when moving
There's a big discrepancy between Google Earth position of my mooring and gps, if I followed GE lat/long, I would end up high and dry. C Map also shows me over the banks when I'm in about 4m of water.
bikedaft
Well-Known Member
W Coast of Scotland, various places charted 100m out - both vector, raster and satellite images. but it is improving a lot over the last 30 years. it can be quite variable between the various info sources. but never quite believing any of them 100% usually helps...Still can't find any areas where the different sources don't match up to within a few M, could they all be out? Cmap, navionics, bing, google & yandrex. Though chart data are a bit sparse.
Anyone know of somewhere where the charts are way off position? Or sat images?
So what would you rely on less than 100m? Not charts , definitely guilty til proven otherwise, from experience sat images IMHO as good as anything with a jolly good eyeball when moving