The Electronic Charting "Zoom Problem"

[colemj]

Originally Posted by Dockhead

3. However, the size and resolution of the monitor enormously influences how well this works. My nav table 23" 1080 line monitor is not all that fantastic -- far less detail than a paper chart. Not all that fantastic, but usable. My bulkhead 4000 line monitor, on the other hand, gives you just about the same picture you would get from paper. So it may be that raster charts and better and better monitors may really replace paper completely, and I think it's already damned good.

6. The "zoom problem" is still a real problem, however, because 90% or more of cruisers use commercial vector charts, or CM93.

…...
"Doing the best you can" isn't really good enough -- surely we can provide somewhat better tools. All the detail is present in CM93 and other commercial vector cartography -- this is definitely not a question of "inferior or incomplete data sets" as someone above claimed. It's just a question of how it's presented -- how the decluttering scheme is implemented.

We are leading this horse to water, now we are having problems convincing it to drink….

3. Actually, it is going the other way - raster charts are on their way out and are being replaced by vector charts. Raster charts are nothing but reproductions of paper charts, and as paper charts no longer get made in the possible future, there will be nothing to reproduce from them.

6. Have you actually tried a real ENC chart, and not an out-of-date, stolen chart set that is not formatted to be used in the manner you wish?

Once again, ENC charts ARE vector charts, and their format is much different than CM93.

Why do you expect OCPN, or any other program, to spend time inventing complicated functionality to work with an out-dated chart format that is only available through pirated means?

While the nature of OCPN is one of casting a wide net of usable chart formats, they are correct in holding the CM93 issue at arm's length and treating it like a red-headed step child.

Unfortunately, the wide availability of CM93 causes people to conflate OCPN with CM93. Even on this forum, people talk about OCPN as a "chart" and indistinguishable from CM93 data.

Do us a favor and get yourself some real ENC charts (S-57 or S-63), play around with them, and report back if you are still having the "zoom problem".

I know you think I am being argumentative with you, but you really don't seem to know what is what here.

If I am wrong in the above, someone please correct me and I will gladly back down from these points.

Mark

[Sun and Moon]

Dockhead, I'd like to commend you on your restraint in addressing the moronic comments made by one poster in particular, who probably navigates about as well as he reads.

[colemj]

Thank you for the personal attack, but can you actually refute anything I posted? I said I was willing to retract an argument if I was wrong about it.

If you cannot present facts against my arguments and statements, can I expect an apology?

Mark

[transmitterdan]

Originally Posted by Dockhead

6. The "zoom problem" is still a real problem, however, because 90% or more of cruisers use commercial vector charts, or CM93. Since the problem really boils down to the method of decluttering used, surely this could be solved with more control over the decluttering process. To restate one aspect of the problem, maybe more clearly now -- the standard scheme of decluttering of vector charts is designed for small, low res screens. It knocks out too much detail too fast, in order to be able to use OpenCPN for planning passages in complicated water, on a larger, higher res screen -- the capabilities of the monitor is wasted. The "Chart Zoom/Scale Weighting" slider exactly addresses this issue, but it doesn't go nearly as far as we need. Is it possible to extend it?

7. The other aspect of the zoom problem is the disappearance of hazards when using large scale views, as discussed in the OpenCPN manual*. I think showing some kind of proxy for hazards as we've been discussing would be just terrific, if it's not too hard to program. Just some kind of symbol or blotch or shading which indicates there are problems which need to be looked at in that area. The makers of commercial chart plotters should think about this too, IMHO.

Dockhead,

I firmly believe that when you come back here a year from now and read this thread you will bang your head and say "what was I thinking when I wrote that?". If you genuinely want to get the most out of a tool you have to use it in the way it was intended to be used. You can't trim your fingernails with a lawn mower. Even though a fingernail clipper and a lawn mower cut things they are not interchangeable. Similarly, you can't refuse to use a tool and then whinge on about how it isn't useful.

This is what you should do for 30 days:

1) Don't look at a paper chart.
2) Don't look at a raster chart.
3) Plan 25 passages using ENC S-57 charts of current vintage. For example, plan a passage from the Delaware river to Virginia through the C&D Canal to lower Chesapeake Bay. Plan a passage along the intracoastal waterway. Maybe go from Seattle to Friday Harbor.

I am sure that if you just use common sense you can easily plan these passages on nothing but ENC vector charts. And you will do a much better job than looking at paper or raster charts.

Nearly all commercial vessels use these vector charts day in and day out for passage planning and real time navigation. If vector charts were not vastly superior to raster there would be a hue and cry to fix them. That has happened in the past and the vector charts now far exceed the detail and precision of raster charts. If ENC vector charts were a "hazard to navigation" commercial vessels would be grounding or colliding every day.

[Sun and Moon]

Mark,
Curious that you should interpret my comment as pertaining to yourself. But if it quacks like a duck [and calls itself a duck] maybe it is a duck.

It appears to me you have misunderstood Dockhead from the start.

Dockhead's basic premise is this: in extremely complex navigational zones, the task of route planning is difficult. If you zoom out enough to see the region, you can't see the individual hazards, you can only see that the region contains some. If you zoom in enough to see the hazards and the safe water between, you can't see enough of the big picture to navigate your way through the maze. The Baltic is one such zone. Dockhead speculated there may be other similar zones in the world. In the Baltic last summer, Dockhead resolved this problem using paper charts. Carsten, who sails the same region, said that's what he does too, and believes that is the overwhelmingly common practice in that region.

There have been several excellent suggestions posed that could assist Dockhead. For me, the most interesting is the two-window approach: open one with a big view, and another with a zoomed in view. This may be an obvious solution to a software engineer that works with multiple related windows all day long, but may not be obvious to others. And certainly not obvious that OCPN can be made to link the windows together.

Boat Alexandra made several suggestions which could assist, including at least one capability that is not yet implemented. And he's likely to push forward to make it a reality.

My thanks to all who have contributed.

[Dockhead]

Originally Posted by colemj

We are leading this horse to water, now we are having problems convincing it to drink….

3. Actually, it is going the other way - raster charts are on their way out and are being replaced by vector charts. Raster charts are nothing but reproductions of paper charts, and as paper charts no longer get made in the possible future, there will be nothing to reproduce from them.

6. Have you actually tried a real ENC chart, and not an out-of-date, stolen chart set that is not formatted to be used in the manner you wish?

Once again, ENC charts ARE vector charts, and their format is much different than CM93.

Why do you expect OCPN, or any other program, to spend time inventing complicated functionality to work with an out-dated chart format that is only available through pirated means?

While the nature of OCPN is one of casting a wide net of usable chart formats, they are correct in holding the CM93 issue at arm's length and treating it like a red-headed step child.

Unfortunately, the wide availability of CM93 causes people to conflate OCPN with CM93. Even on this forum, people talk about OCPN as a "chart" and indistinguishable from CM93 data.

Do us a favor and get yourself some real ENC charts (S-57 or S-63), play around with them, and report back if you are still having the "zoom problem".

I know you think I am being argumentative with you, but you really don't seem to know what is what here.

If I am wrong in the above, someone please correct me and I will gladly back down from these points.

Mark

"Argumentative” hardly does justice. It’s more a hard-bitten hostility to anyone who works or thinks differently from the way you do – like people who like paper charts, which is another really serious failure of the imagination. You are going to universalize your own narrow experience, and impose it on everyone else, or die trying, aren't you? Because you fail to imagine the problem and concepts described, however, does not mean that they do not exist.

CM93 charts are just CMap charts and work exactly like any commercial vector charts like Navionics. Their copyright status and freshness are completely irrelevant to this particular discussion, and mentioning this is mere obfuscation. Read the OpenCPN manual – “The bottom line is: Don't trust CM93. Always check with other reliable sources, meaning, for example, Raster charts (paper charts) Pilot Books, or ENC charts, not another privately issued vector chart, with similar problems ( for example Navionics).”. The OpenCPN authors are talking here specifically about the problem which you deny exists. But the problem not only exists, it is a fairly simple concept: the problem is simply how the screen is decluttered with varying zoom levels -- this scheme is designed for small, low-res screens, not screens large enough to do planning on.

Raster charts have different disadvantages, and I would not want them on a plotter for sure. Whether they are on their way out or not, I don't know, but they are readily available so far for European waters. They are the choice of those European (at least) publishers who provide cartography for PC navigation programs, and my guess would be that that's because they work better -- so far -- for the kinds of things people do on PC nav programs. The price is very reasonable for the UK, Ireland, Belgium and Netherlands (50 pounds or $75 for a complete set with updates through 2016), but for the Baltic you'll pay a couple of grand to NV for full coverage.

What the future holds for raster charts, I have no idea. Since all of the data you need for the big picture on a big screen exists in Navionics or CMap or CM93, I reckon it won't be that hard for someone to figure out how to regulate the decluttering process to suit big screens used on PCs, rather than the little plotter screens they were designed for, and make this widely-used cartography useful for planning purposes.

As to ENC vector charts -- they are not used by any single recreational sailor in Europe that I've ever run across. They are tremendously expensive (you rent them, don't buy them, for European waters) and apparently don't cover all the places recreational sailors go. Whether they declutter differently so as to be more suitable for big screens and passage planning, I don't know – could be, since ECDIS screens are much bigger than recreational plotters. But trying to work better with the normal commercial vector charts virtually all recreational sailors use is a worthwhile pursuit. Anyone who doesn't understand the problem, don’t experience it because of the kind of waters they sail in, or is not interested in a solution to it, is welcome not to read this thread. If you ask 100 people who sail in these waters if they have a problem with details dropping out when zoomed out, 100 will tell you they do. The real experts will have to decide whether there's anything practical which can be done about it or not.

[RhythmDoctor]

I agree with Mark that O's developers should not expend much effort facilitating the use of a pirated chart format known to have serious inaccuracies.

However, I am not so sure about this statement:

Originally Posted by colemj

3. Actually, it is going the other way - raster charts are on their way out and are being replaced by vector charts. Raster charts are nothing but reproductions of paper charts, and as paper charts no longer get made in the possible future, there will be nothing to reproduce from them.

What evidence do you have that rasters are "going away"? NOAA no longer prints them, but they still issue the digital raster files for 3rd party printers to use. Basically they are outsourcing the hardcopy imaging to the private sector, like virtually everything else these days (including space flight). And there are a lot of regulations requiring boaters of various categories to carry printed charts, so the accompanying rasters will continue to be issued.

As I've stated elsewhere, I think rasters are far superior for planning on the desktop, because professional cartographers have decided what to show, and so the charts are more reliable than the ENC display that often requires right-clicking on an object to see the details.

But while underway I generally prefer vector charts in course-up mode, partly because with OpenGL activated, the depth soundings and other text rotate to always be right-side up no matter how the chart is rotated.

These are general preferences - there are some times when I pull up ENC for planning, or RNC while underway. In fact, I'll always do a final review of a route over top of both RNC and ENC, since it's most prudent to use multiple sources. Sometimes I'll even export it to Garmin Bluecharts to view it over my (far inferior) Garmin cartography. There are also some times when I run two instances O, with RNC in one window and ENC in the other. It's great that O offers all these options.

[sailorF54]

Originally Posted by colemj

Do us a favor and get yourself some real ENC charts (S-57 or S-63), play around with them, and report back if you are still having the "zoom problem".

Mark

Pity you did not do us that favor...
Here are a CM93 and an official S63 bit of chart from the same area, both rather large scale.
How long will it take you to tell which is which ?
Can you spot major differences ?

For very small scale (bluewater long range planning), commercial vector charts (including CM93) are hopeless. I have long advocated using electronic RASTER charts for that purpose.

Are you aware that in most countries (the US could be different) S-57 / S63 do NOT cover the sailors' favorite playing grounds, as they are geared towards commercial harbours and fairways ?
Here is the most detailled S-63 view from several very popular anchorages (Glénan, south Brittany, Atlantic France). Compare with the CM93 chart.

CM93 are (were ?) made by a reputable company. The fact is that there are tens of thousand of sailors using them all over the world. If they were that bad, don't you think we'd heard a massive outcry long ago ?

[Dockhead]

Here's the raster chart of Poole Bay. There are about 12 miles between the Needles and the entrance to the Swashway.



Note that the Swashway Channel into Poole, with all of its buoys, isn't visible at all in this view -- so it actually has less detail than the Navionics vector chart at the same zoom on my plotter.


Using the piano keys and the "Hide This Chart" command, however, you can force OpenCPN to show the smaller scale charts, in this case quilted together since we have small scale charts for the whole area:



This is really cool! On my 1080 line 23" nav table monitor, you can't clearly see all the detail. But for planning, it's just the thing -- you can't miss anything. It's not fatal for this particular purpose that the view is over-detailed.


For comparison, here's the Navionics vector chart on the Zeus:



It's a beautiful picture, so different from the paper-like raster chart. More cartoon-like. It even has a visual effect to give an impression that you're looking at the seabed. But it's deceptive.

It's interesting that there is a lot more detail here than was presented on my previous nav table plotter, a 10" Raymarine RL80CRC+ with CMap charts. On that, you couldn't see any of the buoys or recognize the channel, until zoomed in several steps more than this. I wonder if this means that either the cartography or the plotter software is decluttering less aggressively, taking into consideration the fact that plotters are getting more resolution.

[Dockhead]

Originally Posted by transmitterdan

Dockhead,

I firmly believe that when you come back here a year from now and read this thread you will bang your head and say "what was I thinking when I wrote that?". If you genuinely want to get the most out of a tool you have to use it in the way it was intended to be used. You can't trim your fingernails with a lawn mower. Even though a fingernail clipper and a lawn mower cut things they are not interchangeable. Similarly, you can't refuse to use a tool and then whinge on about how it isn't useful.

This is what you should do for 30 days:

1) Don't look at a paper chart.
2) Don't look at a raster chart.
3) Plan 25 passages using ENC S-57 charts of current vintage. For example, plan a passage from the Delaware river to Virginia through the C&D Canal to lower Chesapeake Bay. Plan a passage along the intracoastal waterway. Maybe go from Seattle to Friday Harbor.

I am sure that if you just use common sense you can easily plan these passages on nothing but ENC vector charts. And you will do a much better job than looking at paper or raster charts.

Nearly all commercial vessels use these vector charts day in and day out for passage planning and real time navigation. If vector charts were not vastly superior to raster there would be a hue and cry to fix them. That has happened in the past and the vector charts now far exceed the detail and precision of raster charts. If ENC vector charts were a "hazard to navigation" commercial vessels would be grounding or colliding every day.

Yes, maybe -- my mind is open and I'm trying to improve my methods. I hope that a year from now, I'll read back through this thread and think how little I knew. That would be exactly the result I'm looking for.

I never said any charts are hazards to navigation, and I specifically have no experience with ENC charts. Do you? Are they readily available for your cruising area? I'd like to try them, but they are not practical for our area, and no one uses them. Maybe you could make some screen shots?

I have nothing against vector charts -- on the contrary, a collection of Navionics Gold and Platinum cards updated almost weekly is my primary source of navigation data. They are brilliant in my plotters. But they don't work well for planning in complicated water. I'm looking for better ways to do this electronically. I think features could be added to OpenCPN to give us more control over the decluttering process, which would make even these humble recreational vector charts work for a broader range of purposes. I could be wrong.

As to planning a passage through the ICW -- this is not the task I'm trying to solve. Following buoys along the ditch (I've been there) is trivial, and says nothing about either tools or methods. I am, like others, shaped by my own narrow experience -- I sailed 3000 miles last summer from the UK to Russia and back, through a literally million rocks and islands (I even managed to hit one ). My tools and methods, which have served me well over decades of sailing in a lot of different places, failed me there. I am trying to improve them -- that's partially what this thread is about.

[Dockhead]

Originally Posted by sailorF54

CM93 are (were ?) made by a reputable company. The fact is that there are tens of thousand of sailors using them all over the world. If they were that bad, don't you think we'd heard a massive outcry long ago ?

I have a few thousand miles of experience with CM93 charts in the waters of 10 different countries. I acquired them as a kind of a base map, and have been really surprised at how good they are. I have nowhere seen anything terribly out of date, and the detail is usually comparable, and sometimes better, than that of my expensive Navionics charts. I wouldn't use them as a primary source, but for all other purposes they are brilliant.



By the way, a bit of thread drift, but the raster charts I bought today from visitmyharbour.com are officially FOR NAVIGATION. No Mickey-Mouse disclaimers.

[Dockhead]

Here's a really striking example of the lost detail problem.

This is Selsey Bill, the first big headland you pass on the way from the Solent out the English Channel towards the East. This is what you see on the Zeus:



What a striking contrast to the raster chart:



Even if you zoom in the Zeus, you don't see the crucial pass:



Only at this zoom level do you understand:




An amazing difference. The raster chart gives you all the information at a single glance, even from a further zoomed out view than the first Zeus screenshot. All the way from Portsmouth to beyond Bogner Regis, a complete picture.

This is a revelation for me.

And how beautiful the presentation of the Navionics chart is. A lot of thought and effort has gone into devising this view. This is kind of like -- a technical drawing -- the abstraction -- is always 100x more useful than a photograph. Of course, this is exactly what you want on your plotter, just for a different purpose. It's like a cartoon.

[Dockhead]

Even at this zoom level -- showing a good part of the Channel, all the way to Brighton! You STILL see the pass below Selsey Bill.




A really huge difference to the vector solution.

The same zoom level in Navionics:

[transmitterdan]

Dockhead,

Get the US ENC charts. There is a plugin for O in beta testing that downloads these and other free charts semi-automatically. If you are interested send me a PM and I'll help you get it running on your system. There are about 6GB of free ENC charts for all US waters.

[Dockhead]

Originally Posted by transmitterdan

Dockhead,

Get the US ENC charts. There is a plugin for O in beta testing that downloads these and other free charts semi-automatically. If you are interested send me a PM and I'll help you get it running on your system. There are about 6GB of free ENC charts for all US waters.

OK, thanks, I'll give it a shot. Do you use them? On plotter? Or PC? For planning? Please comment.