Feature Requests

[bdbcat]

All...

I've studied the source code for zyGrib. Impressive piece of work....

One thing bothers me about zyGrib: The direct download function is very handy. However, I find from reading the source that it is accomplished by logging into the zyGrib server (possibly through a proxy) and executing a php script on that server to fetch customized data. There is a login and password to the zyGrib server hard-coded into the source code. Presumably, the zyGrib server is getting the data from a reliable source, like NOAA, and repackaging as needed.

I suppose UGrib works the same way.

Maybe its just me, but does this bother anyone else in the same way?

Generally, why do I need or want a third party between me and the data? How does zyGrib serve up this data or free, and will they always do so? Why are they doing this? Ok, Ok zyGrib is Open Source, and they are all around good guys.

But what about Grib.US? This is closed source. And they are hiring. What could be their business model?

Paid-for grib services I understand.
Winlink I understand, too, since HAM radio interfaces are by definition non-commercial.

But I just don't get Grib.US and zyGrib.....

Maybe this is part of "cloud computing" that is the newest buzzword.

Or maybe I'm just being paranoid...

I also find that there is a method of accessing NOAA databases directly for GRIB data. Could we/should we do that?

My current inclination is to skip the direct downloading function in opencpn, and simply read and display GRIBs from a specified directory, leaving the getting and validating up to the users choice of server.

Comments?
Dave

[jerrymc]

I suspect (not based on any valid knowledge) that the GRIB data is readily available from NOAA, for anyone, but that it is probably a binary or data only download. That would leave it up to whomever wishes to use it to convert into a graphical format, which could be read by a program/application. That may be where zygrib, grib.us come in.

Below is from the "About" tab in the grib.us software(bolding is mine):

Ugrib utilises .NET assemblies from the NPlot project, whose license details are shown below.

NPlot - A charting library for .NET Copyright (C) 2003-2006 Matt Howlett and others. All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

blah, blah, blah, blah.....

jerrymc

[Christian Van H]

Originally Posted by bdbcat

My current inclination is to skip the direct downloading function in opencpn, and simply read and display GRIBs from a specified directory, leaving the getting and validating up to the users choice of server.

Comments?
Dave

I fully understand where you are coming from, and there are plenty of ways to access GRIBS. The only thing I would miss is the graphical way you can select your GRIBs area of coverage before downloading in programs such as ZyGRIB. As the SailMail link suggests, it isnt too hard to request the correct coverage via text though...

[LeaseOnLife]

Dave

I support this approach, the user supplies the grib file

In my post about sources of grib files I forgot to mention: I also can get grib files from Maxsea and in the past Raymarine offered that service too (they gave up though). I believe Furuno has a similar service.

I think we need to collect sample grib files with different types of data from all major providers. I do think there are differences, Ugrib doesn't display some of my winlink gribs...

Let us know what files you need, the group should be able to cover many sources and data types.

[idpnd]

Originally Posted by bdbcat

One thing bothers me about zyGrib: The direct download function is very handy. However, I find from reading the source that it is accomplished by logging into the zyGrib server (possibly through a proxy) and executing a php script on that server to fetch customized data. There is a login and password to the zyGrib server hard-coded into the source code. Presumably, the zyGrib server is getting the data from a reliable source, like NOAA, and repackaging as needed.

The purpose of the zygrib server is presumably proxy/cacheing.

I suppose UGrib works the same way. [...] I also find that there is a method of accessing NOAA databases directly for GRIB data. Could we/should we do that?

The data can be requested freely from the NOAA servers as described here. It may be possible to semi-integrate the logic from zygrib; their license is certainly more liberal than ugrib, whom I find spookier (have to have an account, e-mail harvesting, etc). I would not expect zygrib to have ulterior motives; they're just geeks providing their expertise freely (like your good self ).

My current inclination is to skip the direct downloading function in opencpn, and simply read and display GRIBs from a specified directory, leaving the getting and validating up to the users choice of server.

If one could define a source directory where opencpn automatically loads the most recent grib file, these could be sourced via zygrib/ugrib. But I do feel direct access to the NOAA servers would be preferable.

[Viking Sailor]

Dave,

Your idea of using OpenCPN to display Grib files from a selected directory works for me. That way when I have an internet connection I can use zyGrib for downloading and when I am at sea I can use winlink for downloading.

Thanks,

Paul

[sinbad7]

#83 Why re-invent the wheel when there are so many free grib programs available? They have no doubt spent a lot of time perfecting the software and have crossed most of the pitfalls. Why not allow the users to set up a link to their favorite grib program?

It would of course be great to view the weather overlaid our real charts but then again isn't that asking too much?

Personally I think the task of creating a calibration utility for raster files are much more important. Converting raster charts to .kap is very time consuming and fickle and demands huge files as opposed to compressed .jpg files.

[BillyDoc]

Yesterday I asked on the Help/FAQ thread about determining magnetic deviation for "myship" using openCPN and received three excellent responses. Apparently magnetic deviation information is not a part of the ENC chart information, and any competent sailor can determine it locally by setting a course using the GPS and then noting the variation directly from the compass.

On the other hand, sometimes the GPS doesn't work, but openCPN does . . . which makes this a little more difficult.

The answer to this is at hand, however. In his response to me yesterday GordMay provided a link to the DoD World Magnetic Model (NGA: (U) DoD World Magnetic Model 2005 (UNCLASSIFIED)) and I found when going there that the model is programmed in "C" and is in the public domain. From a cursory look at the model it would seem that it could become an openCPN sub-routine with little effort that was called with a button-click that passed the "myship" lat/long variables and returned the deviation.

This is definitely a low priority request, but it would be a nice touch nonetheless!

BillyDoc

[manimaul]

Originally Posted by BillyDoc

From a cursory look at the model it would seem that it could become an openCPN sub-routine with little effort that was called with a button-click that passed the "myship" lat/long variables and returned the deviation.

The RMC nmea sentence provided by a GPS contains local deviation already. Perhaps a feature request would be a customizable display console with local variation display as an option.

Will

[LeaseOnLife]

Originally Posted by manimaul

The RMC nmea sentence provided by a GPS contains local deviation already.
Will

This really depends on the GPS system. Some GPS receivers have built in look up tables and provide a local deviation. But some GPS receivers don't do that, the deviation field is empty.

And deviation is changing, slowly, but surely. Having opencpn using it's own lookup table, updated more often, would be a good feature.

[jan eulen]

Hello, I would like to make a request for a feature in opencpn regarding route information. For a sailing boat the speed to a waypoint (VMG or waypoint closing velocity) is very important information along the SOG of course. Is it possible to implement this information in the routedata window?

thanks

Jan Eulen
Netherlands

[LeaseOnLife]

Hi Dave

Playing with gribfiles to plan a crossing I thought I post a screen shot. This is about the minimum I think one needs. This is a screenshot of "Viewfax", part of the airmail2000 package. From the grib file is displays:

1. Windbarbs: color and feathers show strength
2. isobars (black lines)
3. wave height (colored background)
4. precipitation / rain (black/gray background)

A mouse over any point shows a box with wind, wave height and barometric pressure.

For the own ship, it lets me enter date, position, course and speed (like I am using it right now from an arm chair), or it can use actual date/time, position course and speed from the GPS.

With the left and right arrow keys I can easily shift time and see the fronts moving.

Dirk

[manimaul]

Single click georeferenced sat-photos. Here is how it could work:

1. An optional toolbar button (internet connection required) is pressed.
2. Opencpn scrapes google maps for png sat photo tiles in current view, joins them into single png and creates .KAP header with accurate PLYs
2 png and .KAP header is passed off to libbsb and a BSB is created.
3. new sat-photo BSB is saved to a sub-directory of the current chart directory named "custom"
4. all BSBs in the "custom" directory show up in the bottom available chart bar as a red rectangle instead of blue (reserved for real chart BSBs)
5. all sat photos previously grabbed off net are available later when net connection is gone

Will

[bdbcat]

Mainmaul....

I like this idea.
Can you point me to a "HOWTO" for the first clause of Item 2?
I don't have any experience with the GoogleMap API.

Dave

[manimaul]

Originally Posted by bdbcat

Mainmaul....
I like this idea.
Can you point me to a "HOWTO" for the first clause of Item 2?
I don't have any experience with the GoogleMap API.
Dave

Hmmm, i think what your looking for is here:
Google Static Maps API - Google Code
"The Google Static Maps API returns an image (either GIF, PNG or JPEG) in response to a HTTP request via a URL"

I'm not sure how to do this without a site key ... but this program is doing it:
Google Satellite Maps Downloader
Too bad it's not opensource so we could examine how it is done. I should mention that this program is can grab sat photo images from google, microsoft, and yahoo.

Will