Expedition Navigation Software 
1 of 1- sparky
- Expert
Offline - From: Cambridge UK
- Registered: 4/27/2018
- Posts: 225
Iso vs Grid routing...
Hi Nick --- lots of great work recently this past year.
What are the current capabilities of Grid and Iso routing and which to do you recommend now?
I think you added the ability to keep marks to port or starboard in Iso in the past year which was the only reason I tended to use Grid.
With all the improvements to ISO is there any reason out side of a double check to run grid?
- Nick
- Moderator
Offline
- Registered: 9/16/2015
- Posts: 2,752
Re: Iso vs Grid routing...
Hello,
Thanks.
I haven't extend the port/starboard to the isochronal routing yet. It isn't necessarily trivial and there have been a lot of other projects.
The isochronal routing is slowly improving though.
Definitely, the isochronal routing is now slightly better. There are some information and logic reasons why the isochronal approach will be better.
That said, now the routing is multi-threaded, it isn't any slower to run both if you have a computer with multiple CPU cores.
As a tip, I mostly use the TG button to run the routing now.
- Kite
- Expert
Offline
- Registered: 1/14/2022
- Posts: 33
Re: Iso vs Grid routing...
Hi Nick,
Am I right that Isochronal won't work with race notes? I'm getting an 'unknown error' which refers to this.
- Nick
- Moderator
Offline
- Registered: 9/16/2015
- Posts: 2,752
Re: Iso vs Grid routing...
It supports race notes.
The paths option will normally tell you what is happening.
- Kite
- Expert
Offline
- Registered: 1/14/2022
- Posts: 33
Re: Iso vs Grid routing...
Problem solved. I misread your po=ost about port and starboard roundings. All working fine now.
- radaroom
- Expert
Offline
- Registered: 1/02/2020
- Posts: 31
Re: Iso vs Grid routing...
Is there an explanation of how Grid and Isochronal routing actually works?
GRID - I am interested in understanding how on an 8km x 8km GRIB (I assume there is one TWS/TWD/MSLP/etc data point inside that grid every hour?) there is an interpolation between each grid. Is it linear?
Isochronal - I assume that the interative paths are defined by the settings in the routing resolution?
Would welcome a high level description please.
- Nick
- Moderator
Offline
- Registered: 9/16/2015
- Posts: 2,752
Re: Iso vs Grid routing...
The isochrone option just pushes an isochrone out. Obviously, the time steps are more regular than the grid option. It is slower, but there are many reasons it should be better - I only use the isochronal option now.
In each weather model, the parameters in each column are effectively an average at that level. WRF uses a quadrilateral cell, ICON triangular, some are hexagonal etc. The wind components are at the boundaries.
The 10m winds are derived quantities from the wind components and the atmospheric structure.
The grib data you get is often interpolated to a regular grid. Exceptions where it is on the original grid are HRRR, RAP, Exp WRF etc.
Exp then needs to interpolate the data. It fits a smooth surface in space and time to the data, which should generally be better than a linear interpolation. It adds computational over-head, but I think is worth it.