Software Suggestions

[atdotde]

Let me also mention that you can do a lot of these things mentioned here with the planner in Subsurface. In particular, in the latest version that came out this week, we also show heat maps both for planned as well as logged dives and for dives planned with VPM-B, we compute GF's that would lead to "similar" dives by doing linear regression on the effective gradient factors encountered during VPM-B deco.

In my blog, I also commented on gradient factors for VPM-B dives here: VPM-B Gradients as Gradient factors The upshot is mainly that the variation in GF's during the ascent mainly comes from the different compartments having vastly different GF's and the further up in the ascent you get, compartments with slower half-times become the leading compartment (and those have the highes effective GF).

Regarding implementing VPM-B on a 21st century computer: Of course you can still compile the FORTRAN code or you can use the C version that you can find on the internet that was machine translated from FORTRAN a while ago. For Subsurface, we chose to do a complete reimplementation and found some curiosities, about which I also started blogging The Theoretical Diver | Scuba diving thoughts of a theoretical physicist (scroll down a bit, don't get confused by the German translation of one of the articles).

Just let me point out, that it is far from straight forward to extrapolate the code for a planning software to real dives: At several places in the code, the bottom segment of the dive is treated differently from the ascent but for a real dive this distinction is of course less clear. Also, because it looks into the future (via the "critical volume algorithm"), it relies on you following the plan in the future, in particular staying below stops for longer violates it assumptions.

 

[dberry]

I gather @atdotde is part of the team coding SubSurface - thank you very much, it's a wonderful project!

In the realm of a dream wish-list, it be interesting to have a GUI that would plot the initially planned profile and deco plan, and let you drag the cursor to trace a possible "actual" profile (i.e. longer at one depth, overshooting a stop, etc.) while watching the deco plan change in real time based on the "actual" scenario up that point. Gas changes could be indicated with a click as you draw the "actual" profile.

(Similar to the Perdix TTS @+5 feature, but interactive and graphical during the planning.

Oh well, ideas are cheap; implementations less so.

 

[atdotde]

@dberry indeed, I contribute to Subsurface, mainly for the planner and deco part.

What you suggest sounds interesting. Let me think a bit how to implement this or similar without too much pain.

In the meantime, there is always the option to do "open dive in planner" and delete all the waypoints starting from some time (can easily be done by holding the control key while clicking on the trash bin icon: this deletes all later waypoints) Then the planner will insert the remaining ascent. Save the result and compare it with the actual dive.

 

[dberry]

@atdotde I hadn't played much with the dive planner in SubSurface, but now I find it does much of what I was looking for. I didn't realize you can drag a waypoint and watch the tissue loadings, deco plan and gas consumption change "live". That's pretty close to what I was looking for in the planner. The only thing missing would be the ability to right-click on the graphic profile to add a waypoint at some time / depth point. (or maybe that feature exists, but I couldn't find it.)

One thing I couldn't figure out: when I added a tank it only has field to enter the working pressure (a constant for a given tank), not the starting pressure. Then after leaving planning mode you can enter a different start pressure But then the "plan" doesn't change. I suppose the intention is that we just use the working pressure of a tank as the expected starting pressure.

With LP tanks especially it's not unusual to get fills >10% above the 2640psi working pressure. And some dive shops give notoriously low fill pressures. Technically these don't change the working pressure of the tank, which is a constant, but definitely change the amount of gas in a tank.

I wonder if Europeans have a slightly different understanding of "working pressure" than North Americans. Working pressure is a parameter for each type of tank. In the imperial system we use the working pressure and the nominal capacity to figure out the actual volume of the tank for SAC calcs and such. Please don't point out how silly this system is - we know, we know!

 

[atdotde]

Double click adds waypoints. And, in the planner, you are supposed to enter the starting pressure in the working pressure field.

 

[atdotde]

BTW, for more fine grained control, you can also select a waypoint in the profile and then move it using the cursor keys.

 

[Michael Guerrero]

If we're registering requests for additional functionality, I was using software called MVPlan for a while, written by a guy named Guy Wittig. His software, not actively under development, has a feature that allows you to create contingency tables for your planned dive. For example, you have your planned dive table and right next to it a plan for staying an extra 5 or 10 min at the target depth (user-definable of course), and next to that a scenario or two for exceeding or falling short of your target depth.

This made it super easy to cut a single table with your planned dive and several contingency scenarios all one one small form factor. It also showed the GFs used to produce the table, so you knew how you were modifying the ascent curve. It's a great little feature that I really miss.

Mike

 

[atdotde]

I would love to have an easy way to do contingency plans. The problem with that is that everybody wants something different (what different depths, which additional bottom times, what kind of gas loss scenarios, does gas loss lead to additional bottom time to sort things out, increased SAC rate, you name it). Of course, you can add all kinds of controls to fine tune this behaviour. But then your user interface looks like the space shuttle cockpit. And nobody will get things right.

So with all these things (another similar issue is some sort of rock bottom gas calculation) the actual challenge is how to implement this in a way that makes at least the vast majority of users happy while maintaining a user interface that does not lead to immediate information overload. Ideas welcome! Note that just "I want xyz" is not a sufficient answer.

 

[alewar]

Another vote for subsurface! It's only lacking DecoPlaner's "cut tables" option.

 

[atdotde]

I don't have access to GUE's DecoPlanner. What is the "cut tables" option?