forum.heinrichsweikamp.com

Interesting results for the average depth during simulated dive to 90 meters:

[pre]
Elapsed | Current | Average
Time | depth | depth
[m:s] | [m] | [m]
----------------------------
0:01 | 90 | 45,0
0:20 | 90 | 78,1
0:40 | 90 | 83,4
1:00 | 90 | 85,5
1:20 | 90 | 86,7
1:40 | 90 | 85,5 (!!!)
2:00 | 90 | 84,0
2:20 | 90 | 82,3
...
3:00 | 90 | 80,5
...
4:00 | 90 | 79,5
...
[/pre]
Results for other simulations are not exactly reproducible, but always after 1:30 - 1:40 the average depth begins to decrease, and should continue to grow and approach 90 meters. For me it is a bug.

Looks strange, the 1.82 seems not to be the best release ever

Have you seen my dive profile in the 1.82 thread where the surface depth decreased while the dive was going on? Maybe its the same issue...

Guy's

They re coded the whole algorithm to fix a couple of bugs and make it more efficient (and introduced a couple of others).

Just be patient and keep reporting any issues.

D

It's been fixed in the next beta by the way:

Re: New 1.82 beta available for the Mk.2
Posted by: JeanDo ()
Date: March 01, 2011 11:37PM

There was a bug in 1.81 with GF handling, moving the GF_low reference depth up as you ascent. This was particularly visible in settings like GF 10/90, and produce was you observed, with increasing time for later stops and augmenting TTS value.

I corrected that and made a couple of dives with the future 1.83 beta (on a backup computer !), and it seems corrected now.

Thank you very much for all those reports that help us to test faster the new beta versions !

~~~~
JD.Gascuel@free.fr, OSTC Mk2 1455 & 1655.

It still appears to be there in 1.83 (see my comment in that thread) although only in the GF model, not the plain OC model.

nakatomi is right, this problem still occurs in 1.83 version too, but now average depth begins to decrease after more than 3 minutes (for my example: GF model, 90 meters depth).