Encore Pro 1.7i vs 1.6 or 1.8

[mhacker]

Too funny, that is the third time I have been in Wiki. Not the independant conformation that I wanted to see!!!!!

I think with several people now checking, we may have an answer and solution but not the root cause.

Looking at the XML dump of the data, it looks like (as I suspected) that the data is stored in "time of seconds" - each "event" is recorded in seconds after therapy started. The "real time clock" time stamps but it is so much easier to just dump the event with a "how many seconds since you started" flag. This also could be the root source as the buffer to keep those seconds may not have the amount of storage required (one of my hits happened a day after a 14 hour sleep in)

I will post a sample later - bought a program just to get at the data... Anyway, it means that to drop off the oldest day, the processer in the card and the xPAP memory needs to reconcile the "time ticks" - if you have a lot of "events" in the last 6 days, it will take time to process the data stream so the old data is gone and the new day's data can be stored, in "seconds since.." format. I suspect that a partial day (of removal) does the corruption. The new day might be writing while the oldest day is overwritten and when it is interupted - well, anything can happen.

I think that a combination of firmware (xPAP device) and software is the real fix but lets face it - we have already paid and they will not make money fixing it.

Now if Repironices was smart... Come out with a new version of EP that included a machine patch... They would sell how many????

[mhacker]

As you can see, the data is similar to the HTML tags when you post - a begin and end tag for the type <tag> data </tag>

- <SleepHistogram>
<ObstructiveApnea>1</ObstructiveApnea>
<Pressure>11</Pressure>
<NRAH>0</NRAH>
<SecondsAtPressure>25830</SecondsAtPressure>
<Hypopnea>0</Hypopnea>
<VariableBreathing>0</VariableBreathing>
<FlowLimitation>0</FlowLimitation>
<Snore>6</Snore>

So for this event, which are recorded at 30 second intervals, happened at 25830 seconds or 430.5 minutes or 7.1 hours. What? Yes, all the data needs to be shifted by a lot in the registers of the card and the program. If 30 seconds later (25860) I had another "snore" then the program knows when to tag it and it shows on the graph as 7 and the time line will calculate the "exact" time down to 30 seconds.

Ya think I have been teaching a class scripting for the last 3 days??? Spell check died on this one....