Oscar data?

[Crimson Nape]

Have you tried creating a new user and see what happens?

[todivefor]

Have you tried creating a new user and see what happens?

Just did that. Even worse. No current data. Most current data is 2015.

[Crimson Nape]

Where are you importing the data from?

Edit: You mentioned 2 different machines. You did store each one's data in separate locations, didn't you?

[sawinglogz]

I have a couple anomalies in the Oscar displays. As I said in my first post with my "Statistics" display, the first 4 lines of the "Pressure Statistics" are all 0s. In the "Overview" display pressures are showing. See screenshots in OP.

On the "Overview" display, there is no data for "Total Time in Apnea." If I go back in time, this data goes missing after 11/1/2015. That coincides with a "Changes in Machine Settings." I am not sure even what all those changes mean. Sleepyhead's "Changes to Prescription Settings" seems more accurate. I had 3 devices. The S9 AutoSet (36005) was a loaner during the Remstar Auto (System One) service (10/21/09 - 9/26/18). I did make pressure changes with the current DreamStation, but changed nothing during the other 2 devices.

Any ideas? Need anything?

The 0s in your pressure statistics do seem wrong. I don't remember who was working on that page.

TTIA is essentially removed for Remstar and DreamStation because it was entirely fictitious. Only Resmed reports duration of events.

To my eye, OSCAR's "changes in prescription settings" seems more accurate. There are all sorts of changes (partly because OSCAR now detects them correctly) that Sleepyhead missed:

• Oct 21 2009 - Oct 12 2014: Your machine went into CPAP-Check mode for one night roughly every 7 months. (I don't know why, maybe your DME triggered that remotely?) SleepyHead didn't recognize CPAP-Check mode.
  
• Oct 13 2014 - Oct 14 2014: It looks like you used the Resmed for one or two nights. Sleepyhead missed that for some reason.
  

Everything else seems to line up, though note that OSCAR correctly identifies your use of A-Flex 2 on your DreamStation.

[sawinglogz]

I just uploaded PRS1_P001329800E06.zip. Let me know if you need anything else.

Thank you, that helps immensely.

On three nights it looks like you adjusted your humidifier during the night, which was the final piece of the puzzle I needed to identify that event and confirm that how it affects the overall session time. (I had seen it before, but didn't know what it meant, so the the alert you saw was to help me find additional sample data to figure it out.) Success!

It turns out I had guessed correctly on how to handle the timestamp in that event, and since we only import whatever setting the machine reports for the overall session, this won't have any visible affect on your data, which is already correct.

However, the night of Nov 30 2014 contains a whole new event that I've never observed before for the 550P series, so that night will currently be truncated.

Other than that, everything should be accurate.

Though you had a lot of data corruption in the files for that machine: there were lots of checksum errors, and lots of waveform data with zero data instead of a chunk header. I don't know if that indicates a bad SD card or buggy software in that 550P, I've never seen more than the occasional zero header, but you have hundreds.

Anyway, the next version, whenever that may be, will no longer complain about surprising data if you rebuild.

[GuyScharf]

You may want to try having OSCAR rebuild the Changes in Machine Settings database.  It shouldn't make a difference but is worth a try.

In your Profile directory on disk, there is a file named rxchanges.cache.  Rename it to something else (don't delete it yet -- just in case).  Start OSCAR and the list of machine settings changes will be rebuilt.

Regarding the Pressure Statistics lines, I think this is a result of having both ResMed and PRS machines in the same profile.  When the program decides which lines to print, it checks whether a channel exists for that item.  The Pressure Set channels are a newer channel introduced to resolve problems inherited from SleepyHead which put different types of pressure settings in the same channel.  And ResMed and PRS machines may have their data in different channels.  So you get reports from both sets of channels, even though one is always zero for the time period of the report.

We can probably eliminate the lines with all zeros.

UPDATE: Logged as Mantis #251.

[todivefor]

It was probably too late for me to start setting up a new profile last night as suggested by Crimson Nape. I think I imported the wrong data. In looking at this this morning, I found I have 14 P-SERIES folders on my computer, not including my SD CARD. I have been playing with FlashPAP, SleepMaster, Oscar, and SleepyHead each having there own P-SERIES folder(s). I need to sort this out. I didn't know I had to have a different profile for each machine. Except for my first machine (1999 - 2009), I always merged data to the new card, ie my data and profile contains data for 3 machines. Do I really need to have a different profile for each machine? I do not want lose my history. Is there a way I can break each machine out to its own profile. Obviously, I need to get a handle on my data. This may resolve some of my issues. Any advice would be greatly appreciated.

[todivefor]

I have a couple anomalies in the Oscar displays. As I said in my first post with my "Statistics" display, the first 4 lines of the "Pressure Statistics" are all 0s. In the "Overview" display pressures are showing. See screenshots in OP.

On the "Overview" display, there is no data for "Total Time in Apnea." If I go back in time, this data goes missing after 11/1/2015. That coincides with a "Changes in Machine Settings." I am not sure even what all those changes mean. Sleepyhead's "Changes to Prescription Settings" seems more accurate. I had 3 devices. The S9 AutoSet (36005) was a loaner during the Remstar Auto (System One) service (10/21/09 - 9/26/18). I did make pressure changes with the current DreamStation, but changed nothing during the other 2 devices.

Any ideas? Need anything?

The 0s in your pressure statistics do seem wrong. I don't remember who was working on that page.

TTIA is essentially removed for Remstar and DreamStation because it was entirely fictitious. Only Resmed reports duration of events.

To my eye, OSCAR's "changes in prescription settings" seems more accurate. There are all sorts of changes (partly because OSCAR now detects them correctly) that Sleepyhead missed:

• Oct 21 2009 - Oct 12 2014: Your machine went into CPAP-Check mode for one night roughly every 7 months. (I don't know why, maybe your DME triggered that remotely?) SleepyHead didn't recognize CPAP-Check mode.
  
• Oct 13 2014 - Oct 14 2014: It looks like you used the Resmed for one or two nights. Sleepyhead missed that for some reason.
  

Everything else seems to line up, though note that OSCAR correctly identifies your use of A-Flex 2 on your DreamStation.

Just a couple observations. As Crimson Nape alluded to, I broke out my 3 machines to separate profiles. The documentation talks about importing multiple machines into a single profile, so I am not sure this is a requirement. I noticed in my DreamStation only  profile, the 0s are missing because "Average Pressure", "Min Pressure", "Max Pressure", and "95% Pressure" are eliminated completely from "Pressure Statistics."  The "Changes to Machine Settings is a little hazy because my Resmed machine was loaner while my System One was being repaired. Couldn't validate the actual dates, but the one System One dates between to Resmed dates didn't make sense. I am not concerned about this, only pointing it out. TTIA makes sense since it is only missing in the DS.

[todivefor]

I just uploaded PRS1_P001329800E06.zip. Let me know if you need anything else.

Thank you, that helps immensely.

On three nights it looks like you adjusted your humidifier during the night, which was the final piece of the puzzle I needed to identify that event and confirm that how it affects the overall session time. (I had seen it before, but didn't know what it meant, so the the alert you saw was to help me find additional sample data to figure it out.) Success!

It turns out I had guessed correctly on how to handle the timestamp in that event, and since we only import whatever setting the machine reports for the overall session, this won't have any visible affect on your data, which is already correct.

However, the night of Nov 30 2014 contains a whole new event that I've never observed before for the 550P series, so that night will currently be truncated.

Other than that, everything should be accurate.

Though you had a lot of data corruption in the files for that machine: there were lots of checksum errors, and lots of waveform data with zero data instead of a chunk header. I don't know if that indicates a bad SD card or buggy software in that 550P, I've never seen more than the occasional zero header, but you have hundreds.

Anyway, the next version, whenever that may be, will no longer complain about surprising data if you rebuild.

Glad it helped. I am no longer using the 550P or that SD CARD, so all should be good. Let me know if you need anything else.

[todivefor]

You may want to try having OSCAR rebuild the Changes in Machine Settings database.  It shouldn't make a difference but is worth a try.

In your Profile directory on disk, there is a file named rxchanges.cache.  Rename it to something else (don't delete it yet -- just in case).  Start OSCAR and the list of machine settings changes will be rebuilt.

Regarding the Pressure Statistics lines, I think this is a result of having both ResMed and PRS machines in the same profile.  When the program decides which lines to print, it checks whether a channel exists for that item.  The Pressure Set channels are a newer channel introduced to resolve problems inherited from SleepyHead which put different types of pressure settings in the same channel.  And ResMed and PRS machines may have their data in different channels.  So you get reports from both sets of channels, even though one is always zero for the time period of the report.

We can probably eliminate the lines with all zeros.

UPDATE: Logged as Mantis #251.

Renaming rxchanges.cache made no difference. I am not too concerned about this. See my description up above. The Resmed was a loaner. Just surprised to see a System One record intermixed.

Your explanation of the 0s in "Pressure Statistics" seems spot on. I broke out the DS into a separate profile and those stat lines go away.