sleepyhead question

My Sleepyhead is calculating time slept wrong....thus my AHI is showing lower then it should be....

It says sleep:22:13 wake: 06:19 hours: 14:19.02 it should be 8:06 hours slept.

It says my AHI is 1.97, but for 28 events in 8 hrs, it should be 3.5

Any thoughts on why its calculating the hours wrong?? The sleep & wake times are correct. Seems to do this
on any if I do to bed before midnight.....

A bug that developed when DST change occurred.

You likely have double sessions being recorded and yes..sometimes related to midnight.
I have seen it myself.
Jedimark is aware of it and until he gets the bug worked out we just have to sort of ignore it.
I found that if I turn off one of the sessions that the hours asleep work out properly and events tab add up properly.
Look at the very bottom of your Details scroll bar and turn one of the sessions off. It may be the top or bottom.. Easy to do..put mouse cursor on green area and turn it to red. If your data goes away, turn it back to green and choose the other.

I have an image of what to look for here. This is at the very bottom of the details section (under the pie chart...so scroll all the way down.

Thanks Pugsy.....as long as I'm not doing something wrong and he knows about the bug.

You aren't doing anything wrong.
Most nights I see double sessions myself but not every night.
One night I did sleep nearly 10 hours and SH said I slept 20 hours. I just go turn off one of the sessions pretty much immediately now.

Mark is aware of it but his plate is full right now and while annoying little bug...it isn't a hugely critical thing. You can always use ResScan if you wish and I do at least go ahead and import to ResScan at least once a week so I have those files as backup if needed. Other than the time thing and some minor rounding off of the numbers differences I haven't seen any glaring discrepancies in data when comparing SH to ResScan.

Pugsy wrote:You aren't doing anything wrong.
Most nights I see double sessions myself but not every night.
One night I did sleep nearly 10 hours and SH said I slept 20 hours. I just go turn off one of the sessions pretty much immediately now.

Mark is aware of it but his plate is full right now and while annoying little bug...it isn't a hugely critical thing. You can always use ResScan if you wish and I do at least go ahead and import to ResScan at least once a week so I have those files as backup if needed. Other than the time thing and some minor rounding off of the numbers differences I haven't seen any glaring discrepancies in data when comparing SH to ResScan.

Would it be possible to send me e.g. a zip file containing a data set that reproduces the problem?

I'm a retired computer programmer with an unusual combination of free time and debug skill. I'll take a look at the code, especially anything involved in deciding which chunks of data belong in the same night, but it would be easier with sample data.

pats wrote:Would it be possible to send me e.g. a zip file containing a data set that reproduces the problem?

Sure tell me what you need and where to send it...send me a private message with your regular email address because we can't do attachments here.
I can make a copy of my SD card files to attach.

Also..Mark has all his coding stuff over on sourceforge website for viewing for those who understand all this stuff and I am not one of them. I test things but don't have a clue about the other stuff.
Links in my signature to sourceforge.

Pugsy wrote:

pats wrote:Would it be possible to send me e.g. a zip file containing a data set that reproduces the problem?

Sure tell me what you need and where to send it...send me a private message with your regular email address because we can't do attachments here.
I can make a copy of my SD card files to attach.

Also..Mark has all his coding stuff over on sourceforge website for viewing for those who understand all this stuff and I am not one of them. I test things but don't have a clue about the other stuff.
Links in my signature to sourceforge.

Yes, I'm set up with the source code.

I've sent a PM with my e-mail address. An image of an SD card that reproduces the problem would be perfect. Perhaps you could check that the data reproduces the bug by the following:

1. Copy the SD card to a "temp" folder.
2. Attempt to import from the "temp" folder into a new SleepyHead account
3. If it works except for the bug, compress the "temp" folder and mail it to me. If it is too big for e-mail I'll send you FTP instructions.

Thanks,

Patricia

pats wrote:1. Copy the SD card to a "temp" folder.
2. Attempt to import from the "temp" folder into a new SleepyHead account
3. If it works except for the bug, compress the "temp" folder and mail it to me.

I did not do the temp folder thing. I saw your PM before I just now saw this post. So you got zipped files directly from the SD card copy and I didn't try a new user but I still have the copy .

I am pretty sure the result would be the same because I have reinstalled SleepyHead Beta several times in an attempt to sort it out and I used a copy of the SD card for importing the older and I have also created multiple users (that was the first thing I tried in fact).

Edit.. I did test my copy of the SD card with a new user and same results..double sessions.. I had DST checked in SH.
I did a test of my copy again this time with another new user and DST unchecked in SH. Still double sessions.