ALL CLEAR about the urgent warning

ALL CLEAR! Well, almost!!

Resscan software has a problem with displaying the data the day AFTER DST goes into effect. There must be some sort of screwed up DST routine in the software. What you will see is a one hour gap in the flow and detail data between 2 and 3 am on March 14. Oximetry data, if recorded, is shown continuous. The start and end times of the flow and detail data are correct. This problem was reported by DreamDiver last year. But the machine was so new then it wasn't widely used and so the problem wasn't a big concern.

We expect, right now, that all users will see this data gap whether or not you reset the time in the S9.

As of now I see no problem with downloading and saving your data in Resscan, unless you set the time on your S9 ahead one hour BETWEEN the last mask session Sunday morning and noon Sunday, 3/13. We have one case of data in Resscan being fouled up but have not yet determined if setting the S9 clock ahead during that time span is the cause or not.

Lizistired wrote:

idamtnboy wrote:...I am of the firm belief the S9 has a problem with its data recording system.

I think we probably ask more of it than is usually asked. I mean really, how many hours have you spent on data today.
No Med Prof is going to look that deeply.
Most of it seems to be general info that can be used to assess sleep disorder pogress. That's good, because a doc doesn't want to ahve to look at breaths. I don't feel that it is that accurate, but am glad it is there to help me learn what I should be looking for.

I agree, wholeheartedly! I enjoy the challenge. Sometimes someone needs to invest (spend?) a lot of time analyzing a situation just to provide substantiation there is no real problem! Once I get a problem figured out I may never look at it again. But I will admit, other issues around the house get neglected at times.

BernieRay's posting this AM did unnerve me quite a bit though, since I was so dead sure I had provided reliable information about resetting the clock in the S9. Can you imagine how I would have felt if a hundred, or a thousand, users had lost precious (to them) data because of what I had posted? It is possible, but not certain, that Bernie's problem was caused by following part of my procedure. We haven't nailed it down yet.

Well, I thought I had things fixed a little while ago, but no luck. I reverted my patient folder back to yesterday, then erased all data on the S9, then forced the S9 to erase the card. If I now try to download the basically empty card, I am still losing summary data and the stats for the last day end up being displayed for all days.

I'm going to see what happens in the morning with tonight's data. if the problem persists, I'll try creating my patient folder from scratch, skipping yesterday's data.

I don't know if changing the time before noon yesterday caused this, but it might be a good idea for folks to wait, just in case, since that seems to be the only thing I did differently.

I changed the time after 1:00PM on Saturday. The gap starting at 2:00PM and ending at 3:00PM is on the data from
Sunday evening until Monday morning. The summary graphs are okay. All the detailed graphs have a 1 hour missing area. My AHI ends at 2:00AM .

Okay, I just couldn't let it go. When erasing the data on the S9 and card didn't help, I reverted back to Sunday morning's Resscan backup. A subsequent download of the now empty card letd to lost summary data and the copied last day's stats to all other days, just like before.

Next, I restored from my Saturday backup, so that the data from the actual DST change was no longer around. I was then able to download fromthe empty card, save my file, restart ResScan and still have all of my pre-Sunday data intact. I've now done this twice and it worked both times.

It would seem that by changing the time on Sunday morning before noon actually caused ResScan to have a problem with my pre-existing Saturday night download, not the Sunday night download.

If my download in the morning runs fine, it would appear that setting the time forward 1 hour on Sunday before noon was the culprit even though I waited 2 hours after I had gotten up on Sunday morning.

Does that make sense?

BernieRay wrote:Okay, I just couldn't let it go. When erasing the data on the S9 and card didn't help, I reverted back to Sunday morning's Resscan backup. A subsequent download of the now empty card letd to lost summary data and the copied last day's stats to all other days, just like before.

Next, I restored from my Saturday backup, so that the data from the actual DST change was no longer around. I was then able to download fromthe empty card, save my file, restart ResScan and still have all of my pre-Sunday data intact. I've now done this twice and it worked both times.

It would seem that by changing the time on Sunday morning before noon actually caused ResScan to have a problem with my pre-existing Saturday night download, not the Sunday night download.

If my download in the morning runs fine, it would appear that setting the time forward 1 hour on Sunday before noon was the culprit even though I waited 2 hours after I had gotten up on Sunday morning.

Does that make sense?

I'd still like to see the STR files off your card from yesterday or today. When the STR file is written there are 10 records of data written into it to cover the entire 24 hr period for the following day. Those data records all have -1 in the fields until data is inserted later. Records 2 to 10 appear to be used primarily to record mask on and off times. The other fields in record 1 have all the summary data and is updated regularly. I think the way it works is when you turn on the flow generator the first time after noon the 10 empty records for the next day are written to the card and the ones written yesterday, for today's use, are then filled in. But why or how this would get screwed up by changing the time before noon mystifies me.

PM sent, idamtnboy.

Well, the problem is persisting. At this point, all that I know for sure is that Sunday morning at 9:00 AM, 2 hours after I got up, I advanced the time on my S9 by 1 hour. Since then, anytime I tried to download new data, once I save my file in ResScan and re-start ResScan, I've lost all summary data in the graphs (the symbol in the browser indicates that the data is still there) and the stats for the last day are displayed for all days.

I've restored ResScan back to where it was on Saturday morning, erased the data on the S9, and erased the data on the card. Didn't help.

I think I am going to wipe out my ResScan install and start from scratch. if that doesn't work, I'll try setting the clock back on S9 and see if that helps.

On Tuesday morning at 0930 I downloaded my data from last night. There were no missing time periods.

Just for the sake of completeness I'd like to mention that the S8 does not have this problem as its clock does not support DST. On the morning on Sunday I downloaded my data and then changed the clock, and that was it - no trouble with any data. Of course, progress is good and introducing automatic DST is nice, but trying to do the simplest thing shouldn't bring about such mayhem. As a software engineer in the area of high-tech data acquisition and control, I must say there is no excuse for the ResMed engineers to make such childish mistakes with the S9 and ResMed.
McSleepy

BernieRay wrote:Well, the problem is persisting. At this point, all that I know for sure is that Sunday morning at 9:00 AM, 2 hours after I got up, I advanced the time on my S9 by 1 hour. Since then, anytime I tried to download new data, once I save my file in ResScan and re-start ResScan, I've lost all summary data in the graphs (the symbol in the browser indicates that the data is still there) and the stats for the last day are displayed for all days.

I've restored ResScan back to where it was on Saturday morning, erased the data on the S9, and erased the data on the card. Didn't help.

I think I am going to wipe out my ResScan install and start from scratch. if that doesn't work, I'll try setting the clock back on S9 and see if that helps.

If you have another computer available that you can put Resscan on, it might be interesting to see if you see the same problem there. Or send me all your files for the past 5 or 6 days and I'll load them into my Resscan. I don't need your Patient.db file. I'll create a fake. At this point I'm thinking something is haywire with your Saturday morning saved file, even though it shows up OK in Resscan, or as you're suspecting, Resscan has gotten screwed up somehow. How about rebuilding your file back to some morning last week, put all the detail and hi-res files since then on an SD card, except for the ones for Friday night, and see what happens on download. Have you tried downloading and rebuilding under a new patient name. One of the pluses of Resscan is you can create all kinds of fake test patient files and experiment around without jeopardizing your real data.

With all you've done I'm getting pretty skeptical that the time change on the S9 is the cause.

McSleepy wrote:Just for the sake of completeness I'd like to mention that the S8 does not have this problem as its clock does not support DST. On the morning on Sunday I downloaded my data and then changed the clock, and that was it - no trouble with any data. Of course, progress is good and introducing automatic DST is nice, but trying to do the simplest thing shouldn't bring about such mayhem. As a software engineer in the area of high-tech data acquisition and control, I must say there is no excuse for the ResMed engineers to make such childish mistakes with the S9 and ResMed.
McSleepy

Do you have the one hour, 2 to 3 AM Monday morning, gap in the detail data graphs in Resscan? Which version of Resscan are you using?

The S9 doesn't have DST built-in. BernieRay did the same thing as you did, set the time up Sunday morning. We're trying to figure out if the S9 chokes when the time is changed before noon. Obviously, from your experience, the S8 does not.

Bobby269 wrote:On Tuesday morning at 0930 I downloaded my data from last night. There were no missing time periods.

Thanks. That's to be expected. From everything that's been posted here the gap is a one-day-year problem, the day after US & AU change to Daylight Time. The EU changes in 2 weeks.

idamtnboy wrote: Do you have the one hour, 2 to 3 AM Monday morning, gap in the detail data graphs in Resscan? Which version of Resscan are you using?

The S9 doesn't have DST built-in. BernieRay did the same thing as you did, set the time up Sunday morning. We're trying to figure out if the S9 chokes when the time is changed before noon. Obviously, from your experience, the S8 does not.

No, I don't have any gaps, anywhere, I just double-checked. I am using ResScan 03.12.016 on Windows XP SP3 32-bit. Another difference is the S8 uses smart-cards, but probably it is all in the data format. Normally, the way to avoid any trouble is to always store data recordings in relative time (using continuous timestamps) together with a starting reference date in UTC (i.e., the date of the origin). Then, whenever the data is imported for analysis, it is reconstructed using those parameters. This way, there could never be any confusion, no matter what surprises (like DTC is a big surprise!) is thrown in the mix. I don't know what the ResMed engineers did to cause this mess but if the S9 doesn't have a DST-aware clock and it's all in how ResScan interprets continuous (i.e., gap-less) data, it's a great shame on them.
McSleepy
P.S. I am not 100% positive whether I changed the date before noon DST, but I almost am. It certainly was around that time.

I'll just try to remember NOT to change the time in the fall.