Hacker News Reader

Comment by dakics | Hacker News Reader

I've bought recently one of those cheap wrist bands.

One of the most important features for me was ability to measure sleep quality in terms of "daily hours of deep sleep," whatever that scientifically means.

Over the course of several months, I've been able to identify things which affect my sleep quality:

* eating late: bad

* drinking late: one beer... not really that bad

* answering work emails until midnight: REALLY BAD (even while drinking soothing chamomile tea during soft rain) etc.

I'm trying as reduce as many "bad" things as possible as I can feel a huge difference between 30 mins and 3+ hrs of deep sleep, as measured by the device.

I feel much better during the day, can work more productively and, most importantly, my "fuse" seems to be much longer.

Probably best $30 ever spent.

(still missing weekend code binges of early 20s, though :/ )

View on HN

What is the brand of the device? I am interested in getting one. I'm struggling today after little sleep, but I would also like insight on how much deep sleep I am getting.

Mi Band 3. It appears its sleep quality measurement through gyroscope/accelerator is accurate enough.

You don’t even have to get a device to get started. A cheap (but useful) app like Sleep Cycle will give you lots of info to start noticing patterns.

Apple Watch does the same thing, but you can choose from 3rd party apps as well.

I’ve found that it boils down to either excess energy not burned off, or not enough calories to slow the mind down yet. So if my mind is racing a bit I get up, eat something small and light on the stomach like a banana or cereal in the dark without tv or the phone. I use some of this time to meditate as well.

Also getting enough sun or vitamin d helps. Trying to burn off as much energy before my mind has a chance to race before the end of the evening.

I also believe sleep quality and carbondioxide level could be related. Need to get a device and give it a test.

https://www.gwern.net/zeo/CO2

One more independent variable you might want to test: when you go to bed. I've found that the earlier I go to bed, the higher quality sleep (I'm sure there is a point at which I would begin getting worse sleep, but I haven't reached it). I've tried anywhere from 9PM to 3AM. 9PM is a little too early to do if you aren't trying to "catch up" on sleep but I've been getting great results going to sleep at 10PM every night. I usually fall asleep around 10:30 or 11PM and wake up naturally to the sun around 7-8AM.

Never feel tired at work anymore, but I do actually start getting tired around 8-9PM when the caffeine from work starts wearing off. The energy and focus is great

Agreed. I get up every morning around 6am. Some mornings I get up at 5am for a group workout thing, but it's mostly around 6am. I'm normally in bed right before 10 because I have found out through experience that I need ~7-8 hours of sleep/night.

I can shift this a bit, say go to bed around 11 or 12 and sleep later, but I am not nearly as productive in the mornings then. If I push this more extreme and say go to bed at 2-3am, even if I sleep until 10 or 11, I am wrecked the next day. I actually feel worse than if I only got 6 hours of sleep instead, but went to bed earlier.

I actually find that a couple of hours of coding on personal projects from 9 to 11 (but not later) actually has a calming effect on me that increases sleep quality.

Mind you - I don't have hard data to support this!

You get 3+ hours of deep sleep?! I seem to regularly crack 1 hour of deep sleep (tracked by my watch) but that’s it.

Anything you’ve done or do to improve the amount you get?

Word of warning, the sleep trackers on the market are not very accurate for tracking sleep phases. Folks like Matthew Walker are pretty adamant about this. You have to measure your electric brain waves to know for sure, not approximations from heartbeat, movement, etc.

My Oura and Fitbit sometimes give wildly different sleep phase readings for the same night. Also, each will sometimes give arbitrarily unexpected readings that leave me scratching my head.

As someone mentioned, those trackers aren't that accurate.

However, if you think you don't get much deep sleep, for the love of everything, get tested for sleep apnea. A home sleep study is quick and easy if you have reasonable insurance, and if you do have apnea, getting it treated could literally change your life (or even save it).

Do you think there are any shortcuts to the months of experimenting? I'd also be interested to learn how you validated your hypothesis (one night of sleep, a few nights of emails at midnight?)

Purely empirical... just figuring out what I did last day/evening, and then noticed how sleep is affected in certain patterns.

It didn't take me months to figure it out (not OP), you might get some insights quickly. Get one of the fitness trackers(Fitbit/MiBand/apple watch) and wear the watch for a week, then take a note of the things you're doing in the evenings and see how if affects your baseline from the previous week. For me, exercise was a clear winner on the first day - I spend less time awake in bed, sleep longer and higher quality _every_ day I exercise. Others, like working late, or playing video games late took longer to figure out.

re 'answering work emails until midnight: REALLY BAD' - true for any screen time. Night shift/f.lux or glasses with a blue light filter may help a little with that

I don't understand why you got down voted. I use also a blue light filter, and it helps. Of course, if you look up stuff until 2AM...there is no device helping you there, except for a battery off! :)

I use Night Shift on all of my devices, and let me tell you, it doesn't help me get to bed any earlier.

it's not supposed to get you to bed earlier, it's to fall asleep easier once you'll get to bed.

It's to help reduce suppression of melatonin production, which can be up to 50%.

Reply on news.ycombinator.com