Release 3.116

RainViewer Radar in Silver!!

A special thanks to our kind friends at RainViewer for the data. Please do go and check out RainViewer on Google Play. Have a look at their pro version too.

Just edit place and add RainViewer Radar data. Yay!

3 Likes

@duane just to be clear clear cause we all know about my thought :station: derails Rainviewer is only in hindcast?

Yeah - I enabled the new radar data - went to the map and couldn’t see any radar. Changed map types, nope.
Finally I was messing around with the map, and I could only see radar data in hindcast - or past time - not present time. Which is weird.
I would highly recommend making the radar visible in current time - and in the past.

Radar for past time is not weird. Radar is measured data - it’s not hindcast or a forecast. By the time the radar scans the sky, the metservice processes and put the data on their website, and RainViewer graphs that data and puts it in the composite image, and Flowx servers processes it (checks every minute) and Flowx app downloads the data (checks every 200 seconds), it can easily take 5-10 minutes.

So it’s extremely difficult to get real-time radar images - hence it will always be in the past and you cannot make it visible in the current time because that would be wrong.

I think the best option is that when you turn on radar and the time is ahead of the current time, Flowx will snap to the latest radar image.

In the future, we plan to predict the future 1-2 hours of radar - this is a prediction, i.e., a forecast, which will solve this problem.

Check out the RainViewer FAQ

3 Likes

@duane Keep Up the Great Work. If I was smart :joy::joy: I wouldn’t predict past 2 hours. This next part might sting a little for others but people :sob::sob: moan and complain that the weather person got it so wrong know it’s a “Prediction” the radar shows a return from particles in the air what it can’t and will never be able to show is “What’s actually hitting the Ground” the old saying goes complainers will always complain and haters will always hate. You are Doing an Awesome Job along with Rainviewer :+1::+1::+1: not saying anyone on here is complaining just a statement I can see the likes on my post being unliked and the hate mail rolling in

1 Like

The forecast will be in gold. I might add a warning for users to understand exactly what the prediction is and that it’s just a prediction.

I might also look at analyzing the error over time and report this to users.

3 Likes
  • Flowx will snap to the latest radar image -
    That would be absolutely great.
    The only issue I had was having to scroll back in time to see a radar image - that was cumbersome…
    Anyway thanks for a great product.
1 Like

@Windsailor, yes, I was finding the scrolling cumbersome too - even the vertical scroll. Were you using vertical scroll?

I’m wondering what to do there. I’m thinking of allowing users to change the sensitivity or base the speed on the speed of the swipe. For example, a slow speed will swipe over a hour and a faster speed will swipe over a 24 hours?? It might be worth a try.

2 Likes

@duane I use both horizontal to get to start point vertical to view hour by hour

1 Like

Double-tap like YouTube.
Double-tap the right side for a set forward jump, double-tap the left side for a set backward jump.

Leave swiping alone until you have a universally brilliant change idea for it.

1 Like

As radar is the context, wouldn’t it be cool if…
Double-tap triggered an animation time scroll back to the original (last) time mark.

So, if you double-tap the left side of the map, and the set backward time jump was 6hrs, the double-tap could slow scroll time (and the map) forward 6hrs, hour-by-hour, min-by-min, whatever; and vice a versa for double-tap on the right side.

This feature-add should allow for increasing increments if multiple double-taps are done.

Using double-tap to replicate an existing gesture seems a waste. You know how much I protect my gestures :slight_smile:

I agree about leaving swiping alone until I get it right, or having the option in settings to change the mode.

I think we’re close so we don’t have to make major changes.

1 Like

Hmm… Not sure I agree. About swiping being close. I’m still always getting lost as I swipe in Flowx. Yes, when Flowx has my full attention and dexterity, I can get pretty good info from it… but my attention (focus) quickly wains and then it’s just an app to swipe out of.

I’m simply struggling with swiping a set amount repetitively. In keeping track of the days of the forecast. Keeping track of day vs night.

Up-swipe vs side-swipe is helpful. The timestamp is helpful. Setting a shorter forecast (is not desirable) but helpful. But the variably of swiping is both Flowx greatest benefit and downfall. It needs some killer refinement.

The double-tap feature might well be the best solution but I’m reluctant to rush into implementing the first idea. I take my time to think about features because they take time to code up and removing a feature is a very painful experience - users hate change and when I take something away, I will get a lot of hate.

I felt we prematurely added “single-tap” to the compare graph screen and then I had to remove it. I don’t really want to go through that again. It takes development time away from other higher priority features.

So just like the quote “hire slowly, fire fast”, I “implement slowly and remove fast”.

It has taken me 2-3 months to get radar working, I plan to release it with in a sup-optimal state and fix the issues over time.

2 Likes

Nobody does.
You are the gatekeeper.
Wise gate, tiny steps.

Wish I believed a forum was a good method to develop a roadmap. I don’t. Hope you have sage methods of your own to track yours.
I can only hope our nuggets of a grand design, inspire a thoughtful grand design.

You’ve got a good track record going. We’re enjoying the engagement.

1 Like

By far the forum is the best method. I know because I’ve seen both sides.

Without the forum, I was repeating myself and having this conversation with many users separately with not interaction between users. So a lot of wasted time.

I’ve seen ideas develop on the forum much faster.

With the forum, it’s just managing many expectations. Different people have different priorities. Maybe this is the thing that is missing. Users don’t see the long list of to-do and their priorities.

Maybe I need to add this list to the forum.

3 Likes

I feel ya there @duane

1 Like

How about the list of to-dos, features in progress, and requests in a format similar to redit or stackoverflow? Meaning that the order of the items in the list is automatically arranged based on votes from users.

Just to be clear, I’m not trying to pull focus from development of the core product, to develop the forum! I just think that this must be a common problem, and a done solution is probably sitting on the internet somewhere just waiting.

Maybe some other forum user here know exactly where?

1 Like

Another sugestion for the use of double-tap is to toggle between different groups of pre-defined combinations of settings, for example:

  • radar only (with the swipe not being able to reach times not covered by the radar data),
  • forecast only (old default, before radar) ,
  • hybrid-1 (as is default now), amd
  • hybrid-2 (with the swipe not being able to reach times not covered by the radar data).

Just a question what is wrong with the way the app works now? Why is tap this and better than swiping? just curious