Updated 14/04/2016 after feedback from the designers and some more BLYNK development
As of a few weeks ago, one of the contenders for the mobile IOT crown, Blynk, has gone properly commercial. Now that’s not necessarily a bad thing – but lets take a look at this in detail.
A couple of pet peeves with Blynk were lack of multiple pages per project and the buttons – which could only say ON or OFF which was a problem if you want them to do UP and DOWN or PAUSE and PLAY etc etc…BUT I’m happy to say this has now all been fixed – with up to four pages per project and full control over button text – also the descriptions above buttons has always appeared to be unnecessarily limited in length. again – that limitation has now been fixed for Android and as far as I’m aware, IOS. Blynk has also added drop-down selection menus – and TABS (for multiple pages per project up to 4).
And how does this cost? Well I assume my setup is fairly standard, on the left you see a number of buttons – 8 data display and 4 LEDS, at the bottom a temperature display. On the right you see one of my two thermostat controls with 10 sliders for a total of 20 sliders. That is for the house and I need to replicate that for the shed so several pages of controls will be needed in the end and that’s for a small house and office – those with big houses may have much greater requirements. Since writing this however those dropdown menus may well replace the sliders (ie 2 sliders and a drop down menu to select the time slot is all I expect I’ll need).
On top of that there are a number of RGB controls needed but right now the ZERGBRA takes up way too much screen space – I’m hoping that will shrink in future – one idea would be an on-off control with a long-press to bring up an RGB or better an HSV control.
So the way this will work is that you need to “top up” the unit with energy (for which you pay) as you want more controls – Imperihome fby comparison costs a few dollars one off payment.
So let’s look at replicating what I have now – 1,000 points costs a dollar, 5,000 points costs 4 dollars etc. and there is a cost per widget – so by my reckoning to replicate what i have now will cost… 9 buttons at 200 points-1800 points, 4 of the value displays at 200 points = 800, 3 LEDs at 100=300, a history graph at 900 and 20 sliders at 200=4,000 points.
In total then just to replicate what I have now and that does not include any lighting controls or anything else comes to 7,800 points. I’ll want to replicate the thermostat for my office, so add maybe 5000 points and then there’s lighting and some controls I’m working on now – I’m thinking maybe 15,000 points without any more graphs… That comes to well over $10 (you can buy 13000 points at $9.99 and the next step up is 28,000 points at $19.99) – but an important point here – if you DELETE widgets you now get your points refunded – such is the power of the consumer as BLYNK responded to our gripes, it was not always going to be this way. I also think with the new controls I can reduce the number of widgets required substantially.
I had a fear that there might be some “draining” of the energy over time but apparently that is NOT the case so we are now looking at one off charges to get more widgets. On the OTHER hand – I seem to be sitting on some 9000 points which should be more than enough for all but the more ambitious projects. And I was concerned about the cost of those history graphs – if it were to pay for their servers that would be one thing but this is hosted on my own Raspberry Pi – on the other hand they do handle 4 inputs per graph so most installations presumably would only need the one graph.
Let’s compare for a second with Imperihome. Around $5 or so (I’m in the UK but let’s keep it all in dollars for simplicity) for the App and that’s it – on the other hand after 2 emails and a couple of weeks I’ve had zero response from them and their App, complex as it is – is not as pretty as Blynk and misses some simple basics like the LEDs and uncommitted buttons.
With Imperihome you can have as many history graphs as you like – in fact as much of everything as you like – but if you’re using Node-Red you have to handle this yourself – in Blynk it is built into the server software they give away for free if you don’t want to use their cloud. In the light of the new charges for Blynk I asked Imperihome what their timescales are for implementing these features – no response weeks later.
I think I’m happy to pay for Blynk widgets as needed – assuming that fixes are regular (right now they are very regular) and new new widgets appear regularly… There are several apps out there but currently one could argue that Blynk is the best looking and thanks to third parties (work which was done for free) there’s good, reliable Node-Red support for Blynk… but then there’s good support for node-red-contrib-ui – all THAT needs is a major visual facelift.
Whatever system I end up using will control 2 lots of heating (3 eventually as I have the house – my home office which is separate and our place in Spain), lights including RGB lights and so much more. I’ve not yet checked if I can put this on multiple devices without paying extra – I hope so.
According to Blynk Bluetooth is next. Not sure I see the urgency for Bluetooth as that is limited range – but what do I know.
Blynk development team have responded to gripes here, elsewhere and on the App store – by increasing the number of points we get, returning 100% points for erased widgets and NOT charging points for new empty projects AND the new tabs cost zero points as well which is a good thing – that is good progress and nice to see they are listening to users. I still think the cost per item for things like graphs could be reduced – but the changes of the last couple of weeks are definitely an improvement and at this point I’m prepared to put more work into this and “see how it goes”. I’m off to Spain in a matter of days and I’ve a lot of work there to update the controls for my various colour LED lighting arrangements.
There are now two types of value panels – each in two sizes the original and a larger one (very welcome) – one type of panel is straightforward and best suited for those of us in control of our data – the second allows a macro replacement so if you can only output numbers you can add for example “C” to the end for degrees C etc.
As for the programmable drop down menu – which is great – I’ve populated it with colours for one of my RGB displays which means colour control is now small enough to fit 16 of them on a page – don’t laugh – I’m heading in that direction.
Thoughts? New readers please note, some of the comments below were made before the current round of improvements.
Hi Chris,
I’m confused… After setting up Home Assistant and realising it’s interface is just too old-school, I’ve been playing around with NodeRed and it’s interface Dashboard. Now reading this I’m wondering where Blynk fits in. Can you give me some insights?
Thanks.
Well, all this talk of poking around with variables to get more tokens for Blynk..
Here’s an easier way that works (I just rebooted and it survived reboot) – let’s say you are running the Blynk server on port 192.168.0.5
https://192.168.0.5:7443/admin
Go to the user – click to edit – change the credit available – save.
Easy peasy! Looks like they meant us to do this after all.
I now have 200,000 credits – which is a bit silly but hey, why not – now I can put in as many pages and widgets as my heart desires…
This admin interface is really useful!
Blynk and Imperihome are merely two interfaces you can use on your phone – both good in their own way. I’ve always had a hankering for doing my own thing however – and straight html/websockets etc was always too much like hard work – so I went for Blynk initially then Imperihome – great if you have no creative ability at all. However, several requests to the Imperihome people over many months later asking for some more plain controls instead of trying to accomodate every system on the planet fell on dead ears – they still don’t have simple buttons and other basic components… and Blynk is great – but only for some things – if you look at their controls and apply some imagery of your own you can achieve now just about anything they have – with Dashboard – this is new – things change – 6 months ago I would have said no to Dashboard but as of the version about to be released which has my last gripe (for now) sorted (swipe) I really do think this is a good way forward for developing phone controls, wall displays, you name it.
There are loads of built-in icons due to Angular – and you can use your own images – I simply have mine stored away in a directory on my main website but if you are using this internally only you can have a folder on the same machine as Node-Red storing images. There are of course touch boards like the Nextions – and they have their place too. It has been pointed out before by others that for the cost of the 7″ Nextion however you could really use a cheap tablet – now in the past I’d say “but that involves a lot of work” – today – any old Android tablet running a browser should do the job with Dashboard.
The story is not yet complete however – to use this externally you need username and password – that is handled in Node-Red for both the developer interface and the UI – however I’m having some slight issues with the UI password in that for some reason the password occasionally gets forgotten – and if you happen to be continually refreshing the page – that can interfere with password input – WHY would you continually refresh the page – well, if there is any chance of an external event such as a timer affecting a global variable you are using (say, the temperature) you have to have some way for the screen to be refreshed.
If the Dashboard is the ONLY way things get updated this is not relevant. However I’m sure at some point, someone will write in and say “oh, that’s simple, this is what you do” – so overall, I’d say that Dashboard is going places.
The Blynk Board looks nifty. And it explains a lot of Blynk’s moves. For me it’s changed my persepctive from “why isn’t Blynk more aligned with what I want as a tinkerer”, to “Oh, Blynk has been focused on this partner product with Sparkfun and support for other hardware is a nice side feature.”
Businesses love subscriptions, a constant stream of revenue. I get that. But a lot of consumers (like me) avoid them. Blynk is competing with Particle (formerly Spark) that only charges a flat fee, no service fee, for now.
Good move by Blynk to refund 100% of energy from deleted widgets. I may tinker with Blynk now as it’s cheap to do so and I’m not penalized for making changes. But to be honest I’ll prefer node-red-contrib-ui and Imperihome because it’s less hassle and cheaper.
I think the Blynk / Sparkfun offering is not really aimed at the likes of us “node-red based” tinkerers.
It seems to me that this tieup consolidates a direction for Blynk to peel away from the need to use a centralised hub system (node red) by pushing the logic down to the devices themselves (blynk board).
For most purchasers who only want to have a few boards, this approach could be just fine and the following link outlines the detail on how to get there – and it all fits on one page!
https://learn.sparkfun.com/tutorials/blynk-board-arduino-development-guide
I don’t think this approach is suitable for those who want large scale multiple device systems (many us lot?), but for the vast majority who just want a small system (e.g.To control the climate in a greenhouse), a single board system like the blynk board will do just fine – especially as the blynk board has the brains to control things locally if the internet closes down. Consider that this solution only uses the communication channel (internet) for receiving control signals and for sending history data out to the cloud.
Consider again that this solution doesn’t require that YOU have to host a central hub system and database to store history data as the data is stored by the Blynk server and the data display handled by the Blynk App.
When you consider what most of us are doing, hosting node-red etc, the Blynk/Sparfun solution needs a LOT less effort to get things off the ground. Also, consider that the blynk board has a lot of built in sensors etc,
Pricing looks fine to me – for one-off systems 😉
Personally I’ve been giving more attention to Imperihome – it has a general API – the only problem is – a lack of general purpose controls – they’re so busy making it work with every system out there they don’t even have general purpose up/down buttons unless you use the blind control. Ultimately, Node-Red Dashboard could be the best of the lot because of the template node which lets you customise – but personally I’m having difficulty getting my head around what you can and cannot do with Angular. Do dout about it – if someone bought Imperihome and brought out a whole swath of new general purpose controls it would get my vote – and with a little database work it does do nice graphs – I have temperature and humidity graphing for several months now – 250,000 records in face (good job I bought decent SDs). On the other hand Blynk IS pretty.
something is changing…
http://community.blynk.cc/t/introducing-blynk-energy-first-paid-services/4975/52
Dmitriy – Co-Founder
Dear all we just did new Android update with STORE FIXES. Please update ASAP. Version is 1.10.1.
What’s new :
On widget removal (recycling) you get back 100% of energy;
New project doesn’t take any energy anymore;
Recycling enabled for old projects;
All Blynkers get +2000 of energy to compensate bugs in previous release;
Sorry for inconveniences and bugs in previous release as usual any feedback is welcomed!
Well, that’s a good start – let’s see how long it takes to get some of the other promised features…. that’s definitely a sensible update. That puts me up to 3,900 points. Not stunning but better than before. Problem is the history graph for example – my server, my electricity and each graph costs 900 points. That’s a LOT.
Hi Pete,
Firstly thank you for you great blog!
I’ve done some test with a local Blynk server and it’s very simple to increase the energy per (local) user to any value.
I’ve tested to change my 2000 initial energy to 999999 and now my Blynk App see like I’ve all this energy!
The only thing to do is to change a variable in the user config file and restart the server.
I’m using the local server and I did spot a var right at the very end of the user file which corresponded to my energy and of course just had to have a go at changing it – however a reboot made no difference and returned the original value – so you must be looking elsewhere or using a different version? Certainly handy to have unlimited energy for experimenting.
mmm… i argued A LOT with them on facebook when they introduced the “energy”, and they clearly stated that that’s only for the cloud service, while the local server has no limitation, FOR LIFE… had no time to try, now you are saying that there’s a limit???
https://www.facebook.com/blynkapp/posts/1002007516556471
I didn’t read the Facebook post before. I didn’t know that there was no limit for the local server. What I said is that I’ve tested with some values.
With all my tests the modified energy remain.
The variable that I changed is “energy”:999999 in the user file (username.Blynk.user) at the root of data folder. The unmodified server tested is the latest blynk-server-0.19.2.jar.
The other approach that works for me is to rebuild the server from source code (tested with Ubuntu) and change the default energy value in the file https://github.com/blynkkk/blynk-server/blob/master/server/core/src/main/java/cc/blynk/server/core/model/auth/User.java
Hi there – Well, I’ve never rebuilt the server from source (Debian) but that could be worth looking at if someone has done a Debian guide – but that ENERGY value – change, save, open to change it has changed – yes…. reboot machine – it’s back to the old value every time for me!
apt-get install maven git
git clone https://github.com/blynkkk/blynk-server
cd blynk-server
mvn clean install -Dmaven.test.skip=true
Build will be in blynk-sever/server/launcher/target folder
Oh that’s handy to know. Thanks. I ended up just over-writing with the latest version. I changed my credit to 20,000 and that was fine but then I noted the LABELS in Blynk had stopped updating – I figured it might be due to messing with the credit – but it turns out (and this has happened twice) that there’s some compatibility issue with my newly updated Blynk App and the old server. I put the latest server file on the Pi (I just rename them to server.jar and overwrite the original) and lo and behold it all works a treat with effectively unlimited credit.
I am sure there is a market for some service like Blynk. Tinkerers and hackers just may not be the intended target group. It is something like gold plated Mercedes – not suitable for me, but perfecly fit for some sheikh or a pop star.
Clever guys those people at Sparkfun.
Thanks for your feedback, guys.
It’s hard to find a model that fits all.
Some people were vocal about subscription, like “NO WAY!” , others just don’t think you need to pay for services. You will always have lovers and haters…it’s ok.
Blynk is a growing platform. We tend to experiment, learn and invent. Nothing is perfect but we are definitely thinking about getting as close as it’s possible. Can this model be changed – surely, can there be subscription – of course. Will there be energy adjustments? – yes.
What’s unlikely to happen is fixed price, because Blynk is a service. You can’t buy a car with a lifetime gas refill (though it would be awesome)
We also hope that our next ideas will be liked, because we are in a constant communication with our users, trying to understand their needs.
Peter, there will be new widgets, of course. We needed to implement payment first due to platform reasons. We really wanted to put them into release, but we had a deadline of launching a joint product called SparkFun Blynk Board: https://www.sparkfun.com/blynk
As always we are very open for a feedback on our community or over email.
I don’t think there are haters Pavel – but many of us have invested a lot of time in this – like the product and want to see it work FOR US. I personally find that coupled to Node-Red it is very reliable but right now it is not fullfilling certain needs – and clearly now it involves costs, decisions need to be made – is this the route to take or not. Do you have any timescales for the likes of button changes, multi-page-per-project etc..
Whilst a simple monthly/annual subscription or a one off fee for the app would not have been popular with everyone, at least it would have been easy to understand.
Whereas what they have gone for just sounds really complicated. On a quick read through, I have no idea what it would cost me if I were to start using Blynk again for my home control project. And frankly I don’t have the energy (!) to bother trying to work it out – especially when Node Red UI does everything that I need and does it for free.
This type of business model seems to becoming popular, the subscription, granted this is a twist on the standard model.
I have no problem with people making money but my problem is that I always have to know what my costs are going to be, this method means that I will not know what my eventual cost is going to be and what if I want to deploy that solution to my wife phone and my son’s or even a visitor is that another fee per phone. With a fixed one off fee I know up-front what the costs will be.
So I think BLYNK is not for me it looks good but there are other solutions that will do what I need for less or even free, if I invest in BLYNK then I would probably be stuck with it.
Their community page says you get back 80% of the energy when you recycle a widget.
I’ve had high hopes for blynk and fully expected to pay for the full app and many extra widgets as they were released.
But I tinker too much. Blynk, you’ve lost a potential customer with this scheme. To complicated, too much chance for waste.
Well I’m with you – I’ll spend weeks or months tinkering with stuff that will eventually not get used – but it is part of the learning process and I think they may have scuppered that. The solution is not impossible – just give people a massively higher number of “points” to start with – but then – especially for someone like myself who is enjoying their later years in semi-retirement – do I want to tie myself into even MORE regular payments… at the end of the day thought – BLYNK could come up with some stunning developments worth paying for, or Imperihome could suddenly start responding (I’m not seeing much of that right now) and provide the extra stuff needed for general purpose use hence rendering Blynk obsolete – OR Andrei who was developing node-red-contrib-ui could suddenly come back and surprise us all (my own view is that supporting dozens of systems is silly – support a central system like Node-Red and let THAT talk to loads of systems – which it DOES and does it WELL – but that’s just my view).
80% back? So they want to discourage experimenting, which usually results in using more (paid) features. This is wrong thinking and a blatant money grab – it costs them nothing (no server usage, etc.) when I add or delete a widget.
These blokes don’t understand their customers. Part of the DNA of makers is we enjoy the building process, usually to do it cheaper than buying an expensive box you just plug in. We appreciate value, and this goofy points scheme with a 20% fee every time we change something will not fly. Not to mention the gall of charging for money for an unfinished product.
I’m out. Let me know if they come to their senses and change to a more user-friendly pricing system.
you are not buying widgets, are hiring them, lol 😀
i don’t want to buy the biggest “energy” packs just to try stuff… add and remove widgets until i find a well looking and functional desing, and even if i have energy back if i remove some widgets, i don’t want to pay for something i may not need at the end… maybe a “setup” free design mode, where you are not forced to pay but have, let’s say, 30 minutes of working project until it stop working before a “redeploy”, and a “production” mode where you pay what you really are using, can be a reasonable way to do… but anyhow, prices are tooooo high, right now…
bye bye blynk… even if all the promises you received by email where already implemented, it’s waaay too much for what you get, really…
they were always proud they 5 stars reviews on play store… let’s see how customers react…
if a one time fee like imperihome is too low, then go for an annual fee, but fixed, not this bu**s**t…
I’m torn.
Blynk does look good. The features it has are nicely done (mostly), but as you noted, it needs a few more things.
I want to support people who make cool stuff I can use. I guess they’ve burned through the $49k they raised on Kickstarter. I like the idea of paying for what you use so small projects don’t cost too much. But the points model seems too much, I prefer a simple one-time fee like Imperihome.
What happens if you delete a widget, do you get that “energy” back?
I’m still just tinkering, and not inclined to go all-in on Blynk and deal with the cost. I can prototype just fine with node-red-ui and Imperihome. Perhaps if I implement these systems in my home I may be willing to pay for a nicer UI (Blynk).
Actually I can answer that – yes you DO get it back if you delete a widget. I think perhaps I’m more concerned at the idea of the points slowly winding down – and what if they go bust or move onto something else – will all of this still be supported? With Imperihome – you pay your money and it is yours – end of story – but then – let’s see how quick THEY are to develop new controls – currently missing some general purpose ones I’ve mentioned in the blog.