DCC and innovation stagnation

[daniel_leavitt2000]

Before the money crunch last month, I had been looking at DCC systems to power the Boston Line. One thing I kept coming up with was that current throttles are totally inadequate for today's market. Like it or not, the smart phone is the universal controller of life. It selects the channel on our smart TV. It turns out the lights or changes the temperature. It starts your car and opens the garage. It streams. It plays. It even makes calls. And so far, the DCC manufactures have totally ignored it.

And this makes no sense at all. There are many, many benefits for a dedicated DCC app instead of a throttle:
1. No FCC radio certifications. Just plug the base station into a router and connect the phone to the network. FCC testing on throttles had been cited many times by manufacturers as to why we don't see new throttles on the market.

2. It costs less for them. All the biggest tech companies are software. Why? Hardware costs a lot to make and margins are small. Offloading throttles and radio to a 3rd party would greatly reduce the cost and complexity of a base station. That in turn would lead to:

3 Lower costs for the consumer. We have cell phones laying around. Chances are you have an old router or two as well. All the new hardware a costumer would need is a base station and a few decoders. That is basically the same hardware as a transformer with a few added circuits.

4. But the company had a lot of opportunity to make profit. If each throttle app cost $50, the consumer would still save an average of $200, while the company would get nearly $50 in profit. More than they would get selling the hardware throttle.

5. And this would lead to a lot of innovation. Think self mapping layouts, real time train detection and tracking on a tablet. Cab video over Wi-Fi directly on your throttle, computer or tablet. Automatic train building by reading the car numbers with the camera. Pre programmed switch puzzles tailored to your track plan produced by the app. Creation of obstacles during operations like a car parked on the tracks or an oil leak in the engine.

6. But most of all it would reduce clutter. I'm aware of JMRI. I hate JMRI. Let's follow the route it takes JMRI to get from your phone to the engine: Phone brand A using app brand B connects to a router brand C that connects to a computer brand D that uses JMRI (brand E) to connect to the base station brand F running to a booster then to the tracks and finally to decoder brand G. That is insane. How many potential points of incompatibility are there? This is just not tenable for a layout my size. And even if it was a small layout, why... Why would I spend  days on end trying to untangle code and chords just to get the trains to run on a crappy app on my cell phone. This complexity could be greatly reduced by using one company from app to base station, using only the router and phone from other companies. Most of the coding would be internal with no need for the modeler to tinker.

The worst part is, I have sent this plea to several companies: NCE, Digitrax, ESU (which even had a smart phone app for its throttle but won't make it available to other Android devices) and no one has responded. Not even a form letter.

[Scottl]

You make a lot of good points.  I'm sure phones or the next equivalent device will continue to make inroads into this, probably by moving to Bluetooth communications to the decoders.

But one thing that I find still remains a barrier is that many modellers are still DCC-adverse, and many of them remain DC only or reluctant to take on the complexity of DCC (there is some, no matter how simple they are to use).  The packaged DCC hardware gets around that by combining all of the coding and hardware together nicely, so it offers something that the JMRI/phone/other stuff does not.

I don't see it is as stagnation so much as DCC systems filling a particular niche.  I love my NCE system and for the $150 it cost me, has been a great investment.  I can still connect with JMRI and make use of a range of hardware options that are pretty seamless to use. 

[Ed Kapuscinski]

I've done the smartphone throttle thing. It's handy in a pinch. But I don't like the user experience of the whole thing.

For one, I don't like losing the use of my communication\information device when running a train. I think I once caused a collision at a friend's layout when I got a phone call.

For two, I don't like the ergonomics of it. I actually LIKE the Digitrax knobs, and the NCE wheel isn't horrible either. But, simply sliding my finger? There just isn't the precision one would want.

[railnerd]

PiSprog One lets you build a fairly interesting standalone DCC system with a RaspberryPi:

   http://www.sprog-dcc.co.uk/pisprog.shtml

Yeah, for now you'd still need to use JMRI, but that may not always be the case.

(This is about to go into volume production this week, and isn't yet available worldwide just yet)

-Dave

[jagged ben]

Where's that thread where I made many of the same points as Daniel?
[EDIT:  here it is. ]

To address a couple points that have been brought up in this thread...

I've done the smartphone throttle thing. It's handy in a pinch. But I don't like the user experience of the whole thing.

For one, I don't like losing the use of my communication\information device when running a train. I think I once caused a collision at a friend's layout when I got a phone call.

So use your old phone.  Or if you don't have one, do what my friend did and buy a junker phone that's a couple years out of date.  Maybe the SIM card reader is broken but the WiFi still works. 

For two, I don't like the ergonomics of it. I actually LIKE the Digitrax knobs, and the NCE wheel isn't horrible either. But, simply sliding my finger? There just isn't the precision one would want.

I agree with this.     For that, there are now knobs that you can afix to your touchscreen.   

...
6. But most of all it would reduce clutter. I'm aware of JMRI. I hate JMRI. Let's follow the route it takes JMRI to get from your phone to the engine: Phone brand A using app brand B connects to a router brand C that connects to a computer brand D that uses JMRI (brand E) to connect to the base station brand F running to a booster then to the tracks and finally to decoder brand G. That is insane. How many potential points of incompatibility are there? This is just not tenable for a layout my size. And even if it was a small layout, why... Why would I spend  days on end trying to untangle code and chords just to get the trains to run on a crappy app on my cell phone. This complexity could be greatly reduced by using one company from app to base station, using only the router and phone from other companies. Most of the coding would be internal with no need for the modeler to tinker.
...

Hmmm.  I get what your saying, but I think you're exaggerating the problems and simplifying the solutions.    One of the advantages we have is the DCC standard:  once you get JMRI talking to the command station, things are relatively foolproof.    And what about the size of your layout makes it untenable?  The bigger the layout, the more tenable it is to have a permanent setup of all those components.    And  what makes you think that having the app come from the same manufacturer will simplify things?  Do you really trust most of the manufacturers (especially Digitrax!) to make something more simple and user friendly than what we already have?   Don't get me wrong, there's definitely room for innovation.   But one of the nice things about JMRI is that it's not proprietary and thus anyone can make an app that works with it.    In fact, I don't know why we don't already have more MR throttle apps to choose from.  Evidently the population of users is small enough that there are not enough people who have the skills+interest+time to make better throttle apps.  That bodes badly for innovation regardless of the general approach. 

I think there's some simple steps the manufacturer's could take to make things much better, especially for entry level DCCers.  I think it's a travesty that the USB connection isn't built into the command stations yet.  [EDIT:  I forgot that Digitrax finally came out with that this year.]   And a command station with a WiFi server that would eliminate the computer would be a good addition.    Ideally there could be an addition to the DCC standard such that any app could communicate with any WiFI capable command station.

[Philip H]

Have you looked at Bluerail? It's been on Model Rail Radio repeatedly.

[jdcolombo]

I'm sure Peteski will suggest this thread be moved to the DCC forum 

But in the meantime - have you looked at the ESU Ecos and the Mobile Control II?

The Ecos plugs into your home network.  The Mobil Control II is everything you'd want in a smart phone app, but has a real knob and buttons.  It communicates with the Ecos via WiFi, and one version of the throttle even has its own WiFi hotspot you can configure separate from your home network.

Like Ed K, I've used the WiThrottle app on my Apple iPhone and have two old iPhones dedicated to this app, but I simply prefer having a knob and hard buttons (actually, a toggle switch is even better) for direction control of a train.  If you really want to use your smartphone, while it is true that JMRI runs through multiple interfaces, I've never seen it NOT work - it's really pretty much foolproof IF you already have a computer interface for your command station.   But with the ECOS system, you only have a connection to your home network to worry about (and can even skip that if you want to set up a dedicated network for train control, as noted above).

The Europeans are much more advanced in their user interfaces for DCC than the US companies.  ESU and Zimo both have systems that use WiFi, not some proprietary radio system that often fails to work correctly, and both have user interfaces that a three-year-old could understand.   And both are relatively expensive, but you get what you pay for (actually, my total investment in a Digitrax DCS command station; PR3 interface; cheap computer to run JMRI; etc., I've probably invested as much as a new ECOS system with at least one Moblie Control II throttle).

John C.

[Maletrain]

Bachmann seems to be headed in the direction you are proposing.  Have you looked at what they are doing?

I think that your idea of using BlueTooth has some merit, but, does that really get around the FCC?  BlueTooth and WiFi are TWO-way communications, so the decoders would be TRANSMITTERS as well as receivers, and that would make them FCC approved devices, wouldn't it?

I also do not like the ergonomics of a "smart" phone screen.  Considering how many "butt-dailed" wrong numbers I get each day from other peoples' smart phones, I can just imagine all of the unintended commands going out to running locos because somebody held the phone in such a way that they unintentionally touched the screen while shuffling car cards and uncoupling picks during a switching move.  Not to mention that 128 steps on a smart phone screen are going to be spaced much closer together than the width of my thumb.  So, simple virtual "slider" controls on a smart phone surface seem like a step backwards in the quest-for-the-crawl. 

And, considering how often my computer hardware and software and cell phone hardware and software become "obsolete" and "no longer supported", I think my NCE DCC system will be reliably useful for a lot longer than any smart phone app.  So, I think it is best to divorce the throttle from the phone, entirely.  That does not mean that the Bluetooth and/or WiFi technology could not still be used, but with a more purpose built device for a throttle.  Perhaps if the beginner could get started with a cell phone with a cheap app and a box that plugged into a pre-existing router, it would help people adopt DCC.  But, I don't think that is going to satisfy the serious operations modelers.

[peteski]

I'm sure Peteski will suggest this thread be moved to the DCC forum 

You think he might? 

The Europeans are much more advanced in their user interfaces for DCC than the US companies.  ESU and Zimo both have systems that use WiFi, not some proprietary radio system that often fails to work correctly, and both have user interfaces that a three-year-old could understand.   And both are relatively expensive, but you get what you pay for (actually, my total investment in a Digitrax DCS command station; PR3 interface; cheap computer to run JMRI; etc., I've probably invested as much as a new ECOS system with at least one Moblie Control II throttle).

Yes, that!  ^^^^^
We are greatly limiting our view of DCC by only looking at US-made systems.  European DCC systems (while some might seem a bit gimmicky and not quite operation-based in the sense of how US modelers run their layouts) are technologically way ahead any of the US systems.

[jagged ben]

Have you looked at Bluerail? It's been on Model Rail Radio repeatedly.

See: https://www.therailwire.net/forum/index.php?topic=39565.0

This is not a replacement for DCC in N scale though.  Not yet anyway.

I should update that thread.

[squirrelhunter]

I'm with Ed, operation via phone is underwhelming, and not precise enough. I don't love any of the DCC throttles either.

I think the real future is for someone to build a minature control stand that is hand held. So 8 throttle notches (or 16), a brake handle, etc. There is a company, I forget who, that posts on the Proto Layouts Yahoo group that is working on this, to the point I think they had a mockup at the St. Louis RPM.

[mmagliaro]

Perhaps my view on this will be disregarded, since I'm not a "DCC guy", but I have operated on DCC layouts, so here is my impression of this human interface issue.

A DCC controller, the vast majority of the time, manages only a few functions - speed, direction, and less frequently, engine selection.  Everything else it can do is rarely used compared to those 3 things. 
 
Why would it be a gain to create a user interface based on touchable screens and menus and submenus when most of the time, we reach for 3 things: speed, direction, engine selection.  Maybe that's why some people in here mentioned that they like the dials and knobs.

Digital interfaces are not human-intuitive things.  They are interfaces that force humans to train themselves to adapt to the interface.  So my vote would be that rather than move to be more like a phone, or put the interface on a phone, DCC controllers should try to look more like human-friendly analog control boxes.
 

[peteski]

Max, you didn't consider all those little function buttons needed to maximize the fun: headlights, ditch lights, horn (even playable horn), bell, whistle, and all the other fun sound bites people love to play with.  Maybe even remote uncouplers.

[Cajonpassfan]

I'm with Ed, operation via phone is underwhelming, and not precise enough. I don't love any of the DCC throttles either.

I think the real future is for someone to build a minature control stand that is hand held. So 8 throttle notches (or 16), a brake handle, etc. There is a company, I forget who, that posts on the Proto Layouts Yahoo group that is working on this, to the point I think they had a mockup at the St. Louis RPM.

+1

I too think the "virtual" interface via a smart phone is lame. It's like a computer game, ho hum
I prefer the tactile experience of controling my locomotive and train with physical appurtenances; knobs, levers, switches, buttons.

That said, (and I use NCE) I wish we had a dedicated diesel cab, with 8 speed notches (with adjustable momentum) brake, and sound controls, and a dedicated steam throttle with a "Johnson bar" (the NCE rotary actually does a pretty goods job of that) as well as brakes and appropriate sound controls. Again, adjustable built-in momentum to emulate real train  behavior  would be desirable. A simple toggle for direction is preferred.
My two cents worth...
Otto K.

[Chris333]

Max may be on to something. As I was sitting on the crapper one day... reading MR I saw some DCC ad on the back page. It showed a photo of the controller and I counted around 30 buttons. This reminded me of my TV remote. The TV remote has 48 buttons, but the only ones I ever touch are 14 of them. The rest mean nothing to me.

When I see a 30 button remote and then think of how simple it is to control a train I just roll my eyes. I mean on/off, Direction, and speed that is all I need. That and I don't think controlling trains should cost more than $50. You can buy a fancy million dollar chandler, but it only take a $1 switch to turn the lights on.

BTW I've never even held a DCC controller or ran a train with it, but I know my TV remote sucks balls.

Let the down votes commence!