Friday, February 3, 2017

Next Generation DCC: When Does It Come to North America?


Since I started this blog and throughout the construction of my current Dunes Junction and Old Line Corridor layout projects, DCC has been a series of surprises and delights: it continues to add unexpected joy and focus to the operation of my trains, and has occasioned a significant bit of fun learning. Nonetheless, every step in my DCC evolution seems to point the way to more and better ways of doing DCC.

I'm not the only model railroader who has noticed that DCC adds to train operation. Tom Barbalet of the Model Rail Radio podcast said it best when he described (more or less, and I'm paraphrasing) analog DC control as operating a circuit, but DCC as control of a model train. To elaborate on Tom's observation, DCC adds additional dimensions of realism such as sound, lighting, and other features, and also other less obvious realism opportunities such as consisting (MUing), speed limits and tables, and various compensations for poor or erratic operating characteristics.

Fun learning first focused on wiring and track for DCC but quickly progressed to decoder installs and programming. More advanced programming and the appeal of wireless walkaround control using smartphones led to use of the Java Model Railroad interface (JMRI). The more DCC tinkering I've done, the more DCC gear I've accumulated. My first DCC system was a Digitrax DCS51 Zephyr, and to facilitate its use with JMRI and WiThrottle, the Zephyr was joined by a Digitrax PS3 module. The Digitrax Zephyr is an ideal starter system, particularly for a small layout such as Dunes Junction or Old Line Corridor.

However, the Zephyr, PS3, and a laptop with JMRI installed--along with all of the 'wall warts' and power cords--was a space-hogging eyesore, particularly for a compact layout like the Old Line Corridor. To streamline my JMRI and control rig, I acquired a SPROG3, which is essentially a small USB box (less than half the size of that Digitrax PS3 module) that turns a JMRI-equipped PC or Mac into a mini command station. It's a fairly elegant solution but requires a little fussing and DIY spirit to get working correctly, and it does require use of WiThrottle for multiple train operation. In other words, the JMRI-SPROG3-WiThrottle combination rarely turned right on and worked without a reboot of host computer, iPhones, or other tinkering with JMRI settings.

I alluded to my next and latest bit of DCC gear in my locomotive mugshotting post: I recently acquired a second-hand ESU ECoS 50200 command station. Trevor Marshall's blog posts on his adoption of the ECoS system pointed me to this system, and subsequent research helped cement my decision to add the ECoS to my model railroading wishlist. The big color touchscreen was a draw, as well as simply executed but extensive controls, including color photo loco icons. However, an even more attractive feature was the ECoS' built in networking and connectivity features--this modern DCC command station would also tame the rat's nest of cables, adapters, and wall warts that accumulated when I connected my previous DCC systems to my network and computers, and hopefully simplify interfacing JMRI with my DCC.

As Trevor noted in his posts on the ECoS, its capabilities are overkill for most North American modelers. It supports graphic control of a whole layout--imagine a digital, on-screen display of an entire layout, not unlike an old-school analog control panel with switches and whatnot. It also supports a number of legacy European alternatives to our NMRA standard DCC--imagine a DCC command station that could also handle Hornby Zero One, PSI Dynatrol, and DIY CTC16, in addition to DCC. It will also handle Maerklin three rail trains and communicate via a few different proprietary European network standards. All of these features and capabilities give the ECoS its hefty street price--around $700 in the US--which made this a 'wishlist' rather than a 'buy it now' item. So when a used ECoS at a bargain price came calling, I answered.

So far, my ECoS is living up to its promise. Its home is a keyboard drawer under Ivy City yard on the Old Line Corridor, and it joins my home WiFi network via a single IoGear Ethernet-to-WiFi adapter--no more ungainly direct cable connection between my railroad and network. It joined immediately with JMRI, warts and all (more on that shortly). When I'm feeling frisky and want to do walkaround operations, I use the TouchCab iOS app on a pair of decommissioned early model iPhones. (The Android-powered ECoS Mobile Control II is still very much on my wishlist, needless to say). And I'm pleased as punch at those color loco icon mugshots!

What hasn't lived up to the promise is access to the ECoS via computer. ESU has made it easy to view and backup configuration settings via web interface, and the web interface is how color photo loco icons upload to the ECoS. But changing the ECoS' configuration via a connected computer is not straightforward--aside from uploading loco icons, there is little that can be changed or added to the ECoS via ESU's 'stock' web interface. JMRI, once properly configured, does access the ECoS over a local network and in its current build will allow what amounts to a one-time upload of the JMRI roster to the ECoS. However, the ECoS and JMRI do not truly synchronize after that initial upload. I am currently working with the JMRI user community to develop a repeatable, on demand capability to upload JMRI roster entries to the ECoS, which will be an important step toward achieving true synchronization between JMRI and the ECoS.

This standout piece of DCC kit and my troubles getting it to work comprehensively with JMRI point the way to the future of DCC.

First, the ECoS' look, feel, and user experience (UX, in industry parlance) are superior to practically everything currently on offer from US DCC manufacturers. The touch screen and graphic user interface should be more or less standard in the industry--Ring Engineering is the only North American train control product I know of with such an interface, but it is not DCC. While the ECoS is admittedly expensive in comparison to US starter DCC systems, I strongly suspect that its compatibility with multiple control and networking standards and dual controls are more expensive cost components than its main UX features, the color touch screen and menu-based user interface.

Second, networking and computer access for configuration and programming should not be an afterthought in DCC products. The ECoS' built-in Ethernet should also be a standard feature of DCC systems, and for that matter, WiFi would likely be easier, cheaper, and more user-friendly to incorporate. Networking and WiFi are currently built into a dizzying array of technology today--the 'internet of things' we've all been hearing about--and it should now be easy to cheaply incorporate wireless connectivity into new DCC products. 

An aside: I'm surprised that JMRI still hasn't been joined in the marketplace by either a standalone competitor, or by a manufacturer-specific alternative to or customized variant of JMRI. The web-based computer access to the ECoS offers a tantalizing glimpse of the possibilities here--I, for one, would pay money for a "it just works" PC/Mac software product with the functionality of JMRI and the simplicity of ESU's web interface.

I'm thinking of a Digitrax Zephyr- or NCE Touch Cab-like starter system with basic, small layout capabilities that dispenses with the keypad and LCD/LED alphanumeric display in favor of a touch screen and menu picks, and adds WiFi. An interface that works with common home computing devices (Windows, Mac, iOS, Android) would allow access to advanced system settings and loco rosters/ If nothing else, such a system would appear more contemporary with other common non-model railroad devices and have greater appeal to beginners and younger model railroaders, in addition to be being more intuitive and fun to use than endless key presses and cryptic abbreviated messages on LCD or LED displays.

Already, European manufacturers other than ESU, such as Piko and Roco are introducing a new generation of basic yet graphic-intensive, network-enabled DCC system that take their design queues from the ECoS--when will American manufacturers like Digitrax, NCE, MRC, and CVP join the trend with next-generation DCC?