The point is that the NMRA should have specified a network so that the product could interplay with each other. Their move basically drives the consumers to a single brand and not allowing them to mix. A Lenz stationary decoder does not fit well into a Digitrax layout without losong a lot of its capability. Had the NMRA specified the protocol and the physical network and then a standard network device interface, then a NCE throttle could be used on a Digitrax layout throwing turnouts using Lenz stationary decoders. But this does not work due to all of the proprietary development done by each manufacturer. It is what chains DCC in the dark ages of technology. From an engineering standpoint, sorry occupational habit, the DCC industry is almost in total technical anarchy. This is what makes this so difficult for most folks. As long as the NMRA shepards the resopnsibility for DCC, Anarchy will reign, because it is clear they know almost nothing about electricty and electronics that are not operating using DC. This is best illustrated by thier desire to chose a CAN BUS, something to place a band aid on a major wound. Until they really take a look at a total re-write of the spec, it will just get a lot worse before it gets better. If Digitrax and NCE and the others were smart, They would all dump NMRA and work on a common interface and a common network and common device messages, then the parts could be interchangeable between manufacturers. And they will still be able to add features for product differentiation. And you can get to where you can plug anything in and have it work. That is where we should be now!
that helps the consumer but not the producer (took marketing classes for quite awhile so this just comes to me.) if digitrax and NCE where to make compatible systems, then more then likely the consumer will choose it based off of price. (law of supply and demand). if digitrax has there reverser going for say $50 and NCE is at $75. than NCE won't be able to sell as many units and be forced to back off on price which will lead to things like job losses, etc back at the plant. right now the DCC manufactures are competing against each other. which is what businesses are supposed to do. The NMRA forced them to make a common decoder type and that had an effect in the business for awhile. I can blame myself for this, I buy digitrax DH123's because there cheap and have decent features. I only buy more expensive decoders because i want a soundtraxx, or because I need something besides the DH123. however my system of choice is NCE. if NCE and digitrax used the same system, then I would just go off price as before, which if enough people do can harm company's. We did simulations back in marketing classes but our products where private jets and parts for them. we did something similar to this and about a month after we went to making products that had similar qualities, out of the 18 companys in the simulation, we where down to 5. I'm not promoting just the businesses here. I can see the NMRA system helping the consumer, but do the guys who work at the factory's a favor and think about this for a second.
Geep_fan is right. The manufacturers don't have a lot of incentive to either be compatible or innovate until someone outside the current ecosystem comes along and disrupts it, for instance the NMRA or a new competitor.
The NMRA standardized the track interface (actually, they adopted one that was offered). All the manufacturers saw the value in this, as arising tide floats all boats. I believe that they do not see similar value in a throttle bus standard for the reasons described above. A little regulation (standardization) is necessary. Too much stifles competition and thus innovation. As to the ethernet/DCC debate and comparing it to cell phone standards, one of biggest contributors to the complexity of the 3GPP specs is the need to provide backwards campatability while supporting forward evolution. A DCC to ethernet migration would have many of the same issues.
They already make PC to DCC interface cards. There's your backwards compatability. I don't buy model trains and model trains stuff, because it employs people. Don't get me wrong, I was unemployed for 5 months (I was lucky) at the end of last year and I want high employment, but I'm sorry, as a consumer, my only job is to buy stuff I want at a price I'm willing to pay. If Digitrax or NCE can't figure out a way to innovate to justify higher marketup and/or reduce costs to keep competitive, then they should go out of business and some people that do have the drive and innovation to make it work will step in. And lets be honest here, The proprietary market works. Look at Apple, but then, the standards based market works too. Look at the PC, or Google Droid, How about HDTV? The US government forced companies to make HDTVs to their standards (The grand alliance was mainly industry engineers) Sony (and Japan in general) had an HDTV system already and the FCC said no and made a better spec that was much more open and forward thinking...I don't think the TV vendors are complaining. Similarly, and standardized forward looking control bus spec can be a spark to innovations and further sales. Companies like Digitrax and NCE can work on making unique value added hardware and software, they will be forced to make better products, because they haven't hooked you in to their system now. I'm sorry, but that's the way almost all of our best systems work. TV, Networks/Internet, Cell phones, Heck, Cars. How about CD players? DVD Players? I mean come on, in the real world outside the trainroom, there are thousands of businesses making billions of dollars on standardized consumer products.
At the risk of branching this off in a totally random direction, I think it is worth noting there are two different camps here: Vendor lock in - vendors who get you to buy into proprietary platforms where it is difficult or impossible to get your data out of the locked-in system. Vendor stickyness - vendors who allow you to get your data out, but make it so convenient to keep your data in that it is worth your while. The key point here is to realize that to many folks these look exactly the same, but in fact are totally different. There are no pure examples of either, so I can't give you clean cut cases. However, I can give some examples. Microsoft Word/Excel update new binary formats each revision, and use tools (patent, trademark, copyright, obscurity) each revision to keep your data in a totally proprietary locked down format. Apple iPhoto manages your pictures for you, but they are in fact standard .jpg's on disk and the "database" is a standard XML file. You can extract all the data in both, but if you just use it your photos appear on your desktop, apple tv, iPhone and iPod automagically. If you want to convert to something wholly different, they are identical. However, if you want to convert to something similar, they are not. With lock in conversion is typically impossible, with a sticky situation it is usually possible, just more work than staying within the ecosystem. Right now we have vendor lock in with all of the DCC vendors. I would much prefer, as a consumer, a model where all the vendors had a sticky situation. Make it so easy for me I don't want to switch, but if I want to enough I can. It's actually more powerful, as you have the same ability to retain customers but also can proclaim your are open.
Microsoft Word/excel is both a good, but also a bad example, because you can transfer your data to open office or other tools, but you risk losing all the nifty unique features that Office has. I'd point to iTunes as a good example of vendor stickiness. Ignoring DRM content which is only partially Apple's fault. You don't have to use iTunes with and iPod, you can use other devices and you don't have to use iTunes to organize your iTunes downloads, you could use another program and there are many with as good or better features. MP3, MP4 M4a(aac), they are all readable by pretty much any music player and MP3 player, but man, once you've chosen iTunes its just easier to stick with it. Yes, you COULD switch, you wouldn't lose much info, but if you used the propretary extended meta data, that would be gone, if you let iTunes sort your library, well good luck rebuilding your playlists. Once you've chosen. It's just easier. I would be thrilled to see a common standard and have the manufacturers work to "plus" what they provide to distinguish themselves.
There's quite a bit of talk about the NMRA not being fit to deal with DCC specs and forcing manufacturers like Digitrax and NCE to use a new bus spec. Well, to take the second point first noone can force the makers to do anything. Picking up the Microsoft analogy the EU might step in if serious monopoly issues arose, but that'd only happen on April 1st in a frozen hell . NCE et al have to be persuaded that it will damage their business not to join in. So who is going to manage the 'new' spec AND represent the modelling community in persuading manufacturers to use it, if not the NMRA? Kalmbach? Trainboard? An alliance of the manufacturers? The latter is most likely, but then we'll get what they feel they can give us without pushing themselves too hard. In the long run outfits like IEEE and NMRA may be sluggish and techinically a bit behind, but they are the best we have that isn't maker led.
Your last sentence there sums the problem nicely. In the mass consumer market there are indeed vast quantities being made and sold by thousands of companies all over the world to millions of consumers all over the world. What do you think all those doublestacked containers are full of? DCC is a cottage industry by comparison. I believe NCE has about 6 employees (correct me if I'm wrong), and it wouldn't surprise me if Digitrax has 10. They don't have huge budgets and R&D departments and if they go out of business there likely won't be a flock of 'drive and innovation' waiting to step in. (I'm not suggesting we keep them in business artificially, just pointing up the reality of a niche hobby.) Incidentally, I'm not sure I'd call the PC standards based since the vast majority of software requires you to use Windows ... And as to DVD etc, well look at the mess that eventually saw Bluray arise as the 'standard'. You really can't rely on market forces to get the 'best' solution (as the recent banking crisis has conclusively proven )
The NMRA has outlived its usefulness. Do you really think NCE, Lenz, Digitrax and all of the others give rat backside about the NMRA? They only agreed to a comon mobile decoder and signal standard and then all ran off on their own. They played nice just long enough to achive a small measure of consistency. NMRA sytem Certification - what a joke. A half thought out spec only means a half backside certification. I will admit some culpability in that when I started, I just bought a system and started tinkering. If I had done the proper research and examined the specification, I doubt I would have converted to NMRA sanctioned DCC. I may have gone to Marklin or something else. The part that is really silly about this whole CAN bus, is that it is not needed if you pull your head back and look outside the normal DCC and wiring tunnel vision. This is also the concept of thinking outside the box. Just think of it as the small child inside us all that keeps asking why. What is the problem - large or complicatd layouts generate so much traffic that it starts to affect overall layout performance and reliability. Suitibly demonstrated by the Derby City Express experiences. What causes the problem - tying all of the auxillary devices to the track bus for communications. The solution is to look to the root cause. Aux devices tied to the track busses. Why? Because that is the way it was done in DC Why? Because that is the way they modeld DCC Why not just have another wire for all of the devices? Good point! The solution is isolation of complicated areas from the rest of the DCC layout. Here is one way to fix the layout so you will never have the issues needing the CAN Bus. This was derived from discussions with the principals at RR-Circuits and Team Digital. This is a plan for Digitrax because that is the system I use at this time. By the way, this was developed to solve a potential problem for NTrak Modules where Occupancy Detection and Signaling are to be implemented. We will start by taking my demostration module set and having it become it's own Power District. This is neccesary for Occupancy Detection. There are bypass busses for the RED. Yellow and Blue lines, as well as the White and the Brown busses for power supply distribution. This maintains the interoperability with other NTrak modules. There are no connections to these busses to any device or track segment on the modules. From the booster, one cable returns to the Command Station using port A and port B goes to a UP-5. Using only the UP-5 throttle port on the side connect to other UP style devices in other parts of the layout to extend an isolated throttle bus. The track power goes to a PSX-Two dual output cricuit breaker. Ouput 1 goes to the power bus between all of the BDL168's. Output 2 goes to what I call the aux DCC bus. I use the color gray. In addition there is a 12VDC internal White bus and an internal Brown bus for the 16VAC needed for the BDL168's and other possible devices. In addition to the BDL168s, we also have DS64's and the Team Digital SIC 24 for the signal control. They get thier power from the White or Brown busses as needed. I use the SIC24 from Team Digital because it does not require a computer interface to resolve the signal patterns like the Digitrax product does. First variation, the BDL168s are the layout masters. They are setup to send Loconet messages upon detection to both the DS64 and the SIC24 and depending on that logic, turnouts are closed or thrown and the signals are adjusted by the SIC24. In this first case, there is virtually no traffic between this power district and the rest of the layout because all of the Loconet messages never go anywhere except to the local devices. The only connection is the commands that come via the track power supplied to the BDL168. The drawback here is that there is no track power connection to the DS64's so they can only respond to Loconet messages. Similarly, the SIC24 also has no track power and only responded to Loconet messages. So this is an auto-run solution and not an operations solution. Second variation, connect the gray AUX DCC bus to the DS64's only. This now allows the throttle to send commands via the track bus interface to control the turnouts within the power district. This allows the operator to select the path he or she wants to follow. Still the SIC24 and the BDL168 are only talking Loconet with the DS64. The Loconet subnet in this case is still isolated since there is no Loconet connection to any external Loconet segment. Variation three - no UP-5 connection from a Loconet port on a booster. In this case you can still have a local throttle bus and control the turnouts via Loconet. You still setup the throttle bus daisy chaining the throttle ports on UP-5 devices, and you connect to the isolated Loconet network to control the turnouts. All Loconet is still only within the power district. There are many other variations that can be derived. But all of these do restrict the amount of traffic on the overall layout. All of the local Loconet never leaves the power district. There is no need for a spec when the problem can be solved in an manner of implementation. It is knowing how the parts can be made to work with each other and looking at a different way of implementing them. While I did illustrate this using Digitrax, I am sure there are parallels using the other manufacturers. The point is take am moment and think about things and remove the tunnel vision for a while and there is always a different solution. And for those interested and semi local to Sacramento, this demostration Ntrak module set will be running at the Gats show in January and also in an Ntrak layout on the NMRA 2011 tour. There is no need for the CAN Bus, they should focus on fixing the broken parts of the standard.
To be fair to the manufacturers as I was a little unfair before, they are trapped as well. 15 years ago, ethernet existed, but it wasn't cheap, Loconet and expressnet may not have been standards, but they were better than nothing. Now though, how do you evolve to a standard like Ethernet without dealing with your proprietary standard. Motorola and others had the problem when they went from proprietary Cable Modem standards to DOCSIS. They had to bend over backwards to support their old standard concurrently and it ended up hurting them. So companies like Digitrax and NCE which are as pointed out, small companies are going to have a big problem. How do you manage the obselecence. I would say that the trick is to get the community involved. Not just the NMRA. As DCESharkman illustrates. The other question I have to ask is "Where is Kader in all this? Bachmann has a DCC system. It isn't any great shakes, but Kader may be the only company big enough to drive standards. If instead of an NMRA effort, they had an industry working group. With a rep from each relevent company, the NMRA and Kalmbach. Maybe reps from the big communities like this one. Right now it's JUST the NMRA as others have pointed out and they aren't equipped.
So what do we agree on? With the "Turnout" of mostly EE engineers here it appears to be another engineering discussion. What's the verdict? What have you guys and gals decided? Good bad, indifferent, could be better, they don't know what the heck they are talking about and/or you all can't agree. I suspect the latter.cute: I've decided to "Switch" off my DCC for now and worry about other problems on my layout. The cost is prohibitive. .
Hey Rick - I expected you to have pop in about the time that ethernet turnouts - er, sorry, switches - were under discussion Weeell, the thread title does kinda point that way I think. I'd say you're right, but on the other hand I could have predicted that before we even started
LOL and still LOL Yep, I thought so. Mike, you should write for Jay Leno, at least we'd have new jokes being told instead of the same old jokes covering different material. I have to say your responses are refreshing, straight forward and with a dash of humor. I know whose bread am I buttering and what am I up to? Nothing yet! Nicely put.
Not true, get us all in a room with white board and plenty of markers and we will come to a consensus, and the implementation strategy. It is here we can not illustrate the ideas efficiently that we have problems converging on a final solution.
Ding Ding Ding! We have a winner! That scene from "Apollo 13" where they figure out how to literally fit a square peg into a round hole was spot on! Internet websites just don't allow for that kind of mind-melding. I see it all the time at work where we try to work together with partners in different parts of the globe. Very, very frustrating.
OK, fine, get them in a conference room with a whiteboard (and no management) and we'd get somewhere....it would probably be over designed and impractical (need management to give a date when it will be released to curtail this) but it would be agreement. BUT, try to get those engineers to agree to a bar for drinks afterwards.... When I worked down in Carlsbad Ca. We had conference room K. aka, the Karl Strauss restaurant down the road. Completely off topic, but I've had pretty good success with Cisco Meetingplace...assuming you can get it to work.
Well it just so happens that I own my own company, and it also happens to have full manufacturing capability, and I have a pretty big white board............ So we can figure it out and prototype it at the same time. And we can have the support of the 200 engineering specialists I employ.
While I am new to DCC, here are my observations. 1. Manuals poorly written. 2. While I have gotten a LENZ system to work with other manufactures block controls and switch controls, why is it not plug and play? 3. When I researched DCC, I thought the layout would be easier to wire for DCC with reversing loops, block control, and I have not attempted signal control or detection yet. 4. For what they do the controllers are to high priced. For what they cost in parts is, well you get the picture. 5. Why is there no standard for Loconet or any of the others. It seems to you have to do work arounds everywhere. 6. Ethernet is not the way to go. To many ports needed which means to many routers/hubs. besides most people can not wire a plug on twisted pair up.
