NMRANet - Issue 4: Migration. Knowing your competition. The big "must have"

[The_Ghan]

Back in my AutoCAD programming days, the company I worked for was small and operated out of a residence.  There was five of us, for about five years.  We then spent a considerable sum with a marketing company.  Within three years we had six offices and ten dealers nationwide.  I set up three dealerships in New Zealand and a distributor in Singapore for S E Asia.  We rebranded products from our ill-concieved, dumb, techno-names to things that an Architect and Engineer would understand.  We dumped half our product range that catered for smaller, niche markets and concentrated on those that were more profitable.  We created a new product for drafting and scheduling steel reinforcement in concrete.

 

At the time, programmers, management and even the MD were concerned about, what seemed to us to be, far-fetched ideas propsed by the marketing company.  Needless to say, within two years Universities were screaming for our software to teach in class.  The 1990's were good to us, our "golden" years, so to speak.  I paid off my house on my own bonuses and I wasn't even in Sales.

 

Three points:

 

1. I don't believe that marketing companies are a waste of time, but you have to do your research and choose the right company;

 

2. Who am I to tell you how to run a business?  I'm a nobody.  Go for it hammer-and-tong ... it might just work!

 

3. Advice is like medicine; it's only good if it is taken.

 

Cheers

 

The_Ghan

[keitaro]

I will place feed back here for what i dreamt up some time ago for train controller and hope it gives you some helpful ideas.

I have no skill in this area so i really have no idea if its feasible or possible.

So the idea is to use nfc primarily.

 

Reasons small, can be used powered and unpowered, should fit nscale, can handle reasonable data throughput and should be able to but im no programmer to add any identofication to the nfc.

 

Ill start with the features envisioned.

Occupany detection, can be budget level or advanced cash cow detection.

The ability to controll lighting motor direction via software or hardware controller.

Any model could be upgraded to this sytem pending modelers skill lvl would depend on amount of controll they may have.

 

 

Ill start with track and layout side of it first.

The idea was to have wired nfc devices connecting to a controller that can then link to a hardware controller or wifi/bluetooth to a tablet or pc for software control.

 

For the budget conscious user the could use a basic setup.

This would work similar to transponding on dcc.

The user would wire a nfc device to the front of the block and the rear under the track of course. More could be added for station blocks for realistic slowing etc.

Benefits

Block detection

signaling could be updated via software command when nfc updates position.

Con

Possible delays in on train updates (lighting train change in speed or direction) as may be out of nfc range till hits end of block.

 

Other budgets,

 

Pending budgets blocks could be completely eliminated and you could have nfc set any distance.

E.g every 15 cm allowing for almost completely real time updates in software and immediate in train changes (lighting, completely realistic stopping) trains 5 seconds apart at a station :evilgrin:

 

You could even have a nfc under switches,perhaps pending budget still wire the switch to a controller as per current or nfc for command to change switch.

 

Now to the trains,

 

Trains could have different levels of support or features, such as budget line model direct from the manufacturer or full spec train.

 

Example kato sell e2 shinkansen with nfc control.

The model features nfc built in cab front and rear and motor car.

This allows for directional lighting control and motor communication via nfc. The train can also run purely from a dc controller as per norm and gets power from track as per norm.

The dc control would be best done via a switch isolating nfc connection to motor to the pick ups directly. Similar to lighting in current trains.

 

If using nfc it will have a built-in dcc like decoder to get commands from the nfc to tell motor the speed or change lights. This is why isolating for dc use is best to bypass the dcc chip detecting and running poorly like current dcc setup models used on dc.

 

For non installed or older models

Next is to buy the nfc decoder 2 types pending budget and use and modellers skill.

 

Budget or unskilled modeler can install a unwired nfc chip into the model.

 

This really requires no skill for most modelers basicly you would have a programing unit to setup a name or identifier for the model.

Then glue tape inside model.

 

Pros

Easy to do

Can be used for a hybrid dc stystem (supports occupancy detection) and user could setup blocks for an automation system in dc. Allowing the system to detect model is in which block and cut power or reverse polarity for automation. Can also use inertia like current dc controller to slow the voltage slowly and cut voltage at the end of the block if model is still slowing as all motors are different.

 

Cons

Low accuracy in stopping starting as system can only reduce or stop voltage.

No model control lighting etc through software

Some models are impossible to open.. Serious any longtime model has had a model they are afraid to open as its complex and afraid to break it.

Cant use on digital powered nfc system as it has no digital control of motor.

 

Type 2- there can be numerous types and sizes like dcc for different models.

 

This works just like dcc now needs to be wired in and requires some skills to do so.

 

Pros

Full control of model if wired correctly.

Interior lighting could also be achieved either being built into the lighting by manufacturer or via decoder install for interior lighting control.

Cab lighting supports reverse polarity for easy 2 wire install and nfc for occupancy.

Cons

Need to be skilled in install

Becomes expensive

Not all models may fit the decoder. E.g steamers.

 

Thats all for now took forever to type on my mobile.

 

 

 

 

 

 

 

PS  ps i agree with marketing is good if you have the right marketing company

 

Hope these ideas help you to make a great product or fill in gaps you had not thought of.

[CaptOblivious]

Keitaro,

 

I need to read more about NFC. You are right that this could be a very interesting application for it! I'm keen to learn more.

[cteno4]

Ghan,

 

I think this is a bit apples to oranges for a couple of reasons. first this is not riding a wave, but the start of a potential new one, and this requires much more of a vision approach than a marketing approach. second nmranet itself is not really a product, but a framework, other folks will be creating the products and markets that will build into an eventual nmranet product.

 

I surfed similar waves in the past. a big one was the cdrom/multimedia waves of the 90s. there was so much more marketing in the wave than vision. i was working with a small company that surfed the wave from a small company of a dozen to going public in less than two years. at the height of it we were a darling, being courted by all the biggies to be bought to become their multimedia division. we had publishing partnerships with apple, but the marketing from all angles forced them to ditch what they did the best and totally stripped away much long term vision for the company. company went public and proceeded to crash and burn w/in a short time. things were very successful on paper and on the surface, but did not last. Its the wave of marketing in the industry that tends to give these big booms to small companies in these situations more than what they may be doing or even how well they are doing it. i was working in the sf bay area during the web boom and got to witness this big time all over to many many friends. After just having gone through it i sat it out. yes i missed in on cashing in some for a few years (i had opportunities) but i saw very, very few of them was probably going to last and i just could not get behind working on some really dumb things for dumb reasons just to cash in.

 

NMRAnet is not surfing a wave like this but is the foundation for a new wave to start, so it needs to do what is good and cool for a vision rather than be told by a market survey what to build or how to package it. the packaging will come later as companies come in around and do products. they will both target their niches and then properly market to them. this will then bring along the whole nmranet concept with it and that can be slowly packaged/presented in palatable forms for the market in specific and general and can be lead by the folks that really pick up supporting the technologies. yes some marketing does need to be done to companies to pick up nmranet and build for it, but the kinds of companies in the beginning will be like captain's and get it pretty quickly and see where they can capitalize. they know their niches well enough to come up with good product ideas.

 

building the nmranet market from other small companies' products that will first start doing little niche things first will allow nmranet to have its over all market form by a nice evolution of what and how the market reacts to better than trying to come up with a grand marketing plan and use that to design the framework. I have always found this approach, especially in this situation, to be better to allow for more flexibility. the greater the scope of the plan the more issues you will run into unless it can grow more organically.

 

one last thing is i dont look at a lot of what you are calling competition to nmranet as actually competition. much of it can play well with nmranet, leverage from it, and let them play well with others (its one of the primary focuses of it) to float many boats. yes some of the individual companies with proprietary systems that have some of the aspects of nmranet may push back and look at it as competition, but i think the marketplace will may force them to play with others thru this or other companies or individuals creating the products to allow them to play with others unless these companies try to totally block this (and looks stupid in the process). yes the nmranet folks did have to know this competition and i think they did their homework from what i have seen to find the needs and frustrations as well as look at forward looking technologies and practicalities as well not being tied to try to work w/in an existing marketing plan/product line. DCC is not competition for nmranet, but a great partner for it. nmranet brings in loads of flexibility for dcc and takes over doing some of the stuff dcc is not good at doing and thus lets it focus on what it does well and let it leverage off all the other things that it can then play with well.

 

cheers

 

jeff

[The_Ghan]

Hey cteno4,

 

Great analogy ... you are of course correct, Autodesk's products were a great wave to surf for two or three decades.  But the market is much smarter these days and the fragmented competition has gone, orphaning a lot of AutoCAD users on legacy systems.  Luckily, with the passion of architecture beckoning, I left the industry at market saturation (pardon the pun).

 

But to have a wave, or create a wave, there needs to be water ............................ 

 

OpenLCB is competition.  I just can't agree with you.  When you buy Pepsi instead of Coke ... that's competition.  When you buy CTI TrainBrain instead of Digitrax, that's competition.  When you buy a NCE decoder instead of a TCS decoder, that's competition. When you buy OpenLCB instead of XpressNet, that's competition.  On the other hand, as I've said before, if OpenLCB will let me connect my LocoNet to a Lenz turnout controller, and run CTI signalling all on the OpenLCB Bus as an out-of-the-box solution then that is complementing the marketplace.

 

Cheers

 

The_Ghan

[cteno4]

Ghan,

 

Yes you're right there is always some level of comperition going on, but I really so see nmranet being much more the complementing thing going on that will get more folks into doing more automation by making it easier, more flexible, more powerful and leverage both existing products as well as the new emerging tech engineering wave that has started and I fear that small traditional train developers might not be able to catch and ride on their own the way the youngans might. So i really do think its something that can float a lot of boats together. I like that sort of approach myself.

 

Jeff

[The_Ghan]

Jeff,

 

If OpenLCB is more feature rich than XpressNet then that is great.  That will be a big draw card.  But it remains competition.  A basic cell phone and a smart phone are miles apart in capability but the basic function remains the same.  I don't know anyone who keeps a smart phone to store their contacts but dials the number on a $40 cell phone.  The two are competition ... 

 

I think we've established that, and Don has agreed.  OpenLCB can be a whole lot more and I look forward to seeing that, as it comes into fruition.  In the meantime, I'm collecting my thoughts with regard to other issues to be posted soon.

 

Cheers

 

The_Ghan

[mwsteele]

I think there would be a problem if model railroad manufacturers only focused on the profitable areas. Since there is such diversity in model railroading many smaller aspects would suffer. For example many people in Japan do not use  DCC at all. But Kato still makes some models DCC friendly with drop in decoders and makes a couple decoders for them. I'm sure that there is not much profit on doing this if there is such a small portion that actually uses it.

So in my opinion I'm happy studies aren't being done to identify PROFITABLE areas. I'd hate to see things dropped.

 

With modern WiFi systems at the 5ghz range would this provide for much less packet loss? Also many products are amplified providing longer range.

 

Also, outside of the residential systems there are many others which offer way better service.

 

I really like the mikrotik routerboard products:

 

http://routerboard.com/

 

Now I don't have any understanding of NMRAnet, I'll have to do some research but I'm sure something like this could be used and have very little packet loss.