michaelkellett ,

just for a simple comparison, I followed the suggestion of Dimitry and checked the components list in their site and related prices...

41 - 8 bit port : 14.00$

Trying to think like a newbie, I can expect 8 I/O digital pins. Instead this is I2C component.

39 - Large [color] LED : 3.00$

In the pseudo-schematics it is just a LED + resistor.

38 - Pushbutton : 2.40$

A pushbutton and resistor. As this pushbutton has its own form-factor, I suppose puts some limitations in the point where it should be wired to be reachable.

37 - RF Connector : 10.00$

... but not sure what do they mean exactly. The "show details" option just show a nice image.

25 - High power 5V supply : 29.00$

Sounds a bit expansive, and it is unregulated power supply. Not exactly what we are seeing in the micro controller boards / projects strongly oriented to the low-power and ultra-low-power features.

23 - Isolated POE 5V supply : 39.00$

Sounds the VIP alternative to #25

00 - Four directly I/O lines of the micro controller : 2.00$

Just a box with wires and LED inside. Need at least three of them for a normal project.

To reach the bare minimum components that we expect to find in a micro controller board we should invest about 50.00$ (!!!) Then, where is the micro chip???

You should buy the Project PCB: the cheapest one cost 55.00$.

Sorry to be co critical but I at least expected that with a so well-done modular system there was to option to setup the maiboard with blocks in different configurations and use the same to make an incredible number or projects ! Instead this is a bare PCB board with no option to make any kind of test. After you have bought your components and your Dream Board you should just solder them, hopefully in the right way because then the board is on, running (or not) as expected.

About this last detail, irony apart, IMHO there is a big big incongruence. What is the sense to create an entire system shown as - citing from their site:

Tibbo Project System (TPS) is a highly configurable and affordable automation platform. Its flexibility is based on Tibbits – miniature blocks that implement specific I/O features. Need a certain I/O function? Install the right Tibbit. Have no use for something? There is no need to have it in your device. This module-based approach saves you money by allowing you to precisely define the features you want in your automation controller. With TPS, we have completely rethought not just the hardware, but also the configuration and ordering process. Read on and see for yourself!

When you can just do this once? And what is the sense to make a so aesthetic coloured block set if the final point is to solder them (hopefully in the right position) then there is no more option but using the board? It seems to me something really away from the concept of experimenting.

Probably the - almost - right application environment maybe the PLC world and industrial applications. But in this case the almost total absence of technical information and robustness specifications and serious environmental test benchmarks plays against the product.

Last but not least, there is also on the site an interesting series of pre-selected projects. An appoint on this: maybe I am too old and I lost the last most recent updates in the world of the development  languages but the choice of BASIC as the easy alternative to the C (No C++?) language sound a bit outdated to me.

Enrico

Enrico,

thank you for taking your time to write this detailed comment. There is one part I really do not understand -- the part about the soldering and using just once. Tibbits PLUG into our mainboards. There is no soldering involved. There are rear-entry connectors on the back of our boards. The entire system is, essentially, soldering-free.

Yes, our devices primarily target industrial control and data collection applications. For example, in one of the systems installed by our customers, TPS devices count the number of cars passing through drive-through lanes of fast food restaurants. In another application, the TPS is used to monitor the operating parameters of a cell tower.

About some specific Tibbits:

- Yes, port extender is an I2C component. Each Tibbit communicates with the main CPU through 4 I/O lines. This Tibbit basically expands 4 lines into 8.

- Large LED. These are specifically for indicating different device conditions in the industrial applications. You don't always need to use such LEDs. You get this Tibbit only when you want to be able to see the status of your device from afar. For normal status indication you have 7 LEDs right on the mainboard.

- The pushbutton doesn't have its own form factor. It is just another Tibbit. Again, these are for industrial applications. Something to push without having to closely look at the device. For "normal" things we have a button right on the mainboard. On the TPS2L device there is also a sensor keypad with 4 buttons.

-  The RF connector is just that -- a coaxial connector, with a wire and a small connector that can plug into our Wi-Fi module.

- High-power power supply Tibbit. IT IS REGULATED. Gives you about 3 amps of current. Very necessary sometimes. Imagine that you filled your TPS with high-power relays. When actuated simultaneously they will draw quite some current! And if you added GPRS, this would increase your current requirements by a lot as well.

Four direct I/O lines. Yes, just the box with wires and status LEDs. Necessary to get the wires to the surface. We are releasing a set of proto PCBs that you will be install instead of Module Tibbits. This will give you the ability to put your own circuit into the box, and still use Tibbit connectors.

Regarding the choice of programming languages. Call me old-fashioned, but I do think that C++ has no place in the world of real-time control. C++ is an excellent choice for application development, but hardcore control stuff should not be done in C++. As for Tibbo BASIC, well, it is really well-suited to the kind of applications that people develop on TPS. I often experience a kind of language arrogance -- what??? BASIC??? I am WAYYYY ABOVE this!!! My answer? It is just a tool, one of many tools in your toolbox. Sometimes it is suitable, and sometimes not... and this is why we have C. C++? I really don't think so. There are better choices to expand our list, I think.

Regarding the documentation. I am not sure why folks are saying that we have no documentation at all. Take a look at this:

Tibbo Document System

Also, we have the software and programming side of things documented:

TIDE, TiOS, Tibbo BASIC and Tibbo C Manual

Now, regarding the publishing of schematic diagrams (as suggested by many comments on this page), sure, we are going to do this.

I hope I was able to address at least some of your questions. Feel free to drop me a line here, or email me at dima@tibbo.com. TPS is my baby. I never get tired of talking about it.

Cheers,

Dmitry

Hello Dimitry,

I am happy that you have take in account my considerations as positive critics instead of something aiming to "destroy" your product. As a matter of fact what I have tried to do was a few words sort of virtual road test on your products. IMHO on top detail that is missed in your post and almost everywhere is the customers target, that as you explain in your previous comment is more industrial than maker and hobbyist.

Now let me answer in detail citing your message.

thank you for taking your time to write this detailed comment. There is one part I really do not understand -- the part about the soldering and using just once. Tibbits PLUG into our mainboards. There is no soldering involved. There are rear-entry connectors on the back of our boards. The entire system is, essentially, soldering-free.

This is my mistake. Frankly this is not very clear in the product specifications and in my opinion, it is one of the first things that should be underlined. Consider that users read few and rarely go in depth if not really very interested.

Yes, our devices primarily target industrial control and data collection applications. For example, in one of the systems installed by our customers, TPS devices count the number of cars passing through drive-through lanes of fast food restaurants. In another application, the TPS is used to monitor the operating parameters of a cell tower.

This explain many things. In the industrial environment, where I have worked just with embedded components and scalable circuits your pricing is more affordable. Remain the point that are not so clear the industrial range specifications and application field (e.g. vibration, dust, humidity etc.)

- Large LED. These are specifically for indicating different device conditions in the industrial applications. You don't always need to use such LEDs. You get this Tibbit only when you want to be able to see the status of your device from afar. For normal status indication you have 7 LEDs right on the mainboard.

This is another point that has sense just in the case of the "industrial view" of your products, that seems to be the right one. Same for the other critics/considerations to the other components.

- High-power power supply Tibbit. IT IS REGULATED. Gives you about 3 amps of current. Very necessary sometimes. Imagine that you filled your TPS with high-power relays. When actuated simultaneously they will draw quite some current! And if you added GPRS, this would increase your current requirements by a lot as well.

Again, in an industrial environment application this has definitely more sense.

Regarding the choice of programming languages. Call me old-fashioned, but I do think that C++ has no place in the world of real-time control. C++ is an excellent choice for application development, but hardcore control stuff should not be done in C++. As for Tibbo BASIC, well, it is really well-suited to the kind of applications that people develop on TPS. I often experience a kind of language arrogance -- what??? BASIC??? I am WAYYYY ABOVE this!!! My answer? It is just a tool, one of many tools in your toolbox. Sometimes it is suitable, and sometimes not... and this is why we have C. C++? I really don't think so. There are better choices to expand our list, I think.

Should place some order. I am an "old" developer and I think I saw a lot of basic languages versions, including compiled ones. It is not a question of "vintage" language at all but it is a question of the target. Rethinking your entire product approach (that IMHO remain to be completed here and there but it is normal as it is just born) the industrial applications target rarely works with the new languages (maybe Python the last generations of hardware engineers) and in this case BASIC will fit a large audience. Nothing to say about C vs C++ in the embedded world I almost agree with this assumption.

BTW, interesting instead the Linux based more advanced boards that are in your catalogue.

Despite the schematics and/or sample circuits are not yet been published, the idea to publish them is something really interesting, especially for the industrial world.

Enrico

Hello again Enrico,

it is great that we are talking. My mode doesn't get ruined because someone has offered an honest opinion. The path greatness lies through a lot of critical comments. Yours were very constructive. I am happy that you took time to write them.

Yes, we have released this new Linux board. Cooking Node.js support for it at the moment. I suppose this is very modern and some users out there will find this appealing.

Keep in touch and thank you for your time!

D.

Looking forward for updates on the already missing or partial parts

Enrico

Looking forward for updates on the already missing or partial parts

Enrico