fischertechnik community forum

Hallo,

da in der offiziellen Produktbeschreibung folgendes steht:
"2 Erweiterungsanschlüsse: I2C und RS 485 zur Kopplung weiterer ROBO TX Controller"
Ist es schon (fast) ein Produktmangel wenn es nicht an dem ist...

Gruß

The TX-C will be a normal I2C master, it has no I2C multi master functionality.
A I2C bus can only have one I2C master on the bus.
So direct I2C connection between 2 TX-C will not be possible for the moment.

@vleeuwen: DANKE !

ft

Dirk Haizmann ft hat geschrieben:Nach aktuellem Stand sollte das Thema
noch im Frühjahr geklärt sein.

Gibt es inzwischen Fortschritte. Metrologisch ist seit gestern Sommer.

Hallo,
ich bin zwar kein ft-Mitarbeiter, schreibe hier jetzt aber trotzdem mal was dazu.
Die I2C-Funktion befindet sich gerade im Betatest, d.h. die TX-Betatester können es bereits ausprobieren usw.
Also: Es wird dran gearbeitet, vielleicht ist es nächste Woche fertig, vielleicht auch erst in 3 Monaten. Aber: es passiert was.
Ich bin mir sicher, dass Drängeln nicht hilft. Vergessen wird das in der ft-Entwicklung so schnell keiner.
Grüße,
Martin

@Masked

I know development takes time, I worked there long enough to know that. As we used to say: you can't make a plant grow faster by pulling at its leafs. So I never ask when a feature will be implemented or a bug be resolved. I do think however that ft could be a lot more open about the problems and involve the community a lot more. Sometimes we get a reply from Janosch that gives some insight about what is happening, the answers we get from Dirk however are useless and almost offensive. I sometimes think that Dirk really does not like his job. The phrase 'shaun mer mal' has become a complete joke. It sounds like a lawyer saying that things will be given 'due' consideration and everyone knows he means 'none'.
For example, in the old forum ther once was the discussion whether the TX software should be open source, ft gave the impression that this was genuinely being considered. Of course nothing happened, after a reminder there was the answer that nothing had been decided yet. After a while the old forum was removed and with it all evidence of that discussion. When asked to reopen the forum they went straight into the same loop: we will look into it, nothing has been decided yet, we will let you know when we know etc.

I truly do not understand the benefit of this kind of behaviour. Explaining that would already be a step in the right direction.

Ad

Hi All,
Because of an internal discussion in the beta tester panel I am as beta tester very curious about the expectations that the TX-C users have about the functionality of TX-C I2C connectivity.
This will help me to perform the beta testing.

-) master or multi master?

-) limitation of the I2Cread and I2Cwrite operations to only 1 or 2 data bytes or more?
More: means a byte array like LEGO is offering.
The more complex I2C peripharals will need more than 2 data bytes.

-) To add data preprocessing functionality below RoboPro level?
--> Above TA level (TA=transver area)
--> Below TA level (between RoboPro and the TA)

-) ?????

If your have the intention to make use of the TX-C I2C bus:
-) Do you have already knowledge of the I2C bus?
-) Will you develop I2C based sensore/actuator hardware yourself or will you make use of 3th party products?
-) Do you have equipment like an oscillioscope or I2C potocol analyser?
-) Do you expect to experment with I2C without soldering?

Hi,

What to expect from I2C?

First, since the TX is a fischertechnik product (a toy if you want), I expect ft to develop sensors/actuators for it. The computing series mainly focusses on robotics so all the typical robot sensors and actuators are obvious candidates. This would bring advanced sensors like e.g. gyroscopes within reach of students and hobbyists who are not necessarily comfortable with electronics. Here good support in RoboPro is paramount.
It could be that for some sensors/actuators there is limited interest, e.g. a GPS or a barcode scanner, In that case homebrew or third party components could be used. Here RoboPro cannot know about the sensors, hence can only provide primitives. Ideally the primitives should be flexible enough to allow any kind of sensor/actuator. When the primitives are too limited many applications will simply not be possible or require a complex interface like an additional microcontroller. From a system point of view the rich set of primitives is much more attractive because it is virtually for free (software only) whereas the interface solution cost money for every unit sold.
I understand that RoboPro is a toy language and that the features are necessarily limited. For the C language this is however not the case. I realise that C is not the language of choice for the majority of the TX users but the users who are able to develop their own sensors are very likely able to program the use of the sensors in C. Therefore I think that at least in C, a rich API should be made available. There are many existing I2C API's, just pick a reasonably good one.

The questions:

-) To add data preprocessing functionality below RoboPro level?
--> Above TA level (TA=transver area)
--> Below TA level (between RoboPro and the TA)
Can't fathom the impact or couldn't care less

If your have the intention to make use of the TX-C I2C bus:
-) Do you have already knowledge of the I2C bus?
Yes
-) Will you develop I2C based sensore/actuator hardware yourself or will you make use of 3th party products?
Most likely both
-) Do you have equipment like an oscillioscope or I2C potocol analyser?
Both
-) Do you expect to experment with I2C without soldering?
Not really, do not fully understand the question.

Hallo,

as the different ß-testers allready have written we are working on this project.

When we have further details we will inform you immediately.

ft

Hi Dirk,
With my more general questions to the big group of TX-C users I hope to get an idea what the end-users are expecting from the I2C bus and how many end-users are interested in the I2C connectivity.

@Carel
When you ask five users you will get six answers.

Hi Dirk,
No problem, I am used to work with interviews of students.
10 students 20 different opinions!!
However only 5 or 6 replies will also show that the I2C bus project (at RoboPro level) is not so worthwhile.

@vleeuwen
Carel, I don't agree with this last "conclusion" of you at all.
I think most people waiting for I2C are just doing that: waiting.

Paul

.. dann warten wir's mal ab.

ft

Hi Paul,
The question is, waiting for what?
(At FtMscLib level)
-) normal I2C bus functionality like LEGO (I2C read/write with a byte array (byte list))?
-) subset, for example: only 1 or 2 bytes of data?
-) preselected list of I2C device?
-) only I2C device with "no soldering iron" connection?
-) ?????
-) or like you are suggesting: no expectation, waiting for the end result?

-) RoboPro I2C operations beside the FtMscLib operations?

This feedback will help me as beta-tester to set up and to perform my function tests of the TX-C I2C connectivity.

Hello together,
if you would like to go on with your discussion, please open a new thread in the "RoboPro"-Section.
This thread should be reserved for the "When"-discussion, not "How".

Thanks,
Martin

Hallo,
weiterführende inhaltliche Diskussion bitte in einem neuen Thread im RoboPro-Unterforum.
Hier sollte eigentlich nur das "Wann" diskutiert, nicht das "Wie".

Danke,
Martin

@masked
Das find ich eine gute Idee !

@Martin, @Dirk,
I agree with Martin. For me, the original question was also:
-) When the normal I2C connectivity at TA/API level, like for example LEGO has, will be available ?
This as described a longtime ago.

But based on the current state, I think that this question needs to be modified and to be extended into:
-) When a limited I2C connectivity will be available?
This question is raising a new question:
What will be these limitations?

Based on the current functional description It looks like that a basic I2C based stepper controller like
AMIS30622 and AMIS30624 is not usable and this is worried me.

The original formulation of the question in this thread leaves enough space for this discussion.
No reason to start a new thread.

See also

Hallo,

die diversen Ideen nehmen wir gerne mit auf.

Aktuelle ist das Thema I2C -wie bereits oben erwähnt- in der Betatestphase.

Ich denke sobald wir hier weiter sind können wir uns um die weiteren Themen kümmern.

ft