Alt. firmware and GUI for Thanos's AMC1.5 & AMC1.6

[prilad]

Hi, Speedy!

Till now ... if I asked you about how big is this FW and the full option program ... proudly say ... like the man in this video
>>>>> It's Russian ... <<<<<

No, no, nooo . My FW is not SO BIG . Yes, it has more options than FW by Thanos, but you can REDUCE that by setting some parameters to Zero...

To continue with the FW troubleshooting edition ... I think it should be an embedded function into the latest working FW ...

Do you think that it really necessary?

Regards,
Ale

[speedy]

Hi Prilad :

To continue with the FW troubleshooting edition ... I think it should be an embedded function into the latest working FW ...

Do you think that it really necessary?
Regards,
Ale

Google перевод ... Да, мой друг, это действительно необходимо ... это хороший вариант, он понадобится нам позже ... как знает? ... и должны быть улучшены, чтобы больше вариантов и кристаллов ... если вам нравится, что ... С наилучшими пожеланиями от меня

скорый

[speedy]

And my firmware - it is the first touch, and almost virtual one. Because I have no any hardware for fast testing and debuging firmware in real life... And I works with x-sim firmware less than a year

But I think that in next generation I will try to make duty cycle shorter than 1 ms. No problem...

I didn't see this till now ... @ Prilad Please don't be Modest in your business ... You don't have any hardware to do tests ? ... you are really genius ... How is that ? ... I'm fair good in electronics but without any power source to test ..!! I don't think so ...
And if you have a good idea, do it at once ... because tomorrow it will be history ... and someone else will do it before you ...

BTW ... did any one noticed that where ever there is a Nuclear Accident happens ... really sorry ... Geniuses comes out ... like Japan or Russia ... If [ yes ] then hit me first please.

best regards and "KMF" (Keep Moving Forward)

spe edy

[prilad]

Да, мой друг, это действительно необходимо ... это хороший вариант, он понадобится нам позже ... как знает? ... и должны быть улучшены, чтобы больше вариантов и кристаллов ... если вам нравится, что ... С наилучшими пожеланиями от меня

Thank you, Speedy... I am touched...

You don't have any hardware to do tests ?

Yes, my friend. When I started to make the firmware, I really haven't any hardware. But I have a computer. And I'm good electronic engineer too. And I have a friends in Russia (and not only in Russia) who help me to debug my firmware before I published her.

Now I have my own board for x-sim (AMC) based on atmega64. Without LCD. But with FT232R. And I plan to create 3dof or 4dof simulator on this ONE board, because atmega64 have 6 16-bit timers (and can create 6 10-bit PWM channels). However, I still can not debug firmware is for atmega8535 and atmega644. Only virtually.

Regards,
ALE

[speedy]

.

And I plan to create 3dof or 4dof simulator on this ONE board, because atmega64 have 6 16-bit timers (and can create 6 10-bit PWM channels).

See that guys 4DOF on one board ... Go get them Tiger ...

spe edy

[prilad]

See that guys 4DOF on one board

This is not unique
viewtopic.php?f=37&t=319

Regards

[tronicgr]

Yes, but with how much delay?

Just the way they transmit the values and the size of the data is making it impossible to keep it at a decent loop frequency.

Just for comparison:

Binary data transmission speed of 2 indentifier bytes and 4 data bytes (2x 16-bit axis): 500us
"AB~a01~~a02~" translates to 6 bytes as "B B BB BB"

http://motionsim.blogspot.com/2012/05/this-capture-i-show-timing-of-capturing.html

Now if they do HEX transmission each hex byte is 2 binary bytes!!!

$<value><axis><value><axis><value><axis>#
~36~~a01~Y~a02~X~a03~Z~35~ which translates into bytes as "BB BBBB BB BBBB BB BBBB BB BB" !!!!!!

How many bytes are there now? 22???? So they have to spend at least 1.85ms in transmission time!!! Don't you also need more time for execution of other code too? So how much should be the total loop delay? 3ms? 4ms perhaps to prevent buffer being full all the time (given they even use buffered serial connection)?? And the PID runs with 10ms intervals...

Sorry for my rant, but I just can't sit back watching all this arduino based controllers do something they are clearly not designed for. They are awesome for robotics, I agree, but I don't think there are too many robots around that need almost realtime control from PC, and most of them use simple RC servos if they are mobile.

Thanos

[vicpopo]

Hi guys,
For the first time I achieved running Prilad FW and SW with Tronicgr amc1.6 with 18,432 crystal.
I hadn't a motor connected yet but Prilad it works.
Happy for you !
A littlle video confirming that.
Best Regards
http://m.youtube.com/#/watch?v=jAWguF-L2Mw

[prilad]

Hi, Guys!!!

Sorry for my rant, but I just can't sit back watching all this arduino based controllers do something they are clearly not designed for. They are awesome for robotics, I agree, but I don't think there are too many robots around that need almost realtime control from PC, and most of them use simple RC servos if they are mobile

@Thanos. I only say that there are people who try to make 4dof on one board. Not more...

And I'll try to make it on my AMC (on atmega64). And I think - it possible. even with main operating cycle = 1 ms. Time-critical code I will write on assembler, not on C... For 4 axis control message will have length 10 bytes max (if to use 16-bit axes: two bytes - header, and 4*2 bytes for data) and 6 bytes min (if to use 8-bit axes) And with baud rate = 115200 it will be have 0.87 ms (max) or 0.52 ms (min) transmission time!

For the first time I achieved running Prilad FW and SW with Tronicgr amc1.6 with 18,432 crystal.
I hadn't a motor connected yet but Prilad it works.

@vicpopo. Congratulations. The good news

Regards,
Ale

[vicpopo]

Hi Prilad,

Look at this video on Speedy topic :
http://www.x-sim.de/forum/viewtopic.php?f=43&t=246&start=80#p3032

Many Thanks

I don't know yet how use 100% your FW and SW but it's a good start.
Best Regards