Go Down

Topic: Will the Due be a dud? (Read 10642 times) previous topic - next topic

retrolefty

As well as slightly higher output voltage for semiconductor Vs tube rectifier, there is the issue of natural 'soft starting' that vacuum tube rectifiers exhibited. As it takes some amount of time for the filaments to reach operating temperature and thus allow full current flow, there was a more gradual rise in output voltage and current flow with tubes, much easier on the downstream filter caps and other components. Semiconductors rectifiers allows much higher initial current surge limited only by the impedance of the transformers windings.


Lefty

smeezekitty


As well as slightly higher output voltage for semiconductor Vs tube rectifier, there is the issue of natural 'soft starting' that vacuum tube rectifiers exhibited. As it takes some amount of time for the filaments to reach operating temperature and thus allow full current flow, there was a more gradual rise in output voltage and current flow with tubes, much easier on the downstream filter caps and other components. Semiconductors rectifiers allows much higher initial current surge limited only by the impedance of the transformers windings.


Lefty

This is true. But modern type filter caps are designed for this type of current surge. It should not affect upstream components too much either since the other tubes will take around 10s to start conducting.
(i see the topic got derailed  :) )
Also, some rectifier tubes are known to go short circuit and take out the expensive xfmr.
Avoid throwing electronics out as you or someone else might need them for parts or use.
Solid state rectifiers are the only REAL rectifiers.
Resistors for LEDS!

Graynomad

Quote
i see the topic got derailed

Well you never know, maybe the Due design will include some valves :)

Quote
IOREF. A Due board will have 3.3vdc on that same pin.

Good point, I guess that answers that question. Pity about the 300+ existing shields but it would be hard to accommodate them and keep the price down I suppose.

______
Rob
Rob Gray aka the GRAYnomad www.robgray.com

retrolefty


Quote
i see the topic got derailed

Well you never know, maybe the Due design will include some valves :)

Quote
IOREF. A Due board will have 3.3vdc on that same pin.

Good point, I guess that answers that question. Pity about the 300+ existing shields but it would be hard to accommodate them and keep the price down I suppose.

______
Rob


I agree it will (is) a pity, mostly because voltage conversion could be performed on the board so that the shield connector pins stay 5vdc compatible, if it was made a design decision.

However if indeed the Due ends up not being 5vdc shield compatible, that will only make a opportunity for some industrious person to supply the world (they will beat a path to his/her door) with a shield adapter board, I would call it the shield Rosetta board. The SRB would just have I/O voltage chips and otherwise just transfer/translate all Due signals so that existing shields will work with the Due, at least electrically. I would well it for around $12 and never have to work again. Oh wait, I'm already happily retired, I guess someone else needs to run with this ball. I release all rights to this specific copyrighted idea.
 

Benji

Well.. if the shields are not going to be compatible, maybe it's time to raise the 'pin spacing' issue again?

mmcp42


Well.. if the shields are not going to be compatible, maybe it's time to raise the 'pin spacing' issue again?

Good idea, that would stop accidentally plugging in an incompatible board
there are only 10 types of people
them that understands binary
and them that doesn't

Grumpy_Mike

Quote
I release all rights to this specific copyrighted idea.

Well you don't have much choice, you can't copyright an idea.  :)

The point is that if something doesn't happen soon increasing numbers of people will be distracted with other boards. As well as the Raspberry Pi, which is more of a desktop machine, the STM32F4 discovery board is much more powerful than the proposed Due. And at less than £10 each in the UK it is worth getting a few to build into projects. The things against that is the tool chain is so difficult to get together especially on a Mac.

retrolefty

Quote
Well you don't have much choice, you can't copyright an idea.  


I may be wrong (IANAL), but I do believe one can copyright an idea, just by writing it down, It's probably a patent you are thinking about?

Yes, the Due is way over due (oh a new Arduino joke!). Reminds me of all the vaporware that use to happen in the early microcomputer days. Announce a new product being avalible (when it really is not) use the advance orders to fund the actual building of the product and hope to hell there are not too many delays before bad publicity burns your effort.


Lefty

Grumpy_Mike

Quote
but I do believe one can copyright an idea, just by writing it down

No that just copyrights the writing, not the idea it contained.

See this:-
http://www.lawdit.co.uk/reading_room/room/view_article.asp?name=../articles/8123-JC-Can-Copyright-Protect-My-Idea.htm
Quoting from it:-
Quote
Can Copyright Protect my Idea? .......
The simple answer is no, an idea in itself cannot be protected. The Chambers 21st Century Dictionary defines an idea "a thought, image, notion or concept formed by the mind" so by its very definition an idea is a totally intangible thing. In order for there to be any protection of an idea, it must first make the idea exist in some tangible way. In the case of copyright law, it is the work that realises the idea that is protected (i.e. a document), and it is the act of recording that work that fixes copyright in the item itself.



Quote
the Due is way over due (oh a new Arduino joke!)

Like  :)

Yes it does sound like that, although they said they had prototypes at the September Maker Fair although I didn't see it. I offered to help with the development and they said I could, wrote my name down and everything. Expressed surprise I had ARM experience, but I had one of the first ARM computers in 1987. And used that and it's successor the RISC PC until 2002.
http://en.wikipedia.org/wiki/Acorn_Archimedes
Needless to say I heard nothing.

It could be many things, chip shortages or difficulties with the cross compiler / environment are my guess.

Benji

My guess would be the same, difficulty getting the IDE working with the different architecture types.

What does 'bother' me (not really, but I can't think of a better word) is that they are not releasing ANY information.
There is no shame in problems, most people here are developpers themselves and know you can run into problems when using totally different hardware than you are used to.
If they have issues, discussing it with the community might offer the solution, since there are a lot of people here that know the IDE inside and out, know the hardware, memorized the datasheet of the new cpu already (ok maybe that's exegurated, but you know...) so it might offer that one solution they are looking for.


retrolefty

Quote
It could be many things, chip shortages or difficulties with the cross compiler / environment are my guess.


I suspect it's that second thing you mentioned. I suspect trying to integrate the 32 bit tool chain into the existing IDE that also has to continue supporting the 8 bits chips is driving them buggy. It would probably be less effort if they were just to develop a totally independent and separate IDE/toolchain to support the Due.


Lefty

GoForSmoke

It is a big step from UNO to Due and solid new software -always- takes many times longer than anyone including the programmers wants to believe. Buggy software only takes 10%-25% as long, as long as it or something very similar has been done before.

Just due to the step up, I think give them lots more time. I also think that the smart route on the IDE is that it should not be a one size fits all package.
Nick Gammon on multitasking Arduinos:
1) http://gammon.com.au/blink
2) http://gammon.com.au/serial
3) http://gammon.com.au/interrupts

Grumpy_Mike


Graynomad

You can tell a few things from that pic, assuming there are no components underneath (I doubt they would do that) then hardware wise it's really just a Mega with a new processor.

But the software opportunities should be extensive.

Re the IDE, I seem to recall that debugging was going to be included (and you can see what looks to be a JTAG header on the board), I doubt that would be retro fitted to the existing IDE and frankly programs written for this will need a much better IDE. I'm thinking they will go the same route as Atmel's AS5.


______
Rob
Rob Gray aka the GRAYnomad www.robgray.com

winner10920

what's that thing next to the word erase? Its right where it says jtagse1
and from the looks of it, id say it'd use the same ide, why hold the same layout if you need different software? And they may have 5v out since also why keep the shape fitted to an mega if its not compatible with mega shields?

Go Up