Will the Due be a dud?

I think a part of what a faster processor can do will be done with an android phone. You can't leave your phone with a project but how many projects actually operate with the designer away? If you want a handheld device then you can just purchase a tablet for 200USD and hook up with an arduino. You need a lot of money on display and storage ect anyway for a handheld and will never get it good-looking enough like a tablet.

This is a bugbear of mine.

Anything in particular?


Rob

Anything in particular?

I make most of the pcb's I use myself and I find it a lot more convenient to have a processor board that plugs into my motherboard/shield whatever you want to call it. As I mentioned above, I have had a few pro minis fail on the other side of the world and it has been easy for me to send a new one with instructions to someone to just pull the old module & insert the new one. Not quite so easy with a board buried underneath a heap of wires & other stuff.

Will it be a dud? I don't know. The reason I think the replies to threads are not that high is because there is nothing to hold and look at yet. Right now the UNO and the Mega are what people use, so that is what all talk is on. As soon as the DUE is launched and people see tinkering going on with it, interest will come.

There has been an announcement and that is about it. How many threads have there been in the last months about ARM compatibility, people wondering if there is going to be a bigger/faster/high memory version of an Arduino? Now there is one, people will come for it and play with it.

I have been working with uC's for about a year now, I don't see a direct application for it, except that it would be convenient for my robot (which is work in progress) to have more outputs, more speed and especially higher capacity for the programming, if I ever want it to do all the things I have in my head. For LED projects it is not needed at all, the 328P works great for it, in many cases is overpowered. When people get deeper into tinkering, want to do real time audio or video processing, they will soon start complaining the DUE is too slow and we need an Arduino UBERMEGAPRO 2ghz ARM with a bazillion gigabytes of ram and quadrillion terabytes of space. Wait and see it happen :)

Hi all,

Any news on when the Due is likely to be out...and the bootloader for the arm. I assume it's going to come out as a new update to the IDE when the Due gets released?

I cant wait to get one! I think it's going to be better than the Raspberry Pi for my applications :grin:

Simon

As soon as the DUE is launched and people see tinkering going on with it, interest will come.

Probably true, although a lot of people seem to be struggling to find an application. Maybe we should be looking at what people are currently doing with other ARM boards for inspiration. You mentioned robotics, that would be an obvious choice I think.

the bootloader for the arm

ARMs have a built-in bootloader and I would assume Arduino will use that. As for when it will be released, only the Phantom knows.


Rob

Robotics would be a great application for the DUE as there is no such thing is overpowered in robotics ]:D But also, look at the things people do with the UNO and its predecessors, make videogames. Yes, that works with an UNO, but imagine what an ARM powered board can get you? Not pixel games, but more advanced graphics. Color even? While I think of it, I can think of a million applications I would like to build with it, if it wasn't for time, money and knowledge.

Robotics, Videogames, Web enabled media player (streaming? operated through phone apps? you name it!) Real time video processing as in security camera with facial recognition Main node for an advanced domotica system

And the obvious: Have it run Linux! Since they get Linux running on everything but a toaster, why not? There's a lot of possibilities I can think of. Are they all useful? Perhaps not, but is that a requirement? I see many projects online that are not really useful but fun or just cool because you made it yourself.

I think the main thing that could be used is better graphics capabilities, the largest screen I can get is a 2.8 in tft, which at best does 3fps, the abilty to play a video or actually decent size screen would be awesome, and I never heard anyone complain about there processor going to fast, which could mean being less restrictive with large plots of code, cutting and making it more efficient to be fast enough also I haven't seen much on the specs but would this have better quality analog digital converters or a higher pwm capability? Most people probably don't need it but it'd be nice to have, when(if?) It comes out I can't wait to get one( assuming I can afford it)

Raspberry PI does seem promising... now if I want to add wifi capability to my project I need at least 50-80$ to buy a wifi shield, but with raspberry I could use a wifi dongle that costs like 10$...

want to use bluetooth on arduino ? again you have to pay XX $ to buy a shield

but wait there is more - need more computing power, basic video recognition etc. ? Wow wait a second, you can actually connect a USB camera to it, or a memory stick... all of those components are actually damn cheap unlike most of the shields

Want better graphics ? it has an HDMI & composite output and it has been shown to work pretty well (QUAKEIII :-D)

I love arduino but rapsberry PI just seems so much better on so many levels... probably it would be best to transfer the whole idea of easy to use IDE to Raspberry PI. Some ppl will probably first use raspberry toghether with arduino for stuff such as wifi etc. But soon it would become obvious that it makes more sense to use just raspberry PI

I am not trying to be negative but that's just what I think will happen sooner or later... ohh and let's face it - they said it will cost 25/35$... that's not much more than the price of arduino UNO! sure even if it turns out that the final price will be like 45-55$ that's still really cheap

ohh and before you start your rant that arduino is a microcontroller and rapsberry pi is overkill for most things... sure I know that but I know that it is cheaper to spend 35$ on Raspberry PI + 10$ on a wifi dongle than 80$ on a wifi shield

overkill for most things

I think that's the way things are going and I have no problem with that. For example using an 8-pin uC in place of a 555 timer, using a $2 32-bit LPC for a simple app that uses 10% of the chip's ability. Why not, it's all so cheap.


Rob

Hi again Rob,

A Wave additive synth is an audio synthesizer that generates sound by "adding" harmonics (waveforms), a bit like an organ. I think Roland made such synths in the 80's. The older variant (soustractive) was using high and low pass filters to "remove" harmonics from some generators (square/triangle/sawtooth).

But back to 32-bit, I'm convinced there is another reason. Probably one big hurdle for Arduino people is to get their software right. 1.0 was a terrible mistake that broke everything, they need to fix it. At least that's how I see it. Otherwise, why would the core team be so silent about these issues ?

On the power of 32 bit, it looks the normal way of evolution to me, and sure no overkill if it's the same price. I made a small 16-channel scope with a MEGA, and it cycles at 1 sample/sec. Not really a big issue for temp sensors, but clearly not the way to go for generic purposes ...

get their software right

I read somewhere that it will support debugging and other stuff, I wonder if that means a new IDE, maybe based on VS as that's what Atmel have moved to.

it looks the normal way of evolution

Yep, when there is a large selection of cheap 32-bit chips right down to 8 pins there will be little or no use for an 8-bitter.

Here's a thought, I recently used an LPC as a 12x12 digital mux in a design. Low speed compared to hardware but fast enough for the application and run-time configurable. When things are farst enough and cheap enough you can use them for all sorts of things previously unheard of. Why not have a 32-bit ARM working as an AND gate?


Rob

Well again I'm not completely sold on the Due upgrade path, at least for myself. My most standard usage of the arduino is to prototype a project and then migrate the code to a standalone DIP chip into the new project, not dedicate a complete arduino board to every project. As this old guy will not willingly build using SMD multi-pin components, there is little use for me for ARM based chips. So I wish them luck, but I most likely will trail far behind any early adopters of the Due.

Lefty

I have no specific use for an ARM board I admit, but I like the idea of designing one and the neat things one could do with the SAM chip. So I'll persevere with my "Due compatible" design for the time being in the hope that we get more details soon.


Rob

DUE compatible ? Did you see the design specs somewhere ? I'm definitely interested. Even a short list of features.

I tend to think to migrate towards some STM32 based chip. Not clear how it works, but there are boards with serial bootloader like Arduino. On some other forum I also saw some link to 4x4cm cubish stacks with 32-bits from ... Atmel by a German startup. Maybe you could be interested...

Did you see the design specs somewhere ?

I wish, that's why I used the quotes :)

I decided to just go my own way for the time being. It won't be physically compatible and maybe not even electrically compatible but I hope to have it compatible with the tool chain. Like the rest of us though I have to wait and see.

Even a short list of features.

• SAM3U 96MHz ARM Cortex-M3 processor. • A 72-pin backplane with 8 addressable daughter boards. • Ability to individually reset daughter boards. • 48-pin connector for memory expansion, fast IO and debugging. • Three interrupt lines allow stackables to provide vectored interrupts to the CPU. • Software control of power to the system and external devices. • 512k bytes of external SRAM (expandable to 16.5Mb). • Battery backup for SRAM and the SAM3U’s RTC/RTT. • Serial PROM for non-volatile system parameters. • High-speed microSD interface. • Serial Synchronous Controller I2S interface for CODECs, DACs etc. • External watch dog and power monitor chip. • Eight 15v-tolerant analogue inputs. • 32 digital IO with 5v or 3.3v support in both directions. • 16 high-current digital IO with 5v or 3.3v support in both directions.

The schematics are nearly done, here's a render of a rough PCB layout.

I probably won't commit to anything until I see the real thing, it depends on how long that will be. Also I'm still leaning about the SAM and may make changes as I get to know more about the chip.

I have asked for input with little response but am still open to ideas.

I guess I'm, saying that I don't think an ARM Arduino will be a dud and I'm putting my time (but not any money yet :)) into it, it's a logical extension to the Mega. As for the Due itself we have no details so that remains to be seen however just being part of the Aduino fraternity will help a lot.

On some other forum I also saw some link to 4x4cm cubish stacks with 32-bits from ... Atmel by a German startup. Maybe you could be interested...

Very much so, if you can find it please post a link.


Rob

Graynomad:

I want to make a wave additive synthesiser

I have no idea what that is :) but yes it needs some neat projects (audio or visual or even audio-visual I would think) to show what it can do.

I am trying to figure out what precisely to do with it that I could not do on a 1284-based Arduino OR a Linux capable board like Beagle bone.

I have a feeling that 90% of apps can be done with less and the 10% that need more have other options. Still there are other options to the standard Arduino and it does well. Just being an official Arduino port will count for a lot.

I've always thought the mega1284 was about perfect, I think I'll be holding off on my SAM design for the time being and might go back and revisit something similar I did a while back with the 1284.


Rob

i'm just learning about the due right now and i'm very new to programing and the arduino so excuse the ignorance. but would the due be better fit for more intense lcd displays? something that can actually handle some more intense rendering/graphics?

Yes, I imagine for people who want to have a large display it’d be awesome, it would be really cool if it had some capabilities to do what processing does on the computer, you know instead of sending to a computer to graph data, maybe it can be done directly with a nice size screen
it’d be awesome for people like me who really have no previous programming experience other than arduino, to be able to use the same language to do more powerful applications
I know I could learn more powerful processors and figure it out the long way, but I’ve found it much easier to see something like digitalWrite at first then dig deeper and see how it does it later once I have a basic understanding,
Without the arduino platform it’d be like trying to explain how to use bitwise and direct port access to blink an led, while at the same time figuring out timers and other hardware that is hidden at first, it gives people a chance to learn the flow and basics first then if they need or want to dig deeper
just for that reason I think its worth it to develop the due,

it gives people a chance to learn the flow and basics first then if they need or want to dig deeper

That is the beauty of the Arduino concept I believe, you are isolated from the nitty gritty but on any line of code you can deal directly with the hardware.

something that can actually handle some more intense rendering/graphics?

The SAM is probably not the best ARM choice for high-res graphics rendering, many ARM chips have LCD interfaces built in and the SAM doesn't. I would not expect to get 3D rendering in anything like real time, but I don't think that's what Arduino users need from a board. The sort of thing I reckon people would like such as bar graphs, soft instruments etc would be a walk in the park for it.


Rob

I’ve seen a lot of projects and code here that would be more suited for the Due than an UNO.

When I first checked out Arduino I saw the IDE and was almost sold just on that. I see the UNO as a development board, not as a primary platform. With the UNO I should be able to use the IDE to program lots of different ATmegas and ATtinys to run on their own and it even comes in handy to test an idea now and then. And the newer ATMEL MCU’s have their own clocks even (3 less parts for <= 8 MHz), so I bought.

I’m not up to surface mount so for me the Due might be a way to get the power without building what I can’t.

I did just order a Teensy++ because of the full USB speed and the extra memory and the price. I got w/pins so I can just set it in a breadboard and wire it up.