XMEGA

Go To Last Post
440 posts / 0 new

Pages

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It's been proven already that certain pieces of silicon in the industry will always return the same RNG's at a certain point in time after powerup.

No one's perfect!

Regards

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You could do like some companies do and use a Lava Lamp for RNG. Here is a tutorial from MakeZine.

http://blog.makezine.com/archive...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

sxpilot450 wrote:
It's been proven already that certain pieces of silicon in the industry will always return the same RNG's at a certain point in time after powerup.

No one's perfect!


But if the random source depends on parameter like temperature, supply voltage, aging... how can a RNG repeat its number series. Maybe we're talking about a different kettle of fish. Have you read the artcle of the link I have posted above?

Regards
Sebastian

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The smaller the feature size, the more problems I'm seeing in silicon!

Regards

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Maybe we're talking about a different kettle of fish.

I think sexplot450 is talking about pseudo RNGs

Quebracho seems to be the hardest wood.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

yes, without running these values u pull from e.g. a smartcard processor through a proven random number algorithm, I wouldn't trust em!

Regards

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

sexplot450 wrote:
yes, without running these values u pull from e.g. a smartcard processor through a proven random number algorithm, I wouldn't trust em!

Arrrggghh! That's the whole point! There is no such thing as a "proven random number algorithm"!

Quebracho seems to be the hardest wood.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I think it is Martin Gardeners "Mathematical Carnival" that has a particularly interesting proof of the non existence of random numbers.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

theusch wrote:
Digging around in the docs, it appears that STK500 and ATAVRISP2 and similar "serial ISP" tools are dead for the Xmega. So if I get one into my sweaty palm, is it JTAGICE2? STK600? a tool similar to ATAVRISP2?

Does anyone have any idea on the programmer/debugger support planned?

As I understand it the XMEGA:
1) No longer supports ISP over SPI making the conventional programmers obsolete
2) There is JTAG support but the only device suggested at the moment is JTAGICE2. Could the Dragon be used for downloading via JTAG. It seems unlikely it will be supported as a debugger, even on those processors within the 32K limit.
3) There is the new PDI interface. Is there any device out there that currently supports PDI. Is there potentially an "XDragon" under development?

The replacement of debugwire with PDI is fine but where is the hardware to support it?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Trevor_G wrote:
theusch wrote:
Digging around in the docs, it appears that STK500 and ATAVRISP2 and similar "serial ISP" tools are dead for the Xmega. So if I get one into my sweaty palm, is it JTAGICE2? STK600? a tool similar to ATAVRISP2?

Does anyone have any idea on the programmer/debugger support planned?

As I understand it the XMEGA:
1) No longer supports ISP over SPI making the conventional programmers obsolete
2) There is JTAG support but the only device suggested at the moment is JTAGICE2. Could the Dragon be used for downloading via JTAG. It seems unlikely it will be supported as a debugger, even on those processors within the 32K limit.
3) There is the new PDI interface. Is there any device out there that currently supports PDI. Is there potentially an "XDragon" under development?

The replacement of debugwire with PDI is fine but where is the hardware to support it?


Actually, besides the JTAGICE2 is the AVR ONE.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Trevor,

As far as I know your options are JTAGICEmkII (later models), STK600 or AVR One. Of these the $190 STK600 is the cheapest way to get code into Xmegas but it won't allow debug.

Cliff

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quite a difference in price, considering you can program atmega's with a parallel port and a few resistors. And you can debug them with $50 jtag mkII clones. Hopefully some lower-cost alternatives will pop up for the xmegas as well, for hobby developers.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The datasheet I glanced through mentioned a bootloader, perhaps this is the method hobbyists will have to use.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

CountZero wrote:
The datasheet I glanced through mentioned a bootloader, perhaps this is the method hobbyists will have to use.

Which datasheet was that? I just looked through the one for 64A1/128A1 and it doesn't even have the text of the section about bootloader support (the usual SPM stuff) let alone any mention of an inbuilt bootloader.

I was hopeful, after reading your post, that they might be like the AT90USB chips and come factory loaded with the FLIP bootloader but sadly that does not appear to be the case as that seems to be a USB only mechanism.

Cliff

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It was in "ATxmega A Manual Preliminary" So perhaps not a datasheet per se

to call the information in that sparse would be too much :) only says TBD under bootloader, self programming.

Edit: And the document was found here: http://atmel.com/dyn/products/pr...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

CountZero wrote:
TBD under bootloader, self programming.
My expectation is that section will document the capabilities of the device to support a bootloader -- not that the device actually ships with a preinstalled bootloader.

But, the AT90USB AVR's ship with a DFU bootloader and maybe XMegas will as well.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You are probably correct, but one always has hope...:)

A low cost programming solution could perhaps be a generic JTAG adapter and a svf file. Assuming one can generate one.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The only 2 Xmega devices are defined in AVR Studio 4.14 at the moment are:
ATxmega128A1 STK600-RC100X-13 STK600-TQFP100
ATxmega64A1 STK600-RC100X-13 STK600-TQFP100
Both are 100 pin TQFP devices difficult to prototype without an STK600.
So to prototype and program would require an STK600 ($199) and a STK600-TQFP100 ($99) total ($298) available now at Digi-Key.
The STK600 can program Xmega's using JTAG or PDI, however to do debug with JTAG or PDI requires a JTAGICE MkII ($318) at Digi-Key.

At present the AVRONE does not support any 8 bit MCUs and one can only guess at the price ($$$$$) :shock: .

Although I do have a STK600 with STK600-TQFP100 and a JTAGICE MkII I don't expect to do much debug as I've only used it once in the last six months and that was to re-load the bootloader on the AT32AP7000 on my STK1000 :wink: .

I'm ready for Xmega samples :) Send SOON........

Dyan...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

My reading of the data sheets and the ATXMegaManual (from the memory section of both) is that there is a separate bootloader section as in the Mega. As a result you have still got to find a way to load the bootloader in the first place.

I am looking at the ATXMega...A4 family. On these data sheets there is no mention of JTAG it appears to be PDI or nothing. Which really does pose a problem. Perhaps by the time the chips are available......

Trevor

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

KV9R wrote:
The STK600 can program Xmega's using JTAG or PDI, however to do debug with JTAG or PDI requires a JTAGICE MkII ($318) at Digi-Key.

So STK600 and JTAGICE mkII can do PDI. I wondered about that but couldn't find anything in the AVRStudio4.13 documentation (other than a label on a connector on the ST600). I asume the support comes in AVRStudio4.14.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes AVR Studio 4.14 supports Xmega128A1 Rev D
See below...

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

See Atmel document 8077 on the ATXmega A. It has documentation on the PDI opcodes. I haven't studied it in detail. At first glance, it would appear to be similar in nature to SPI, but different in the details of what bits go where. So anyway, cheap programmers are probably possible. Somebody with spare time on their hands can go study that document and tell the rest of us :)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

KV9R wrote:

At present the AVRONE does not support any 8 bit MCUs and one can only guess at the price ($$$$$) :shock: .

Expect it to be $599.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Does anybody know, what the power consumption of an XMEGA is, if it is running at 32 MHz at 3,3 Volt?

The datasheet is empty, regarding this question.

Would be important to know, to get a feeling how efficient the XMEGA is, regarding the MIPS/mW.

Regards:

Uwe Fechner

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

New Xmega flyer on Atmel's Web site:
http://www.atmel.com/dyn/general...

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for the notice, Lee!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

davidpr wrote:
Quite a difference in price, considering you can program atmega's with a parallel port and a few resistors. And you can debug them with $50 jtag mkII clones. Hopefully some lower-cost alternatives will pop up for the xmegas as well, for hobby developers.

If there's enough demand I'm guessing the same companies that make cheap hardware for the regular AVRs will eventually support the xmega.

But, being realistic, hobbyiest purchases of the xmega will likely never be more than a tiny drop in the bucket for Atmel. So, from their perspective, a $200 tool (the STK600) is very reasonable. I have my STK600 and can say they're not likely making any money on it. It's a much more complex product than the STK500.

And the xmegas, so far at least, only come in surface mount packages which aren't very hobby friendly either when you have to design a custom PCB or buy an expensive adapter to use the part. Perhaps they'll offer an inexpensive "USB stick" with an xmega on it someday that can be programmed via USB like their ATS90USB parts?

Plus, with all that code space, you're not likely to develop any sort of real large application without wanting an ICE anyway. I have my complaints with the AVR team, but the price of the tools isn't one of them. For what you get, and 99% of their target market, they're a good value.

My biggest complaint right now is THERE ARE STILL NO XMEGAS! The end of March is almost here... Will Atmel make their promised date?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

From the Xmega introduction slideshow:

Device       Eng Samples Samples Mass Prod
ATxmega128A1 Q1-08       Q1-08   Q2-08
ATxmega64A1  Q1-08       Q1-08   Q2-08
...

Since we are now into Q2, those samples we signed up for during introduction week should be on their way. And we should see listings at the distis any day now...

Lee

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

1) Atmel is up to old tricks again, locking the Xmega A family manual DOC8077A against copy.

2) "Darn, those Xmega output buffers sure say nice things about each other."

Quote:
Any WG output from the Timer/Counter 0 are split into a complimentary pair of outputs when any AWEX features is enabled.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

What do you expect, probably written by a poor university graduate, smelling is not one of their strong points...or perhaps the smell checked new butter. :?

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

if you save 1$ per day when those Xmega will be availaible at our favorite online store, you will be able to buy the complet chain with dozen atxmega.
Do you remember when the atmega328 was announced?
I still wait.
i'm ready with my jtagice mkII.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello new to the forum. Well browsed it but haven't posted. I do however have some ATxmegas. Just awaiting the STK600....

These are prices i was quoted for qty 100 (AUS prices).

ATXMEGA64A1 Approx A$5.80 each
ATXMEGA128A1 Approx A$7.65 each

Ciao

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Where you we able to get the XMegas?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello i was able to get them out of Arrow Asia through a contact.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for the info!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

No worries, in fact I most likely will have more than i need... I will know for sure very soon. If this is case and someone is in desperate need to get these and willing to pay freight I would be happy to pass an extras on.

I'm located in Brisbane, Australia.

Ciao.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

ATXMEGA64A1 Approx A$5.80 each
ATXMEGA128A1 Approx A$7.65 each

With the US$ roughly equal to the AUS$, the XMega compares quite favourably in price with current models roughly the same "size" but with all the XMega toyz. DigiKey qty. 100 for a Mega64 is US$6.88; '128 is $8.75.

Farnell-Australia shows prices for the Mega64 and '128 about twice that.

Lee

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Farnell-Australia shows prices
...which of course one tries to avoid unless absolutely necessary.
https://www.soanarplus.com shows about half the Farnell's prices.

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

This is a combination Xmega/STK600 question:

DigiKey and Mouser have a nice stock of STK600, as well as the -TQFP100 adapter kit. The Xmega 'A1 models that will be first available, as mentioned above, are in that package.

But the pinout is quite a bit different from other AVR Megas in TQFP100. Is there an Xmega adapter plate in the STK600-TQFP100 kit? I'm guessing "yes" as that is listed as "Related Device"s on the Atmel Web site page.

Lee

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

theusch wrote:
Is there an Xmega adapter plate in the STK600-TQFP100 kit? I'm guessing "yes" as that is listed as "Related Device"s on the Atmel Web site page.

I can't answer that question.

But when I received my STK600, it had one 40 pin PDIP "Top-Card " and one 64 pin TQFT "Top-Card ". There were 6 or 8 adapter cards, as well. I actually inserted each of them into the STK600 "Top-Card " sandwich. For each adapter card, a list was provided with the currently supported controlers that were supported be each adapter card.

So, maybe one of the adapter cards that comes with the STK600 will also support the Xmega "Top-Card ", as well. That, or there is a pending AVRStudio/STK600 software upgrade to accomodate the XMega.

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes with the STK600-100TQFP there is an adaptor plate it is the RC100X-13... Waiting for samples :)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

theusch wrote:

...
Quote:
Availability and Pricing.
The first devices, ATxmega128A1 and ATxmega64A1 are both offered in 100-pin TQFP and BGA packages and are available now.

...
Lee

Sigh. Above quote from the February release notice.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

theusch wrote:
Darn, John, you just ain't excited?!? With just the Xmega and a MMC you can make a 1Msps datalogger and have enough time left over to do digital filtering and/or data compression.

Or add a display and have a Msps digital 'scope with the only additional components being the ranging/protection on the probe inputs.

Or add on to that with logic analyzer.

Or finally have the AVR be able to do high-quality sound capture.

Or finally have the AVR have enough oomph to drive streaming peripherals like AC97 codecs.

Or finally be able to match Jesper's miniDDS without resorting to cycle counting.

After the Xmega learning curve all of the above are going to become starter apps that can be tossed together in a day or two. We already >>know<< how to drive our displays with AVRs; we already >>know<< how to communicate; etc.

And you aren't excited???

Lee


I'm like a little kid waiting for Christmas to finally get here. I've been doing more reading into the Xmega datasheets. Some items:

-- No longer will Peter/danni be able to tout the x51 settable interrupt priority as an AVR disadvantage. [Though I've never run into a case where it mattered on an AVR anyway.] Besides three levels, within each level a priority can be assigned. An interesting feature is that the low "level" group can be serviced round-robin.

-- Adding to my list of apps above to tackle with the Xmega, let's exercise another subsystem and add an IrDA link on one of the USARTs to remote-control our 'scope/logic analyzer/datalogger/DDS. Still a near-single-chip solution.

-- What other subsystems could we possibly use? An obvious little-bitty one is RTC using the internal "high accuracy" 32kHz (datasheet has no numbers as of yet) oscillator.

-- Use a "spare" PWM to recharge our own batteries when an external wall-wart is plugged in?

Lee

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sounds like the drool machine is running full speed. I'm waiting, also!!!! Not holding breath, though.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

For the toyz above, it should really have a graphical LCD, right? Hmmmm--if enough pins in the right places, could use a memory-mapped one...

Oh, yeah, I forgot one item to add to the list above:

-- Built-in quadrature decoder. So add "endless" knobs like on my TDS2xx 'scope. Those are pricey, though--$6 and up. But for US$0.99:
http://www.goldmine-elec-product...

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Recently posted on the Atmel Web site is the updated datasheet (rev. B) for the A1 series.
http://www.atmel.com/dyn/product...

There are now at least a few numbers in the datasheet--the power consumption numbers during sleep are especially impressive.

Quite a list of errata. Notable to me are the bootloader section not working, and the 12-bit ADC being more like a 10-bit. :(

Lee

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Darn, Lee, you just ain't excited?!? :)

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

the drool machine is running full speed.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

Pages