Good news for those waiting on the ATmega1284p and 16k sram!

Go To Last Post
56 posts / 0 new

Pages

Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Contacted Digikey USA for the ATMEGA1284P-AU-ND today quoting 320 pcs minimum qty and delivery 4/2x/09.. about to order the 320 pcs minimum qty for a product beta run and decided to ask about the "non-stock" status and was this going to change to a "stock item" in the near future...got this news :!:
Digikey..
"Our Prod Manager says he was glad you checked on this. He is going to change the status to regular stocking and parts should be ordered Monday. You could place an order then and it would be backordered. You may then order as little a qty as 1."
that should make some Freaks happy :D

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

Just my opinion:

I would not believe DigiKey or Atmel about stocking/shipping dates. I would not design-in a product unless it was already shipping and stocked in large quantities. I got royally screwed by Motorola, which was one reason I moved to Atmel. They made glorious claims about this new micro that was going to ship in a month, I used it in a design and then on ordering it found that it would actually be shipped in 6-months. I never found out if it actually shipped in 6-months, and I've never used another of their products. I suspect that they changed their name to Freescale hoping they could fool the slubs they had screwed into thinking they were a different company.

Now Atmel is playing at exactly the same game. I will give them the benefit of the doubt and assume that their problems are due to incompetence, not malevolence, but I am beginning to wonder.

I briefly considered using an Xmega in a design a year ago. Good thing I lost my virginity with Motorola, so I didn't make that mistake. I have consistently received optimistic forecasts from DigiKey for delivery of Atmel products (not Xmega) that did not pan out. One order was promised in October, moved to December, then January, then February, now it is promised for mid April. I'd move on to some other company if I could convince myself that any other company wasn't as lying and incompetent as all the rest. I'll never specify any device that hasn't been in production and delivered in quantity for at least 3 months.

Again, just my opinion.

Smiley

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

Good point Smiley, In a similar vain, I've spec'ed an Xmega for a project because Digikey had a receive date of 2/14/09 so I ordered 25 pieces for a test run the week before that.

As everyone knows, they didn't arrive. Due date became the 2/15/09 then jumped to 4/2/09.

Luckily I didn't bet the bank on it, and I've shifted over to Atmega2560's for now.

I figured if I was going to have to learn a new chip inside out, it might as well be the Xmega with all neat toys inside just to play with in my spare time (crypto engine, 12bit a/d, better interrupts). Nice dream.

-Raliegh

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

Quote:
Now Atmel is playing at exactly the same game. I will give them the benefit of the doubt and assume that their problems are due to incompetence, not malevolence, but I am beginning to wonder.

Quote:
I'd move on to some other company if I could convince myself that any other company wasn't as lying and incompetent as all the rest.

I agree with you Smiley..ATmel marketing seems like one big lying sack of sh*t these last two years with regard to delivery of Avr's...I sought of thought it might be because they supposedly were going fabless I thought about a year ago..and maybe they forgot that when you go fabless you must remember to order some wafers (and transfer the recipe)! I do not know but the double wammy of bs about delivery, a perception of insufficent stock qtys of many popular AVR chips in distributors , and many new chips with alot of Errata (thats why I would think its a no brainer to stay away from Xmega for at least another year or at least until Errata X No Known Errata).. with a shrinking global semi market and the fact they seem borderline financially ,makes me wonder if Atmel doesnt get it act together soon IMSO they may find the grim reaper on their doorstep early 2010 !..o well its all speculation until the sh*t hits the fan....actually I hope they find a partner so they would be stronger financially!

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

from afar, the Xmega makes little sense, competitively.

I hope that Atmel recovers from Microchip's unethical or worse attack akin to that suffered by Nancy Kerigan

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

Right now, I begin to wonder if for many applications, specially for short runs, an ARM cortex M3 like STM 32 from ST would be the way to go. Cheap in short runs, more powerful in many fields than Xmega, and, most of all, available right now. Cheaper dev tools, like JTAG, nice eval boards, like STMcircleI and II.

I only know of one commercial product that will be developed around Xmega, but given the board of directors and inertia in that company, probably Xmega's will be obsoleted before that product is released to the market...

Guillem.
"Common sense is the least common of the senses" Anonymous.

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

Quote:

I only know of one commercial product that will be developed around Xmega,

Well, make that 1 1/2 products. ;) If they ever become readily available and they [Xmega] work the way I think it will according to the datasheets I've got a quad high-speed encoder app that would not be suitable for a single AVR otherwise.

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

Ugh, Digikey changed the Xmega backorder from 4/2/09 to 8/29/09. Doesn't look good for any Xmega projects this year either :(

-Raliegh

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

If you want to play, Avnet has a supply of ATxmaga128A1 and ATSTK600-TQFP100 as of this date.

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

How soon we forget just how long it took for the mega256 family to actually go from vapor to silicon.

Actually we had buggy samples very early on (the warnings on the sealed bags about moisture exposure destroying the parts and careful handling required was somewhat scary), but the fact that Winavr didn't support the chips (also nutos) kept the samples in the bags.

Today we could actually use the things in product, but we also are using the Philips Arm7 family too. I still like the AVR's better than the Arm7 for the same code size, the Arm is just harder to use and has a worse jtag IMHO.

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

been waiting on the atmega644p from digikey which was estimated in stock 28/02, still nothing (one person said 18 march, another said 1st april when i enquired). sent an email cancelling the order (no response though...) and ended up going with nuhorizons as their website says they had 450 in stock (but it hasnt shipped yet).

if only mouser would stock this part would make things a lot easier.

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

How can a semiconductor company not have a chip fab? Is that like a car company not having an engine factory?

Imagecraft compiler user

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

Quote:

been waiting on the atmega644p from digikey

???

Atmel ATMEGA644P Distributor Inventory Check
Device 	Distributor 	Country 	Stock 	Date 	Buy
ATMEGA644P-20AU 	DIGI-KEY CORPORATION	US	1929	03/12/2009	Buy
ATMEGA644P-20AU 	AVNET ELECTRONICS MARKETING	US	1043	03/12/2009	Buy
ATMEGA644P-20AU 	NU HORIZONS ELECTRONICS	US	887	03/12/2009	Buy
ATMEGA644P-20MU 	AVNET ELECTRONICS MARKETING	US	310	03/12/2009	Buy
ATMEGA644P-20MU 	ARROW ELECTRONICS	US	200	03/12/2009	Buy
ATMEGA644P-20MU 	DIGI-KEY CORPORATION	US	113	03/12/2009	Buy
ATMEGA644PV-10AU 	DIGI-KEY CORPORATION	US	1191	03/12/2009	Buy
ATMEGA644PV-10AU 	AVNET ELECTRONICS MARKETING	US	780	03/12/2009	Buy
ATMEGA644PV-10AU 	NU HORIZONS ELECTRONICS	US	125	03/12/2009	Buy
ATMEGA644PV-10AU 	ARROW ELECTRONICS	US	15	03/12/2009	Buy
ATMEGA644PV-10MU 	AVNET ELECTRONICS MARKETING	US	330	03/12/2009	Buy
ATMEGA644PV-10MU 	ARROW ELECTRONICS	US	127	03/12/2009	Buy
ATMEGA644PV-10PU 	NU HORIZONS ELECTRONICS	US	2201	03/12/2009	Buy
ATMEGA644PV-10PU 	DIGI-KEY CORPORATION	US	149	03/12/2009	Buy

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:
another said 1st april when i enquired
Well then....
Quote:
if only mouser would stock this part would make things a lot easier.
especcially now when you can order via X-on and don't pay huge shipping costs.

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

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

Quote:

if only mouser would stock this part

??? Flavours in all package types are shown in-stock at Mouser.

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:
???

should have been more specific. atmega644p-20PU - not listed in your results. it is also a non-stock part at mouser, minimum 225 quantity and your normal 3 months lead time.

thanks for the tip on X-on John, had no idea about them! (though in this case <$800AU from digikey for 100, >$1000 from x-on, plus no stock)

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

NewKleer wrote:
should have been more specific. atmega644p-20PU

http://de.farnell.com/atmel/atme...

Peter

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

de farnell only ship to germany unfortunately. the australian farnell ran out of stock just the day before i was about to order (although it was ~$300 more than from digikey)

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

Quote:
de farnell only ship to germany unfortunately.
HAHA...so all you need to do is to get someone in Germany to buy them for you and then ship them to Oz...now I would never do anything like that of course to get around red tape. :wink:

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

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

bobgardner wrote:
How can a semiconductor company not have a chip fab? Is that like a car company not having an engine factory?

You are aware of the economics e.g. multi billion $$ to build a modern fab. Fabless is the only way to go for anyone unless you are getting up to the level of a 50 billion dollar plus company.

Back to the original post DigiKey do now have the ATmega1284p listed in small quantities but the ship date is 25/06/2009 so not too soon....

Ifor

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

How do you get your chips made at someone else's fab? Ask them nice? What if they are all booked up? You just keep pushing back your release dates? Hmmmm... that's exactly whats happening. Microchip have any fabs with capacity left? (haha. Dark joke)

Imagecraft compiler user

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

I guess I'm just too stodgy an old guy, only running apps at 7.37MHz or 3.69MHz except for a couple. So -10 or -20 (or -8 or -16) work fine for us.

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.

Last Edited: Fri. Mar 13, 2009 - 02:39 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

How do you get your chips made at someone else's fab? Ask them nice? What if they are all booked up? You just keep pushing back your release dates? Hmmmm... that's exactly whats happening. Microchip have any fabs with capacity left? (haha. Dark joke)

I don't know but ARM seem to have become possibly the most successful semiconductor business on the planet (with the possible exception of Intel) using this business model. What they sell is simply intellectual property rights, not silicon - they leave the grubby stuff to others.

As the world moves (does it?) to more and more widespread usage of FPGAs I'm guessing it's the VHDL coders, not necessarily Altera or Xilinx who are going to make all the money out of it.

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

The most likly cause of delays is actualy failing to fix critical bugs properly and having to do more re-spins. Your release schedual may well have one re-spin in it but if you still have a critical bug your looking at 3+ months if your lucky. The real problem is letting the marketing guys get going and anounceing products befor getting confirmed good silicon. I don't know which fab Amtel use but in the curent ecenomic times I would be very suprised to find that getting fab time is the issue.

Ifor

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

Marketing people sell vapour, pretty much always have. I never trust the marketing releases - talk to the FAE's, they usually have a much better idea.

We know why the Xmega was announced so early - it was to stall of other competition. Who else has an 8 bit processor with such flexible IO and so much memory?

The problem with moving to a new fab is that fab's don't tend to have compatible processes. Atmel uses their own fabs, and their own processes. Chip designs are inevitably targeted to a particular process - re-spinning them for another fab and process could take 6 months and in fact may result in a very different chip! A fab generally supplies lots of information about its process that is used in the creation of the design.

JD

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

So if a half dozen other companies wanted to make AVRs with their own mix of ram/flash/eeprom/ioports/iodevices, every different companies' chips would have diff pinouts, like arms, but mktng could brag that they were 2nd and 3rd sourced. Is there ANY arm cpu part number from two diff companies that is pin compatible? So does this mean that they are compatible, or incompatible?

Imagecraft compiler user

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

clawson wrote:

I don't know but ARM seem to have become possibly the most successful semiconductor business on the planet (with the possible exception of Intel) using this business model. What they sell is simply intellectual property rights, not silicon - they leave the grubby stuff to others.

Qualcomm may have more revenue on sheer Patents/Intellectual Property than ARM. But Qualcomm needs to or chooses to spend most of their income on lawyers to try to expand their clutch, e.g., W-CDMA.

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

bobgardner wrote:
So if a half dozen other companies wanted to make AVRs with their own mix of ram/flash/eeprom/ioports/iodevices, every different companies' chips would have diff pinouts, like arms, but mktng could brag that they were 2nd and 3rd sourced. Is there ANY arm cpu part number from two diff companies that is pin compatible? So does this mean that they are compatible, or incompatible?

While they may exist I cannot, off hand, think of a dual-sourced ARM chip design.

But the key thing about ARM though is that the core of all the chips, made by about 100 different silicon fabricators, aimed at various different markets is that the same toolchains and, possibly most importantly, developer skills apply to them all. This is what makes ARM programmers some of the most valuable out there. It's also like a self supporting market. Once you have invested in the tools (ARM Multi-ICE cost $3,500 each for example though I think a lot of people prefer Lauterbach's these days) then when it comes to the next design the temptation is to pick ARM again as you already have the tools/skills and because they are so popular there's almost certainly one of those 100 companies that have produced a chip doing what you need from the lowliest sub-$1 NXP LPC devices to $50+ behemoth devices for specific designs such as HD TV reciever/decoders. Things like the Coretex cores are just widening their spread even more so now it's possible to lay down an ARM core in an FPGA for example.

Cliff

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

ifor wrote:
bobgardner wrote:
How can a semiconductor company not have a chip fab? Is that like a car company not having an engine factory?

You are aware of the economics e.g. multi billion $$ to build a modern fab. Fabless is the only way to go for anyone unless you are getting up to the level of a 50 billion dollar plus company.

Back to the original post DigiKey do now have the ATmega1284p listed in small quantities but the ship date is 25/06/2009 so not too soon....


There's several fabless semiconductor companies around. Also cutting edge ICs like nVidia (has always been fabless) and AMD/ATI (who are now going fabless too).
AMD’s fabless status official http://www.fabtech.org/news/_a/a...

TSMC (Taiwan Semiconductor Manufacturing Co.) is the world's largest dedicated semiconductor foundry:
http://www.tsmc.com/english/a_ab...

Quote:
Fab Capacity
TSMC is the world's largest dedicated semiconductor foundry, providing the industry's leading process technology, library and IP options and other leading-edge foundry services. TSMC operates a six-inch wafer fab (at fab 2), five eight-inch wafer fabs (at fab 3, 5, 6, 8 and WaferTech) and two 12-inch wafer GigaFabs (at fab 12 and 14). In addition, the company has a $1.2 billion joint venture fab (at SSMC) with NXP Semiconductor, which is operated in Singapore. Keeping pace with the trend of clear industry growth, TSMC is committed to significant capacity expansion: an eight-inch fab in Shanghai, China (fab 10.) In 2006, total capacity at TSMC exceeds seven million 8-inch equivalent wafers.

http://en.wikipedia.org/wiki/TSMC

Quote:
Various fabless high-tech companies such as Qualcomm, AMD, Altera, Broadcom, Conexant, Intel[2], Marvell, NVIDIA, and VIA are customers of TSMC.

http://en.wikipedia.org/wiki/Fab...
Quote:
The credit for pioneering the fabless concept is given to Bernie Vonderschmitt of Xilinx and Gordon A. Campbell of Chips and Technologies.
...
As with most technology-intensive industries, the silicon manufacturing process was and is cost-prohibitive, especially for these small start-up companies. These companies relied on using excess capacity from Integrated Device Manufacturers (IDMs) to manufacture the chips they were designing.

This was the birth of the fabless business model. Companies were manufacturing integrated circuits (ICs) without the need to own an internal fabrication plant. Simultaneously, the foundry industry was established by Dr. Morris Chang with the founding of Taiwan Semiconductor Manufacturing Corporation (TSMC). The foundry industry would become the cornerstone of the fabless model - providing a non-competitive manufacturing supply chain partner for these innovative and pioneering fabless companies.

During the 1990s, industry pundits acknowledged the financial success of fabless companies, such as Nvidia, Broadcom, and Xilinx, and such companies as Cyrix produced competitively-priced products, benefitting consumers and driving the global market for computing devices. As of 2007, the fabless model is the preferred business model for the semiconductor industry[citation needed].

When FSA was established in 1994, there were only three fabless companies - Cirrus Logic, Adaptec, and Xilinx - each with revenues in excess of $250 million. In 2007, GSA tracked 10 separate fabless companies that have each surpassed $1 billion in annual revenues.

The model has been further validated by the conversion of major IDMs to a completely fabless model, including (for example) Conexant Systems, Semtech, and most recently, LSI Logic. Today most major IDMs including Freescale, Infineon, Texas Instruments and Cypress Semiconductor have adopted the practice of outsourcing chip manufacturing as a significant manufacturing strategy.

Even Intel is cooperating with TSMC and have them manufacture SoC's based on Intel Atom CPUs.
TSMC To Build Intel's Atom-Based Chips: http://www.forbes.com/2009/03/02...
http://www.tomshardware.com/news...

(And btw. there's also car companies without engine factories. Many car brands share the same engine and it becomes more and more common to buy an engine from someone else or develop it togehter because engine development is very expensive. But most car companies are owned by another car comapny today, so they can buy engines and car bodies etc. from each other inside the corporation).

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

Fabless firms move up in top 20 IC vendor ranking
http://www.eetimes.eu/215800072

2008 Top 20 Semiconductor Sales Leaders ($M)
Companies marked with *** are fabless (Qualcomm, Braodcom and nVidia).

AMD recently spun off its fabs to GlobalFoundries. But that could actually cause problems from Intel, since Intel says says GlobalFoundries deal breaches AMD cross-licensing agreement

Intel Could Shut Down AMD's CPU Production Completely in Two Months

Jansen Ng (Blog) - March 16, 2009 12:27 PM
http://www.dailytech.com/Intel+C...

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

Digi-Key has posted new Ship Date Estimates for the ATmega1284P.

ATMEGA1284P-AU -- 5/11/2009
ATMEGA1284P-MU -- 4/11/2009
ATMEGA1284P-PU -- 4/11/2009

I believe these were posted on Monday.

Stan

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

ifor wrote:
Back to the original post DigiKey do now have the ATmega1284p listed in small quantities but the ship date is 25/06/2009 so not too soon....
Experience shows that you shouldn't hold out too much hope on the Digikey dates.

Even when they have chips from the fab, I still cannot get what I want. Several months ago, the mega328p-pu and -mu became available. However what I really want is the -au (TQFP-32) variant.

The Digikey dates for the 1284p show a similar pattern. PDIP and QFN before TQFP. Anyone know what the problem is here and why I cannot find the mega328p-au. Even AVNet and NU Horizons don't have dates for the mega328p-au.

--Mike

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

It seems obvious that fab capacity allocation is by $$ amount of job. Big jobs win. Bad for small companies.

Imagecraft compiler user

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

Quote:

Experience shows that you shouldn't hold out too much hope on the Digikey dates.

Well, on new stuff made out of Unobtanium I'd agree--but let's not drop it on DigiKey's doorstep. If they have the orders in to the manufacturers and the manufacturers give them a lead time or a ship date, surely that is the best they can do is to post that information. And repeat for the other distributors.

In particular re your Mega328P-AU situation: I would assume that if the little beasties were in fact "real" that one of the North American distributors would have at least some. The distributor check off of Atmel's Web site gives no sniff of any -AU anywhere (and few '328s period). So mere mortals cannot get any into their sweaty palms. Either it is the Unobtanium factor, or as suggested a big customer may be getting the first millions off the production line.

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:
Well, on new stuff made out of Unobtanium I'd agree--but let's not drop it on DigiKey's doorstep.
I agree. Digikey are just reflecting what they are being told by Atmel. It seems like AVNet have learned the lesson and simply don't publish Atmel dates anymore.

--Mike

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

sbennett wrote:
Digi-Key has posted new Ship Date Estimates for the ATmega1284P.

ATMEGA1284P-AU -- 5/11/2009
ATMEGA1284P-MU -- 4/11/2009
ATMEGA1284P-PU -- 4/11/2009

I believe these were posted on Monday.

Stan

Well, as of today, the date is still today, but the available quantity is still zero.

Phoo.

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

It looks like the Mega328 situation is easing up a bit

Atmel ATMEGA328 Distributor Inventory Check
Device 	Distributor 	Country 	Stock 	Date 	Buy
ATMEGA328P-AU 	DIGI-KEY CORPORATION	US	150	04/11/2009	Buy
ATMEGA328P-AU 	AVNET ELECTRONICS MARKETING	US	646	04/10/2009	Buy
ATMEGA328P-AU 	NU HORIZONS ELECTRONICS	US	300	04/09/2009	Buy
ATMEGA328P-PU 	MOUSER ELECTRONICS	US	1501	04/11/2009	Buy
ATMEGA328P-PU 	ARROW ELECTRONICS	US	306	04/10/2009	Buy
ATMEGA328P-PU 	AVNET ELECTRONICS MARKETING	US	46	04/10/2009	Buy

but '1284 is still unobtanium.

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:
It looks like the Mega328 situation is easing up a bit
The 150 "AUs" at Digikey need to be purchased as a whole tray. I would expect that situation to clear up in a few days. New Horizons and Digikey both have a $25 order minimum. AVNet is better in that regard.

I'm going to wait a few more days for Digikey as I have some other parts to get as well.

--Mike

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

I have just foud this post after doing a hunt for the 1284p (-mu : 44qfn case).

Can't find stock anywhere. We are planning to take them off our Raven boards just to get the first couple of prototypes going!

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

well since I started this thread actually believing this chip might actually ship and in spite of all the warnings from the Freaks to the contrary..need to update the mega1284p and add it to the "Atmel Hall of Shame" I have on the wall....as in the the list of Atmel Sales delivery date bullshit from various distributors and in this case Digikey (one of the poor Atmel Vaporware Distributors)
Atmega1284p- 4/11/2009 - Delivery Date Past Due
Atmega1284p- 5/11/2009 - Delivery Date Past Due

starting to understand why Atmel distributors seem to list so many Atmel parts as "NOT A STOCK ITEM" !!!!
as in unreliable vendor maybe?

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

Quote:
the list of Atmel Sales delivery date bullshit from various distributors and in this case Digikey (one of the poor Atmel Vaporware Distributors)

How is Digikey responsible for Atmel missing their ship dates?

Regards,
Steve A.

The Board helps those that help themselves.

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

Koshchi wrote:
Quote:
the list of Atmel Sales delivery date bullshit from various distributors and in this case Digikey (one of the poor Atmel Vaporware Distributors)

How is Digikey responsible for Atmel missing their ship dates?

Digikey is not responsible and didn't infer they were...quote "poor Atmel Vaporware Distributors" rather I feel somewhat sorry for the distributors who sell Atmel and quote delivery dates that the distributor says Atmel provides them with...especially when Atmel does not meet those dates!

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

On May 11th Mouser posted an Estimated Ship Date of 8/7/2009 for 300 each of the ATMEGA1284P-PU (DIP-40).
Apparently they are not going to bother stocking other packages.

Stan

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

interesting on the view order dates from mouser:

Factory Lead Time
50 Weeks

seems excessive, but at the same time seems just about right :)

good to hear mouser will be stocking it though, annoyed i could get all parts from mouser i need for my projects bar the picopower DIP avrs.

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

Is there any update on where you can get Atmega1284P

DJ

Thanks

Regards

DJ

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

Quote:
where you can get Atmega1284P
At yesterday's seminar it is still showing as "in development" rather than available now (which only means that you may be able to get samples, not necessarily production qty)

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

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

So whats need to be developed futher on this?

Thanks

Regards

DJ

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

Don't know, maybe they are just prioritizing production or bug removals, so it gets put in the back burner for a while.

It's amazing the number of products that Atmel pops out, I think too many, and I'm afraid that things may go bust.

I guess the Xmega in all flavours would have a higher priority than the Atmega1284P, and we know the painful delays encountered with it.

But if they rushed things out with bugs we would then complain about that.

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

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

29 sept
pr says

Samples of the ATmega1284P are available now. The device has a suggested retail price of $3.31 in 10K quantities.

ATmega1284, a non-picoPower version of the ATmega1284P, will be available in Q4 2009.

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

Whatever the reason for Atmels many delay is, I am sure marketing would benefit of being more open about the reasons for why we have all these delays. Not only we, but I believe that also the distributors may be leaded to believe that delivery is sooner than later even though they have the power to deliberately give us the most promising delivery dates, given by Atmel, to keep us as customers.

But as already said, shifting to another brand wont help since they are just as reliable. Problem is then that Atmel do not have to care, since if you leave for freescale, someone using Frescale is leaving them in hop that Atmel should be more realible.

Another reason for why xmega still is so hard to get is that marketing is holding it back until better times (okay, my opinion an belief) or they see that it wont compete against arm and thus xmega will never be released and reach the market in full scale due to economical policy. I hope I am wrong, but...

Regards
Vidar (Z)

----------------------------------------------------------

"The fool wonders, the wise man asks"

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

js mentioned on another thread that Atmel stated at a recent seminar he attended that the problem with the Xmega was that they tried to make the most complex part first. A high risk strategy like that can pay off, but can be disastrous if it doesn't work.

There are companies with very reliable deliveries who don't announce devices until they are actually available.

Regarding fabless: there are very successful fabless companies, and very successful companies with fabs. It seems to depend on good management.

Leon Heller G1HSM

Pages