[NEWS] The first Cortex-M7 is available now!

Go To Last Post
85 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A few days ago, ST launched the first Cortex-M7 and now it's available in markets.

 

Discovery kit with STM32F746NG MCU

 

And two Evaluation boards:

 

STM32746G-EVAL (Evaluation board with STM32F746NG MCU)

 

STM32756G-EVAL (Evaluation board with STM32F756NG MCU)

 

Six STM32F7 are available in Mouser. Some of Cortex-M4s have higher/equal cost than Cortex-M7s. e.g. STM32F746VGT6 and STM32F437VIT6.

 

Not so bad.

Here Cliff asked:

 

Am I the only one who sees M7 as a bit of pointless technology? If you have reached M7 why not simply move to A5?

and Alex answered:

 

Just power supply design for A5 is a complete nightmare. Then you need DRAM and Flash. They take board space and require careful routing.

 

 

M7 has a lot of performance and much simpler to use, not quite as simple as ATmega8, but still way simpler that A5.

Good answer

 

I'm not a big fan of Cortex-M7 but I like it.

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

Last Edited: Sun. Jan 3, 2016 - 07:53 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Surely Atmel has had a Cortex-M7 for some time?

 

https://en.wikipedia.org/wiki/AR...

 

And, apparently, also Frescale.

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

awneil wrote:

Surely Atmel has had a Cortex-M7 for some time?

 

https://en.wikipedia.org/wiki/AR...

 

And, apparently, also Frescale.

I could find ATMEL's M7s in DigiKey but for Freescale nothing.

neither Mouser nor DigiKey

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

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

Rohalamin Wrote:

I'm not a big fan of Cortex-M7 but I like it.

Have you ever used it?  Any other Cortex processor?

 

 

JIm

If you want a career with a known path - become an undertaker. Dead people don't sue! - Kartman

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB user

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

jgmdesign wrote:

 

Have you ever used it?  Any other Cortex processor?

 

Havn't work still with M7 but M3 (STM32F1) is the core that I'm working with it about 1 -1.5 year. in many cases an M3 is enough.

some guys sometimes say me that it's an old MCU. I don't know why they think it's getting old. it's so nice.

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

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

In addition to the Alex's answer, I have to say the package of an MCU is an important factor.

in many cases A5s have BGA package then this would be difficult and expensive for designing and making and soldering this type of MCUs/MPUs.(FreeScale has this type of MCUs with LQFP package though).

 

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

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

That's why we have the likes of RasPi - high end chips on a board at a price point the hobbyist can afford.
We just need more companies to do this with a sub $10 price.

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

Rohalamin wrote:

A few days ago, ST launched the first Cortex-M7 and now it's available in markets.

 

Discovery kit with STM32F746NG MCU

 

And two Evaluation boards:

 

STM32746G-EVAL (Evaluation board with STM32F746NG MCU)

 

STM32756G-EVAL (Evaluation board with STM32F756NG MCU)

 

Those Eval boards look pricey, but the Discovery STM32F746G-DISCO is impressive.

Sub $50, with Cap touch 4.3" LCD mounted, on a Board not much larger than the LCD.

 

I think these upper-end parts will come more with "LCDs attached", as that gives a stable code base to grow from.

 

 

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

The ST offerings make Atmel's look lame and overpriced. With the AVR, Atmel had a low (for the time) entry with the stk200/300. That worked for me.

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

Kartman wrote:

The ST offerings make Atmel's look lame and overpriced. With the AVR, Atmel had a low (for the time) entry with the stk200/300. That worked for me.

I agree. I like the STM32F4xx series. Using '415 and 407 for some time. I'd like to help out Atmel but they need a HAL like ST's, and a CubeMX like ST's for we pro's, and for commercial apps.

One just doesn't do bare metal on ARMs as on 8 bitters

 

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

Kartman wrote:

The ST offerings make Atmel's look lame and overpriced. With the AVR, Atmel had a low (for the time) entry with the stk200/300. That worked for me.

 

Yes, Atmel's M7 Eval offering is looking expensive, but they do have some nice good value XMINI offerings, (sub$10) just rather sparse ATM.

Seems everyone is now moving to do the dual pairing of [USB-Debug] and [MCU] which gives the best user experience.

 

The new Micro:BIT from BBC in this thread follows the same trend.

http://www.avrfreaks.net/forum/m...

has a Freescale USB/Debug chip and a Nordic Semi MCU/RF - should be nice to use, done properly.

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

Who-me wrote:
Seems everyone is now moving (sic) to do the dual pairing of [USB-Debug] and [MCU] which gives the best user experience.

 

"moving" ?! It's been well established as The Way To Go for some years now!!

 

Atmel were late to the party, but have now had their act in gear for a couple of years.

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

Who-me wrote:

Those Eval boards look pricey, but the Discovery STM32F746G-DISCO is impressive.

Sub $50, with Cap touch 4.3" LCD mounted, on a Board not much larger than the LCD.

...

Yeah, indeed that Discovery board is impressive.

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

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

... but low stock though its lead time is short; it's already in mbed.

Atmel's board is in stock; Mouser in-stock starting 2015-Aug-04.


http://www.findchips.com/search/STM32F746G-DISCO

https://developer.mbed.org/platforms/ST-Discovery-F746NG/

http://www.atmel.com/tools/ATSAMV71-XULT.aspx

http://www.findchips.com/search/ATSAMV71-XULT

Edit : added mbed URL.

"Dare to be naïve." - Buckminster Fuller

Last Edited: Tue. Jul 14, 2015 - 04:26 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Rohalamin wrote:
A few days ago, ST launched the first (sic) Cortex-M7

As already noted, it was not the first Cortex-M7.

 

And, in fact, that wasn't even ST's claim - What the ST marketing blurb actually said was,

World’s First ARM Cortex-M7 Core-Based STM32 F7 Series MCU

 

http://www.st.com/web/en/press/p...

So all they're claiming is that it's their first M7; so the "World's first" is just complete nonsense - as nobody but ST is going to be creating ST chips!!

 

 

I find all of these "World's First..." claims to be pretty spurious - but ST must win an award for this one!!

 

surprise

 

Last Edited: Tue. Jul 14, 2015 - 09:40 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

awneil wrote:

So all they're claiming is that it's their first M7; so the "World's first" is just complete nonsense - as nobody but ST is going to be creating ST chips!!

Is this the World's first deconstruction of ST's "World's first" claim?

 

- S

 

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

Rohalamin wrote:
A few days ago (sic), ST launched the first Cortex-M7

Why do you say that?

 

I just got a DM from ST on Twitter pointing out that it is significantly more than just "a few days" ago: it was September 2014 - at which point it clearly was the world's first!!

 

http://www.st.com/web/en/press/p...

 

So why are you posting such old "news" just now?

 

I was caught out because I just received the latest Anglia newsletter which also had the "World's First" claim

 

blush

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

 ... and I guess that was the World's first reconstruction of ST's "World's first" claim?!

 

blush

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

awneil wrote:

 

Why do you say that?

...

Do you want to luck this topic, Andy!?wink

I dunno!cheeky

I couldn't see any Cortex-M7 in some distributors like Mouser a few weeks or months ago.

maybe that would be better to change the tittle of thread to this "The first STM32F7 is available in markets now!" or "...

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

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

Specifically, why did you say, "A few days ago"?

 

What was your basis for that?

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

awneil wrote:

Specifically, why did you say, "A few days ago"?

 

What was your basis for that?

Andy,

STM32F7 MCUs were not availbe at markets untill a few days ago seriously.

 

http://www.eeherald.com/section/new-products/owns201507060001.html

 

or:

 

http://www.stm32duino.com/viewtopic.php?t=392

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

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

awneil wrote:

 ... and I guess that was the World's first reconstruction of ST's "World's first" claim?!

 

.. and here is Atmel's 'World's ...' claim... not to be outdone... they spin for

 

"Atmel Ships World's Highest-Performing ARM Cortex-M7 based MCUs"

http://www10.edacafe.com/nbc/art...

 

but their eval board price really does not line up well at all.

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

One thing I don't like about ST eval boards is that they are not putting SWD connector on them and debuggers contain ST-Link firmware and I was not able to figure out if there is CMSIS-DAP firmware. I have custom tools for CMSIS-DAP and messing with anything else is not in the books.

 

I'm also not a fan of the long though hole connectors.

NOTE: I no longer actively read this forum. Please ask your question on www.eevblog.com/forum if you want my answer.

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

alexru wrote:
One thing I don't like about ST eval boards is that they are not putting SWD connector on them

same could be said of some (many?) Atmel Xplained boards ...

 

sad

 

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

alexru wrote:

One thing I don't like about ST eval boards is that they are not putting SWD connector on them ..

 

Do you mean as IP to attach an external ICE, or as an output so the Eval can Debug onto a user board ?

The better eval's do allow access to the Debug-firmware Chip on the board.

A no-brainer really, as it is just PCB traces and some pads. 

Sadly, not all vendors grasp something this simple. 

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

awneil wrote:
same could be said of some (many?) Atmel Xplained boards ...
Yes, and I'm complaining vigorously about it :) I'm pretty sure that R21 Xpro has this connector because of me.

 

But at least Atmel boards have CMSIS-DAP EDBG. I can live with that.

NOTE: I no longer actively read this forum. Please ask your question on www.eevblog.com/forum if you want my answer.

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

Who-me wrote:
Do you mean as IP to attach an external ICE, or as an output so the Eval can Debug onto a user board ?
External tool to a target device. I'm not that cheap, I can afford a proper debugger, I don't need to turn my dev kits into debuggers.

 

Who-me wrote:
Sadly, not all vendors grasp something this simple. 
Exactly.

 

PS: And especially with CMSIS-DAP, you really just buy one tool for all the parts. And CMSIS-DAP is the reason I'm only considering ARM parts for new designs. Well done, ARM.

NOTE: I no longer actively read this forum. Please ask your question on www.eevblog.com/forum if you want my answer.

Last Edited: Thu. Jul 16, 2015 - 05:49 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Rohalamin wrote:
STM32F7 MCUs were not availbe at markets untill a few days ago

Maybe - but that's not what you said:

 

On Jul 12, 2015 - 01:42 PM, Rohalamin wrote:

A few days ago, ST launched (sic) the first Cortex-M7

 

They launched it - see the press release - back in September 2014.

 

Sure, you couldn't actually buy them at that point - but that is when they were launched.

And that's when the "World's First" claim was made - and valid.

 

As we know, Atmel are also pretty "good" (sic) at launching stuff long before it's actually available to buy ...

Last Edited: Thu. Jul 16, 2015 - 06:23 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The debug connector provision on Atmel Xplained-Pro boards is rather inconsistent:

 

  • Some have a connector to which you can attach an external debug probe;
  • Some have a connector to allow the XPro board to be used as a debug probe;
  • Some have neither!

 

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

awneil wrote:
Some have a connector to which you can attach an external debug probe;
And we see more of those and it is good.

awneil wrote:
Some have a connector to allow the XPro board to be used as a debug probe;
Only D20 Xpro and only because it was the first Xpro board. At a time it was released there was no Atmel-ICE and using JTAGICE3 required an adapter that did not exist. So it made sense back then.

awneil wrote:
Some have neither!
Hopefully they will go away soon. I think all new boards have a connector for the external programmer.

NOTE: I no longer actively read this forum. Please ask your question on www.eevblog.com/forum if you want my answer.

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

alexru wrote:
Only D20 Xpro and only because it was the first Xpro board.

Ah - I see!

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

alexru wrote:

awneil wrote:

Some have a connector to allow the XPro board to be used as a debug probe;

Only D20 Xpro and only because it was the first Xpro board. At a time it was released there was no Atmel-ICE and using JTAGICE3 required an adapter that did not exist. So it made sense back then.

 

It still makes sense today :)

I can understand ICE having higher performance than Debug-Bridges on Eval Boards, but there is no excuse to not bring the Eval board Debug to pins - even non-fitted, empty holes are ok.

 

One of the smarter designs I see is from Nuvoton, who make it very clear you can do both on their NuTiny EVAL-PCB designs, with two sets of holes and break-off slots between the Debug-Bridge and Target Eval device.

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

Who-me wrote:
but there is no excuse to not bring the Eval board Debug to pins - even non-fitted, empty holes are ok.
There is an excuse, actually. You will either need to pay for extra active switching components as on Atmel board with automatic target detection, or you need to place jumpers as on ST boards. Both options take space, time to assemble and cost money. And as everyone has noted, ST is not having any margins on their kits as it is (not that you should make money on the kits by today's standards).

NOTE: I no longer actively read this forum. Please ask your question on www.eevblog.com/forum if you want my answer.

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

Interesting philosophical discussion here as to what constitutes "World's first". I am hereby announcing a teleportation machine that will soon be available. OK "soon" may actually mean 200..300 years but I claim this announcement is "the world'd first"!

 

BTW I believe I am responsible for a world's first cheeky. We produced the "world first" first pen driven handheld with hand writing recognition in the early 1990's. Sure the Apple Newton was announced before our product (we were in the habit of not announcing until it was real) but we had our device in the stores 3 months before Apple actually delivered and I claim that means we were first.

 

So I would say that it's general availability not vapourware that would constitute a first in anything.

Last Edited: Thu. Jul 16, 2015 - 09:10 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I always laugh at the "worlds first cellphone call" and similar nonsense.

 

Clearly, the engineers working on the stuff would have been making (test) calls long before the big public "worlds first" demo...

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

alexru wrote:

 

Who-me wrote:

but there is no excuse to not bring the Eval board Debug to pins - even non-fitted, empty holes are ok.

There is an excuse, actually. You will either need to pay for extra active switching components as on Atmel board with automatic target detection, or you need to place jumpers as on ST boards. Both options take space, time to assemble and cost money. And as everyone has noted, ST is not having any margins on their kits as it is (not that you should make money on the kits by today's standards).

 

Nope. Take a close look at the Nuvoton design - they manage without any of what you claim.

A few examples, with nicely zooming images here :

http://www.digikey.com/product-s...

 

 

Last Edited: Thu. Jul 16, 2015 - 09:31 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:

 

Rohalamin wrote:

STM32F7 MCUs were not availbe at markets untill a few days ago

 

Maybe - but that's not what you said:

 

 

On Jul 12, 2015 - 01:42 PM, Rohalamin wrote:

 

A few days ago, ST launched (sic) the first Cortex-M7

 

 

They launched it - see the press release - back in September 2014.

 

Sure, you couldn't actually buy them at that point - but that is when they were launched.

And that's when the "World's First" claim was made - and valid.

 

Alright, you won. your majesty!cheekydevil

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

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

clawson wrote:

Interesting philosophical discussion here as to what constitutes "World's first". I am hereby announcing a teleportation machine that will soon be available. OK "soon" may actually mean 200..300 years but I claim this announcement is "the world'd first"!

 

BTW I believe I am responsible for a world's first cheeky. We produced the "world first" first pen driven handheld with hand writing recognition in the early 1990's. Sure the Apple Newton was announced before our product (we were in the habit of not announcing until it was real) but we had our device in the stores 3 months before Apple actually delivered and I claim that means we were first.

 

So I would say that it's general availability not vapourware that would constitute a first in anything.

I was responsible for the world's first coin operated public access Prestel terminal, and the world's first computer game cassette duplicating/vending machine. I believe they were both also the last of their kind, so a double distinction!

 

Quebracho seems to be the hardest wood.

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

Who-me wrote:
Nope. Take a close look at the Nuvoton design - they manage without any of what you claim.
Ok, break away debugger concept also works in some cases. NXP does this fro their Xpresso dev boards. But now you are limited in your board design.

 

The width of the Xplaube Pro board is over 2.5", breaking that would not be easy. It also carries a dozen of traces to the debugger chip (some extra serial interfaces in addition to the programming pins), so breakaway tabs must be quite wide. So there are limitations to that concept.

 

Also, on Xpro boards for targets with USB, USB connector is located on the same side as the debugger connector and the rest of the sides are taken by the breakout pin-headers. This will not work in your concept.

NOTE: I no longer actively read this forum. Please ask your question on www.eevblog.com/forum if you want my answer.

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

Oh my ... you should see this:

 

http://www.rogerclark.net/gd32f103-a-stm32f103-on-steroids/

 

and especially this one:

 

https://olimex.wordpress.com/tag/gd32f103/

 

This part is soo interesting to me!surprise:

 

Dear Tsvetan,

Those GD32xxx devices are effectively a kind of clones of our STM32, but it's a pure piracy: no agreements of any kind between ST and Giga Device, no license... nothing.
ST legal people are in charge of this problem.

 

Have you were aware of it?

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

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

Rohalamin wrote:

and especially this one:

https://olimex.wordpress.com/tag/gd32f103/

 

This part is soo interesting to me!surprise:

 

 

"So it appears that the GD32 has been designed by Gigadevices to be pin and generally register compatible with the STM32)"

 

Interesting, the purported ST claims of 'piracy' seem a little creative - certainly they  cannot brand it STm, or claim ST endorsement, but the CORE is not ST's property, that license is with ARM.

 

Likewise, numbers like f103 are not ST property, (but the full part code may be copyright).

 

Also, vendors have been making "compatible" 8051's for decades, so the idea of an MCU that is somewhat Pin and register compatible, is already widely  proven in the MCU market place.

 

It just may be a small surprise to some ARM licensees.

 

The GD32F103 sounds a superset design, not a direct copy, and the Flash speed claims suggest it is a generation ahead of ST.

Some of the controls and specs read like a first-release part, especially around USB speed settings & PCB crystals.

 

The GD32F20x parts look interesting, 120MHz and up to 256k SRAM

 

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

Moved to ARM forums. - JGM

If you want a career with a known path - become an undertaker. Dead people don't sue! - Kartman

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB user

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

Apart from the f103 compatibility, what is this part giving us that we can't get elsewhere? 256k ram? Kinetis k64 is one device that comes to mind. 120MHz? K64 again and sym32f4. Atmel have 300MHz with their m7 parts. Where's the excitement here?