Mcu32

$50 dev board : https://www.microcenter.com/product/665366/arduino-nicla-voice ($53 DigiKey) Or direct ARDUINO.cc Key Features Component Type: Development Board Components: Mainboards Processor: Syntiant NDP120 Neural Decision Processor Clock...

A dev board / module is available: https://www.digikey.de/de/products/detail/syntiant-corp/SYNTIANT-TINYML/15293343 (original with mouse bites ; ) Nowadays, not even the Chinese dare to do that anymore.

5 year ago Article ends with "Volume production is scheduled for summer 2021," The link is wrong, but they have that NDP120 and newer: https://www.syntiant.com/hardware

Going to jump in here again about running on multi-core processes and using multi-thread setting. Reason -- was poking around the web and asking AI :) CoreMark Multi Core Reporting Requirement According to the official EEMBC CoreMark Run and...

Right now only have a Teensy that I can run it on - don't have an RP dual core. But its on my to do list which seems to be getting longer and longer. :)

It does not matter what @whitebengal meant or not. I suggest that any thread on Adafruit vs Sparkfun (and this one) is closed immediately. If there is a need to enter into a Teensy development due to non PJRC events, than this should be named...

That looks pretty good and doable. Are there any volunteers? :)

https://www.hackster.io/news/syntiant-s-ndp120-deep-learning-chip-offers-a-25x-performance-boost-for-always-on-applications-ad648bc51cd4 Syntiant looks very good and is modern. A board with one of their chips (for offline wakeword-detection for...

Ah, okay, that sheds some light on the matter! I would first use a logic analyzer to obtain the actual numbers.

It was clear to me from the start that this would not be merged. No, it would be too tedious to get the code back into a form that you would merge. Of course, you can merge it if you want. Also, the numbers in the readme... I updated Teensy 4 and...

You can find a similar table in the datasheet. I'd trust that more :)

Data ist valid "Min 100ns". - Just wait 160ns-100ns = 60ns after setting WR->0, then read. And isn't WR an input to the chip? Am i'm reading the diagram that wrong? In a German forum, you would have been criticized if you didn't think it...

Additionally, while the specific context matters, I believe phases "version of Teensy" or "variant of Teensy" for products not from PJRC are extremely problematic.

Please upload clear photos of the top and bottom side.

If that gets merged in, mine (+ESP) will follow.

I've added Coremarks for Teensy 4 / Teensy 3.6 . both are a lot better now (newer compiler?) (see github) Q: Is -O2 or -O3 still the best? T4 seems faster with -O2 , T36 faster with -O3. Should i add some switches/flags?

I'd reduce this to #1 - but keep in mind, that even a new "Teensy 5" would require code changes :)

Cache size also really matters. An 8K instruction cache is the reason why Teensy 3.6 scores 66% faster than Teensy 3.5, even though is has the same Cortex-M4 processor running at only 50% higher clock speed. If instruction and data caches are...

The other approach would be to fix measurement time and each core should be allowed to do as many iterations as it can within given time. Then the sum of number of iterations done on each core should be used to calculate total performance. This...

Yes, you always have to be aware of what is being measured and how in order to obtain a reasonably reliable answer. Coremark also has many critics. For example, “floats” and “doubles” are not measured (so you don't see an existing FPU in the...

FWIW, not long after Teensy 4.0's release, the ARM folks were unhappy with me for publishing a score less than 3000.

That depends on the definition and how individual manufacturers define the safety margin. It could be that some want a large safety margin and others don't. “Overclocking” is a very vague term. We can't know that. And yes, I agree, I would say...

Yes, the version of the compiler has a major impact. Optimizations are constantly being worked on.

Good question.

I'm wondering why all the "RP2350 Due Core" results end in .00?

Let me just add that there are indeed MCUs that have I/Os with different electrical specifications.

To answer the original question about the maximum I/O pin load of the Teensy 4.x: in this case, the information is actually documented. PJRC provides the relevant specifications here: https://www.pjrc.com/teensy/techspecs.html While detailed...

more coremark measurements is always great. would love to see the new ESP32-E22 measured - its 2x 500mhz risc v cores and dual-band wifi...

Feel free to use my repo (link above). It should work just with adding the board to the platformio.ini. Would be nice to add your results to the readme.

There is a #define MULTITHREAD that must be defined via platformio build_flags to allow multithreading. For two threads: build_flags= -DMULTITHREAD=2

Thank you Tomas, i've added your Benchmark results to the readme, and made the detection code more robust. Indeed, build_flags= -DMULTITHREAD=2 (or 1) in platformio.ini is enough.

Lady added the Dual-Core Code for thr RP2350, an I took her code and added support for ESP32/freeRTOS. Good idea—I'll make some adjustments to the code to improve automatic architecture detection.

Are you using Platformio by chance? If you have ESP32-S3 code/project I can run on newest ESP32-P4 and post the results.

Platformio, yes (see link above) . You have to add the P4 yourself - can't test it

It appears that the RP2350 ("200MHz Overclock") is only marginally faster than the older EP32 model. I will acquire a newer one (P4) and add the results.

In the other thread, Coremark was brought up again. Lady Ada seems to have expanded it for the RP2350 so that it takes both cores into account. I have further expanded the code and it now also runs with freeRTOS – i.e., “real” preemptive...

Sad story. I fear that this gives less sellers in the EU and thus more difficult to get a teensy with some other stuff. A Teensy alone and thus with shipping, will be a lot more expensive.

Yes, I agree, this conversation needs to migrate to Adafruit's forum. We've allowed it to go on long enough here. We're far past unproductive and now a few solid days into disrupting this community. @ptorrone - Please create a thread on your...

Discussed over the years ... IIRC ... Given the PJRC bootloader it may be compatible, or work with TeensyDuino ... but naming and such cannot violate the PJRC naming rights of the Teensy branding, or suggest it is a PJRC sanctioned item. Thanks...

.... Or i.mx rt 1170 + MIPI + more / other pins broken out.

My assumption is that Sparkfun purchased exclusive licensing rights to produce and sell Teensy boards, and Paul would be violating the royalty contract he signed by doing what you suggest. Honestly though if you’re going through the trouble of...

I call myself a hobbyist (even though I have EE degree) and bought plenty of Teensy 4.x and going to buy way more ! @ptorrone - initially I wasn't understanding what this is all about, but after what you have said and done it become clear that...

Trying to understand this - the third party would want a reseller discount so they can make a profit, and then Adafruit would also want a reseller discount so they can make a profit. I struggle to see how Adafruit would then be able to remain...

Not sure why I'm posting... I've bought things from every company involved and generally have had nothing but positive experiences with all of them. My first PCB design used a Teensy 2.0. Does feel a bit like a kid wondering why mommy and daddy...

It sounds like Sparkfun wants to keep Teensy retail margin to themselves & not allow Adafruit to distribute Teensy profitably at the retail price - basically not selling them to their competitor at wholesale pricing. I think the reason this...

Yes, that could be a problem. And no one knows how long Sparkfun will continue to produce Teensy. I fear that if difficulties arise—such as renewed delivery problems with certain parts (which don't even have to affect the CPU)—it will disappear...

Just for fun, Coremark Results for ESP32-C3 (160 MHZ, RISCV with FreeRTOS in background): 2K performance run parameters for coremark. CoreMark Size : 666 Total ticks : 14644 Total time (secs): 14.64 Iterations/Sec : 409.72 Iterations...

That's probably the main reason. Adafruit is competition for Sparkfun. Sparkfun isn't covering itself in glory in this matter either.

I agree this probably isn't productive for anyone, but for now I'm going to allow this thread to continue. However, I'm absolutely drawing the line at creating more duplicate threads promoting Adafruit's new competitive product or RP2350...

Do not create more threads to advertise RP2350 or products unrelated to Teensy.