Suggest next Teensy with Cortex M7

Status
Not open for further replies.
En français:
bonjour Paul,
J'aimerai savoir si une "teensy 4" était toujours envisager

Google trad:
good-morning Paul,
I would like to know if a "teensy 4" was always considering
 
We are about exceed the processing power of a rpi or pi-zero without the madness of a huge OS, and we need to stay calm in our anticipation? An entirely unreasonable request.
 
We are about exceed the processing power of a rpi or pi-zero without the madness of a huge OS, and we need to stay calm in our anticipation? An entirely unreasonable request.

I would be very interested to know what you would do with such a fast processor what you cannot do with a Teensy 3.6? ;-)

For my applications, a real time software defined radio, the Teensy 3.6. only barely reaches its limits when overclocking the audio codec with 192ksps . . .

So you would have to have very good reasons to call for a faster Teensy . . .

Have fun with the Teensy, bonne nuit,

Frank DD4WH
 
There's a lot you can do with audio effects processing with a T3.6 which is great!

I also like to do a lot of subband processing, where the audio signal is split into multiple bands using high stage-count IIR biquads which consume a lot of processor time (even using the CMSIS library).

With real-time audio processing, you ALWAYS need more processing power.:)
 
Next step after audio is video.. ;) So I really hope there will be an efficient interface that can be used for video or other fast data streams. At least a 16 BIT (better more pins for handshaking) Port.. it could be used for a large number of other interfaces, too. Memory for example.
And I hope, VGA (or better..) will still work.
 
en :
Why do I want a teensy 4? Not only for speed, there is the number of pins, and I'm not talking about the one located at the end of the teensy, I'm talking about real pins. Then there is the audio part, to increase the quality and the number of channels was cool. In addition I am limited by its amount of available memory, there is no bus memory available to increase the number of RAM MB.

fra:
Pourquoi je veux un teensy 4? Bah pas que pour la vitesse, il y a le nombre de broches, et je ne parle pas de celle situer endesout de la teensy, je parle de vrais broches. En suite il y a la partie audio, augmenter la qualiter et le nobre de cannaux serrait cool. En suite je suis limiter pas sa quantitr de memoire vive disponible, il n'y a aucun bus memoire de disponible pour agmenter le nobre de MB de RAM.
 
Before asking for more pins and memory, do you already have finished and working projects to show here, pushing the limits of the Teensy 3.6?
 
Eng:
yes precisely, I can not use the pin located below the board

Fra:
Justement, je n'arrive pas a utiliser les pins situer en dessous de la carte
 
pourquoi pas? il y a des adaptateur pour utiliser ces pins, ou vous pouvez les souder aussi comme Kurt a fait pendant le Beta
moi je fait meme que Kurt mais pour les breadboard j'utilise des extensions, tres facile :)
 
Although I largely prefer speaking French, too, we should respect the international character of these forums and the fact that PJRC is a company based in the US and thus make an effort to write exclusively in English. I'm aware of the fact that many French and Italian people are victims of an education system which does not take globalization into account, but that's not an excuse, it should be a challenge!

I'm not sure about the intentions of Armadafg asking for more pins, more RAM and whatever, but up to now, we haven't yet seen successfully developed and working projects realized with the actual Teensys, so we can't actually judge if it's simply his play instinct or if he managed really to build something which can't be fully done with a Teensy 3.6. It's easy to ask for a SUV, but one should have proven before that one made already a few 10000 miles on a bicycle... ;)
 
Bicycles are not seasonal, so one must upgrade to an SUV for all seasons (that and more power, less footwork)
:D
 
Last edited:
eng :
I have actually considered this cards:
https://www.tindie.com/products/loglow/teensy-3536-breakout-revision-a-standard/
She is too expensive and too big. I use 2 teensy and it is 120 € + shipping costs instead of 80 €, and for a portable project.

fra:
Je les ai effectivement envisagé cet cartes :
https://www.tindie.com/products/loglow/teensy-3536-breakout-revision-a-standard/
Elle son trop cher et trop grosse. J'utilise 2 teensy et ca fait 120€ + frais port au lieu de 80€,et pour un projet portable.
 
eng :
I understand, in terms of French I use a translator, but sometimes it can make mistakes, I try to correct but it is not impossible that there are still problems of translation. That's why I post the original version for those who want to translate it with another translator or directly read it in french

fra :
je comprends, en ce qui concerne en francais j'utilise un traducteur, sauf que parfois il fait un peut n'importe quoi, j'essaie de corriger mais il n'y pas impossible qui y ai encore des problème de traduction. C'est pourquoi j’affiche la version original pour ceux qui voudront le traduire avec on autre traducteur ou directement le lire en francais
 
Since you didn‘t tell details about your project, nobody can help you with alternative solutions or thoughts. A need for many pins can exist if there are lots of controls like buttons, encoders, or potentiometers, or if there is a need for driving peripherals in parallel mode like ADCs, DACs, or displays, but there are most times smart solutions to get around that, i.e. by serializing several things with shift registers or multiplexers, or by using a Teensy LC as a sub processor and port extender. The world of embedded computing has its limits and constraints, but there is most times a creative way around. But without knowing more about the project specifications, no qualified advice can be given. Just increasing the hardware resources looks sometimes like the easiest way, but it not always the most efficient. Think of booting Windows and launching MS Word on a modern 8 core processor with 8GB RAM anda 1TB harddisk. It takes around ten times longer than booting MS-DOS 2.2 and launching Word Star from a 360kB floppy disk on a 8086 IBM PC with 640kB RAM from 1985...
 
eng :
Regarding my project, I really hope that will see the light of day, but I'm currently with the last step before I make it public: the 3d printer. It will be at least 2 years that I oriented my project with teensy 3.6 and FT80x, FT81x and soon I hope BT81X .

Fra:
En ce concerne mon projet, j'espère vraiment qui verra le jour, mais je me bas actuellement avec la dernière étape avent de le rendre publique : l'imprimante 3d. Ça va faire au moins 2 ans que j'ai orienté mon projet au toure des teensy et des FT80x, FT81x et bientôt BT81X, en fin j'espère.
 
eng :
I totally agree and I love retro computing. Regarding my project I am obliged to use 2 teensy 3.6 for the moment and I do not really like it. it consumes more, it is more expensive and super painful to devloper. If I want to use a single card I would need four buses spi: one for the EVE (Embedded Video Engine) + sdcard the other for the teensy audio, the 3rd for the ram SPI and the last one that will be used for the following. Moreover it would mean that you have to move teensy audio on the bus SPI2 and for the moment I do not have the level for that. in addition to the four SPI buses it would take about 15-20 pins available in addition to the 2 USB ports

fra :
Je suis tout à fait d'accord et j'adore le retro computing. En ce qui concerne mon projet je suis obligé d'utiliser 2 teensy 3.6 pour le moment et je n'aime pas vraiment ça. ca consomme plus , c'est plus cher et super pénible à devloper. Si je veux utiliser une seule carte il me faudrait quatre bus spi : un pour l'EVE (Embedded Video Engine)+sdcard l'autre pour les teensy audio, le 3e pour la ram SPI et le dernier qui servira pour la suit. De plus cela voudrait dire qu'il faut déplacer les teensy audio sur le bus SPI2 et pour le moment je n'ai pas le niveau pour ça. en plus des quatre bus SPI il faudrait environ 15-20 broches de disponibles en plus des 2 ports USB
 
I must admit, for my "car project", which i neglected for awhile, Teensy 3.5 didn't have enough pins. I ended up using a bank of 8x SPI MCP23S17 port expanders on a single CS SPI2 bus, as well as another 6 I2C MCP23017 on a single Wire1 bus, in order for it to be what it is for today. The display refreshes of a dual UART displays were lacking performance running side by side for gauge uses, thats when I started using TeensyThreads and its been like that since working great :)

So yeah, I pretty much surpassed the pin limits of a 3.5, but knowing that, I ran most lines off the chips and not the teensy, so only limited wiring goes to the T3.5 for bus usage. This (although unlikely for current project), allowed future ease of MCU replacements/upgrades
 
Message #144 talks of needed 4 separate SPI, one of them for audio. That doesn't make much sense to me. Maybe it's the French to English translation?
 
eng:
yes sorry you're right it's the i2s not the spi, but the teensy audio uses the same pins as the bus spi1 if I'm wrong

fra :
oui désolé tu a raison c'est le i2s pas le spi, cela dit le teensy audio utilise les meme pins que le bus spi1 si je me trompe pas
 
Status
Not open for further replies.
Back
Top