I am going to buy a new toradex product. This product will only act as a hub, I will use it to transfer the incoming data to another computer. Also the most important thing is that, i want to develop my code on .NET Core framework and deploy to toradex board. I do not prefer c/c++ or other language. I want to only .net. Which toradex product would be the best?
If you’re only requirements is to pass on data then I’d imagine most of our products would suffice. If you’re also doing some kind of data processing or such on the board itself then we’d have to factor in CPU compute power as well. Another aspect to consider is since you want this to act as a hub, what kind of connectivity are you looking for? Ethernet? WiFi? I mention this since some of our boards have built-in WiFi and some don’t.
But in general most of the compatible modules listed here sound suitable: Torizon | Toradex Developer Center
As for your .NET Core requirement, you should be able to develop with .NET Core no matter the module chosen.
I will transmit it after a converting and masking operations. As you say, i’m going to choose optimal CPU. The data bus will be serial port, maybe ethernet in future, but firstly it’s serial port now. I need 3 serial port, there are at least 3 serial port for each card, right ?
Just to be clear are you talking about serial UART like this example here: How to Use UART with Torizon Visual Studio Code Extension (.NET Core) | Toradex Developer Center
Or are you referring to the SPI interface?
Either way I’d suggest looking at the product page here: Computer on Module (CoM) - System on Module (SoM) - Arm
If you browse each family of modules there will be a table showing how many interfaces are available per each module. Though I believe you are correct that most/all our modules should have enough serial interfaces (depending if you’re talking about UART or SPI).
I am talking about UART, not SPI. Also, i revived your comparison page, between torizon and yocto based operating system (Operation Systems) The memory optimization item made me think a little bit, why is memory optimization lower on torizon and what are the disadvantages of this situation for me ?
why is memory optimization lower on
torizon and what are the disadvantages
of this situation for me ?
This is due to how containers inherently work. With Torizon we except customer to develop and deploy their applications in containers. A container is a sandbox (like a light VM) that has it’s own set of libraries separate from the base OS. This can lead to situations where there’s duplicate libraries loaded and therefore more memory is used up. There is also the memory overhead of having the container system in the first place. Generally speaking this total overhead is relatively minor compared to the memory needed by most applications. The only real disadvantage is if you’re right on the edge of not having enough memory and the overhead is what’s putting you over that line.