Search:

Hello,

i agree with @KlausST. If you are looking for theoretical information see links to these algorithms:

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

Hello,

what datatype do you mean ?

Regards

Hello,

this is unfortunately true and I am afraid of such taxes :|

Thanks for opinion and regards

Hello @KlausST,

now I can say you more about assumptions to my design. I find in Xilinx Vivado free IPCore with FPU module implementation. This implementation is using DSP slices and I suspect it...

"Really huge" means in this case up to few millions. The first dimension of tensor is usually the number of samples used to training "artificial neural network". of course vector coprocessor will be...

Hello Klaus,

related to 1) Yes you are right. There be many instances of 16 bit floating-point units (only two operations: multiplying and adding). Them will be used in module for making...

I am working on some kind of vector floating-point coprocessor. I would like to use it for multiplying and adding matrixes with floating-point numbers. Yes I am aware that FPU for floating-point...

Hello,

I am looking for best way (in sense of speed and simplicity) how to connect STM32 MCU to FPGA board (Artix-7). The comunication must be two-directional. I am experimenting with some kind of...

Hello,

could you please expand on this? What is wrong with them to stop you from buying much advanced products?

Thanks and Regards

Hello Guys,

I would like to ask for opinion. For some time I am looking for cheap FPGA board with big number of "logic cells" and "DSP blocks". I do not need very high clock frequency and super...

Hello,

it is impossible to answert such question. If you don't specify type of operations - I am citating:


You list three diferent families of FPGA's from two different vendors (them differ...

Hello,

as far as I understood you correctly - you waana design a matrix multiplier (2 dimensional arrays). You don't give many important assumptuions related to your design. We don't know how big...

Hello,

I am not expert in this subject, just find some information on the web. Maybe it could be anchor point for your research ;)

See this document:
...

Hello,

you can transfer file by using "WinScp "application (it uses ssh connection to the Zynq board - which you used with TeraTerm). Here is link to their website:
...

Hello,

I would like to relate to your opinions:

1) It is possible (in real environment) to achieve bit rate 12 Mbit/s on "CAN FD". I agree that it is not possible for such long distance.

2)...

Hello,
yes you are right :). The assumption to emulate more complicated CPU on smaller and simpler one is bad choice. But you can see how things are done in such examples.

Best regards

Hello,

here you are few links to existing 8086 CPU emulators:

https://github.com/TheFox/i8086emu

https://codegolf.stackexchange.com/questions/4732/emulate-an-intel-8086-cpu
...

Hello,

you can use STM32 MCUs in arduino IDE. See these links:

https://github.com/stm32duino/wiki/wiki/Getting-Started
...

Hello,

I found this project:

https://opencores.org/projects/keras_to_fpga

, but this doesn't look to be finished/completed, and rather is only runtime for running ANN from Keras not tool for...

Hello,

as in title "Has anybody tried to run Keras or "Tensor flow" frameworks on FPGAs. I don't ask for runing trained ANN on FPGAs, but I rather asking about running neural networks in training...

Hello,

what about "CAN FD" (the newest CAN standard). It has max. data transfer rate close to your requirements (8Mbps). I am using "CAN FD" in my newest project based on STM32MP157A MCU and it...

Hello Guys,

I am just reading article from EEJournal titled "High-End FPGA Showdown – Part 1 A Tale of Three Cities". Here is link to this article:
...

Hello Klaus,

I don't want to argue width your arguments - I agree with all of them, but on the contrary why not?

What argumenst are against to use floating-point: occupation of FPGA resources...

Hello,

here is next part of this post:

To make sine function generator in VHDL we need two components:

ROM memory width sinus table for one quarter
Compnent which is changing angle 10 bit...

Hello,

I need for resolving Forward/Inverse kinematics problem (in context for robotic arm) to calculate trigonometric functions values. I am going to use in every arm joint quadrature encoders...