Ethernet communication speed?

Hi guys,

I am trying to compute what is the communication speed of Arduino Ethernet? The Ethernet board will be connected to PC via Ethernet port. I will send a data from PC and Arduino will return it to me and with that I will know how fast it is communicating. For my project, serial communication is too slow, that's why I am trying to do this. But, I am quite not sure how to do it? Any help....

The shield and the Arduino are communicating at about 4 Mbit/s. I guess in reality you could reach about 1 Mbit/s network traffic giving you roughly 100 kB/s data transfer rate, about 10 times what you get by a serial connection.

I was actually talking about this board and not the shield. I guess, there will not be much difference.

http://arduino.cc/en/Main/ArduinoBoardEthernet/

Thanks

And, are there any ways I can check it by myself. Using labview or something similar. I am a noob and I couldn't find related info's on the forum

Thanks for your help.

The Arduino Ethernet is basically an UNO with the Ethernet shield integrated on the board and the USB part removed. So there is no difference regarding your question.

You can check the speed yourself by programming a simple sketch that accepts a connection and returns every byte it receives. Then you have to write a benchmarking application on the PC side and let the two communicate. I don't know labview so I cannot give you hints there.

Thanks you, but I am not sure about the programming part that you mentioned.

Can you please tell me where I can find some sample code or what should I search.

Thanks

To my knowledge there is not existing code that does benchmark the speed of the Arduino Ethernet Shield. Theoretically you could just use the webserver example of the Ethernet shield and use web server benchmarking programs but you'll probably would get other results if you would do the same with direct socket programming.

If you wanna do the above google for "web server benchmark". You'll get lots of results, you have to decide which one is best for your case (there is no benchmark that can do that "in general").