I would say that this is not so much the amount of data as the number of requests . In other words, if you have 100,000 connections that transmit 100 bytes of data, you will have more tcp / ip overhead than if you had 10 connections of 100K each.
This does not mean that there are no associated costs associated with data transfer through tcp / ip against named pipes. There is. But as a rule, I would say that the decision about what you are going to use should do more with the architecture of your system than worry about overhead.
If you are going to transfer data between physical servers, you should go with tcp / ip; named pipes are not an option. If you transfer data between processes on the same server, then the named channels are certainly the best performers.
One of the reasons why you can go with tcp / ip when you are on the same physical server is the likelihood that you will be splitting processes into physical servers at some point in the future.
To answer your question: if you do not transfer a lot of data and often do not, you probably will not notice the tcp / ip overhead if two endpoints are on the same physical machine.
Hth,
James