X-Surf 100 - Best TCP/IP for Speed?

[danbeaver]

Using RoadshowPPC and the X-Surf 100 in OS4.1 shows not more than 24% CPU utilization with the overhead browser (IBrowse) or 18% with the overhead of AmiFTP transfers. Like all of the referenced numbers in this thread, this is Anecdotal and has no statistical validity. Test Bed numbers apply only to the testing hardware, software and network at hand and thus have no more than a casual relationship to Real World numbers.  I feel you have to ignore the numbers unless you perform quite extensive real world testing and then calculate a meta-analysis of those results to get a generalized pictured of the throughput "in the wild."  

My conclusion is that it "feels" fast, in the same ballpark (but not over the fence) as a USB 10/100/1000 dongle off a Deneb in Z3 PIO under OS4.1.

[danbeaver]

OK, installed the x-surf 100.

Using Miami right now and it is barely faster than the Ariandne.

Ariandne with MNI gives 155,000 cps and 25% cpu.
X-Surf with SANA2 gives 170,000 cps and 15% cpu.

Oh well.  Hopefully when I figure out how to install AmiTCP with Easynet, the results will double...

The whole point of EasyNet is the ease of installation, given a few router parameters; although Roadshow lacks a GUI, it is pretty much a dead simple install: install the software, copy one file (the NIC card settings from storage to device), make sure the correct driver is in the Devs:Networks folder, and reboot.

[danbeaver]

This just tells me that AmigaOS need a new network driver infrastructure..

It already has that in OS 4.1.

[danbeaver]

The lpr.device works with every stack but AmiTCP v3.  In the case of Roadshow, it works great and switching printers is as simple as changing the ENV variables.

[danbeaver]

A base and random file transfer rate in cps is not a fair way to evaluate NIC card transfer speed, one needs to control all the hundreds of other variable factors that impact that number.  I would go with the card's designer recommendations that AmiTCP with his tweaks gives the best performance.  His testing was in a controlled environment setting where he limited the random variables.

[danbeaver]

The catch is that real life usage isn't laboratory conditions..

In that case, numbers mean nothing and you should go with intuition on TCP stack decisions and not bother quoting any numbers.

[danbeaver]

It means one should try a variarity of situations to test. Even the Monty carlo method comes to mind.

And your validity comes from where?  A guess? A feeling?  

You can't account for the hundreds of variables that affect performance that way.