Having watched this thread with a growing sense of anger (I hate it when corporations/businesses deceive in this fashion), I was wondering: has anyone contacted ZDNet and had them justify this benchmark? I would love to hear the response from the "engineers" who compiled this information. | -----Original Message----- | From: linux-net-owner@vger.kernel.org | [mailto:linux-net-owner@vger.kernel.org]On Behalf Of Walter Zimmer | Sent: Thursday, April 05, 2001 4:43 AM | To: Lee Chin | Cc: linux-net@vger.kernel.org | Subject: Re: How do we interpret this ZDNET benchmark? | | | Hi! | | Lee Chin schrieb: | > 100000000 (ie. 100 Mb) / 1 Million = ~ 13 bytes | Some further math: | We can serve a maximum of 200,000,000 bits/second | | One frame takes at least 64 bytes = 512 bits | preamble = 8 bytes = 64 Bits | between frames there are 96 bits interframe gap | | So the smallest unit we can serve has 672 bits. | | If we divide 200,000,000 bits/second by 672 bits/packet we get | a maximum of 297619 packets per second. | | Right ? | | Off by factor 4. Looks like someone took a specific benchmark | (or not even that :) to let one product shine above another. | | Cheers, | Walter - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org
