|BUT it has everything to do with download times. A slower PC will take longer to download something than a faster PC will. |
The bandwidth rate is controlled by the Network Interface. If your NIC is 10 Mbps, you can download to a maximum of 10 Mbps. If 100 Mbps, then a max of 100. If 1000 Mbps, then a max of 1000 Mbps.
This is regardless of processor speed. The ONLY thing that might not be as "fast" would be the rate the CPU recombines the packets as they're coming in. While I don't have any statistic on hand, I'd wager even the slowest Pentium IV class CPU can recombine faster than a 1000 Mbps processor can download.
4. OK CurtR, referencing download times here again, I downloaded 23 Mb Office 97 SP2 in IE 7 - it took me 1 minute, 7 seconds. The same download, same PC in Chrome took 31 seconds.
Did it ever occur to you that the difference could have been cause by:
- How busy the server you were downloading from was in each instance?
- How many packets had to be retransmitted during the download process do to contention based issues (ie: packet collisions causing resends)?
I realize you don't know very much about networking and I really wish you would listen to me and stop misinforming others who know less than you do. You'll only confuse them.
RE: DSL: Only meant to imply that DSL isn't shared the same way that Cable is and speeds are more constant is all I meant.
Wrong again. They are "shared" exactly the same way. A network is a network is a network........period, full stop. Every provider, be they ADSL or cable, have segments. Each segment has "X" number of users (each one being a "home" connection). The segments will be separated by subnets (usually using a Private Class addressing scheme). When one segment becomes saturated (as in, has the maximum number of connections it can support while still providing decent bandwidth to all connections) they then start another segment. Each segment is "shared" between the users connected to it. It's worth noting at this point, a segment could theoretically be comprised of up to 254 connections (the maximum number of netowrk IP's on a segment).
Actually, you wouldn't want them running at the same time to have it be a true test. You would want to run one from a particular connection, then run the other on that same connection as the first one, preferably off hours with less traffic.
Actually, I know what of I speak and I was correct at the outset and you are once again....wrong.
Try this on a LAN while downloading from one PC to two others. Even if your LAN is only 100 Mbps, you should still be able to download a 1 GB file from PC A to PC's B and C at the same time and achieve close to the same download rate and have the download take approximately the same time to download.
The same is mostly true if downloading from the internet. BUT, and I stress the word but, there are variables that can adversely affect one of the two download streams without bothering the other. See my answer to #4 above for two of the "variables" than can affect download rates.
I don't have the time or energy to educate you on networking so you're either going to have to do some reading/research on your own, or you're just going to have to trust that a guy with 15+ years working in industry, with the last 5 solely in Enterprise Level Networking (including LAN, MAN, WAN and wireless) actually knows what he's talking about.
It matters not how straight the gate,
How charged with punishments the scroll,
I am the master of my fate;
I am the captain of my soul.