The breakdown into pages that you get when you click the number of posts (or the little d) is by a specific number of posts. The "page" loaded may vary a little depending on whether posts in the range are long or short, but averages about 200K per page.

MOAB is currently at 542 posts, so opening the entire thread in a single click would run to something like 110MB (a rough estimate).

In order to receive the download, your computer needs >110MB of free space in its Temporary Internet Files folder. Since most recent Windows systems default to use no more than ten percent of the drive free space for the Temp folder, you would need at least something more than 1GB of free space on your system drive, just to open the "page," if nothing else is already in the Temp space.

Even if you had an "exclusive connection" a download is sent in "packets" of a few KB. Your computer asks the server to start, and the server sends the first packet. When your computer has received the first packet, it sends a confirmation to the server and the next one is sent, or if the packet was not confirmed as "successful" the same packet may be re-sent until a successful receipt is confirmed.

If others are using the same server, each packet request may be "queued" to wait until the server can accept it, which may cause some slowdown in the download, and once your computer has sent the request for the next packet it's "waiting for a response" which in any program gives that program a "priority" that prevents you from closing the program.

If someone chose to open in pages and save each page, the disk space required would be about the same ~100MB (plus a bit of cluster loss) if they noticed that all the pages have the same default title and edited the page names to add a unique page number to each save. If you don't "rename" each page, each successive page will write over the previous one since the filename is the same, and when you get done (after a very long time) all you'll have is the last page.

(The same "same name" problem is common, notably when one downloads decisions from some US Supreme Court archives where all downloads are named "file.pdf.")

Even with a high speed connection, a 100MB download may take a few minutes up to a half hour or so. Servers intended for downloading large files may give "priority in the queue" to computers that are in the process of a download, but at mudcat each of the 2 or 3 thousand packet requests probably will go to the end of the line. If any one of the requests fails, your computer will be "locked up" because it's waiting for the server to reply^*, and you will usually be unable to close the browser. Since, at mudcat, each packet request probably goes to the end of the line, if traffic is heavy mudcat quite probably has a "time out" on the queu (in case someone gets tired of waiting and goes away), and the request may be dropped and will never be answered; but even at best it will take a very long time to complete a download of this size. The opening of the page may not have failed - it may still be trying to finish loading.

^* A similar result happens frequently at MSNBC News when one of the dozen or so remote servers that feeds one of the hundred or so "embedded ads" on the home page doesn't respond. The loading of the page is done, but the browser is "waiting for a reply" from the server that provides the last ad, and the browser is locked so that you can't scroll, click, or close the browser until the last reply is received. (If the adserver sends an error message, the download proceeds, but MSNBC allows idiots to advertise there.)

John