Quote:
Originally Posted by Jiaz
JD will first find the next link. Here it takes in count the order/priority/forced state/waiting times/limits. In next step it finds mirrors to the link and then chooses best one of them (no waiting times, no captcha, premium if available)
|
And in here lies the problem. When JD is choosing the mirror, it leaves out a key factor from the decision making:host download speed. For example, JD chooses mirror A (100 kB/s) over mirror B (2-5 MB/s) just to avoid 1 minute waiting time. A rather annoying illogical choice
, especially with large files. Or in order to just avoid captchas, JD chooces mirror C (30 kB/s, no captchas) over mirror D (4-5 MB/s, captchas). A clear example of another bad decision.
So in other words: JD should prefer over 10 times faster hosts with short waiting times and/or captchas over slow hosts with no waiting time and/or captchas.
This is why I would prefer JD to choose the mirrors based on the order
I put them in stead of just trying to figure out the right one based on
bad logic.
Quote:
Originally Posted by Jiaz
So when you do not use mirrors, the order should be 1:1 the same as JD1
Please note that when you sort/filter the list it does NOT show the actual order of links that JD will check/try them. When a column has colored background -> you are in filtered/sorted view.
JD1: click on column header sorted the complete links
JD2: click on column header only sorts the VIEW, not the links
|
Yes, I'm well aware of this difference. That is not the problem.