easier portability would be the only thing I can think of. For example, I had 2 laptops, One is faster and used for demos. The second one is a little slower and is just used for remote connectivity wit the office. I had complete battery failure on the faster one. No time to get another before people had to leave for the demo. I switched the hard drive that has a complete setup, ready to go to the other laptop. I made a couple fo quick changes and verified settings and it was ready to go. I did not have to configure, compile, package, and install a kernel to do so. This is great because laptop compile speeds stink. I have also used distcc for this but still the process is fairly long. This is just one example. Changing out motherboards would be another. It removes a bit of hassle and lends some flexibility.
Robert On Sunday 20 August 2006 10:07, [EMAIL PROTECTED] wrote: > Since long I have two questions, and this is the first one: > > I see initrd kernels all around, and i can imagine the benefits > for 'hijacking' systems (like installers) which need to discover the > hardware first to select the appropriate kernel modules and settings. > > But when i'm going to configure a custom kernel, on known hardware, > why should i use initrd at all ? I mean, what is the advantage of initrd > over non-initrd in this case ? Really, any argument ? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
