Hey, stop repeating that. Following this logic, flasher-3.5 is "bad software damaging nand memory", too.

multiboot idea is to transparently flash new kernel, everytime you choose to do so from start-up menu - saving you need to use flasher-3.5 for all such cases, allowing to do so on the go, etc.

It wears NAND *equally* the same as using flasher3.5 by hand, for every kernel change, and honestly, there are 0 (zero) reports, of anyone hitting NAND wear limits by using either flasher-3.5, or multiboot.

Of course, u-boot is *much* better idea and way of achieving running different kernels - but it is not good enough reason to depreciate multiboot, which was present and used *much* longer before u-boot got available for our device.

/Estel

// Edit
Don't take personal offence (Pali) - it's just that it's 4th or 5th time I see you repeating mantra about "multiboot being bad software damaging NAND", so decided to give a matter at least little just.

Again, u-boot is much better way and I'm eternally grateful to you, for bringing it to us. You've also made a "tactical" step to drop multiboot support from kernel-power (which could support multiboot *and* u-boot flawlessly, from technical point of view) in favor of U-boot, to pro-actively encourage people to switch into U-boot.

Good decision and thinking - u-boot is better. But, it isn't fair to religiously depreciate multiboot (and by extension, it's creators), as "bad software". It was as good software as we could have for that purpose (faster, on-the-go flashing of different kernel) for the time, until better solutions arrived - nothing more, nothing less.