How Linux friendly is this build?

[AlbertP]

What worries me by Googling is that according to the Asrock website the MB supports Windows 8. That would imply that is supports Secureboot but that it can be disabled. But as far as I can tell there isn't the slighest mention of secureboot, much less disabling it in the manual.

Windows 8 is not preinstalled, so it doesn't need Secure Boot. Even if it were available, it wouldn't be enabled by default to allow people to install Windows 7 or XP. It's really not a problem to use a Windows 8 compatible computer, at least if Windows 8 is not preinstalled or included with the computer.

After all I think this is a very good choice. The hardware should all be working with Linux if you install the NVIDIA proprietary driver for the graphics.

[archibald]

It was meant to be amusing, I should have added a smiley to it :-)

[frantaylor]

The marketing department doesn't always adhere to grammar (or logic).

The "marketing department" is actually 100% brilliant for creating a new word that can be trademarked so that its usage can be controlled and the consumer can know that they are purchasing a drive that meets the specifications.

[DanL]

Blue-ray (yes, I did mean to type it like that, the colour has an 'e')

The marketing department doesn't always adhere to grammar (or logic). The name is 'Blu-Ray', so trying to convince everyone that it's 'Blue-Ray' is like screaming, "I'm a moron!" and then trying to convince everyone that it's a good thing to be a moron because you use a different definition than the rest of the world uses..

[frantaylor]

you gotta try it

I've put together systems where all of the components individually were right off the RHEL supported list but when assembled together, no go. There are obscure interactions between kernel versions, motherboard chipsets and drive firmwares. Sometimes firmware upgrades fix problems and sometimes they break things. I've spent far too much time debugging these things trying to make a stable system. There is no rhyme or reason, only empirical results.

You actually have to plug all of those things together, install the OS and run some torture tests before you can call it "working".

[Kano]

If you want to try UEFI the most secure way is to have got a backup of your firmware chip. Sometimes you can do that with the integrated flash utility. When you have it you could order in case of problems a correct replacement or the mac adress of your integrated nic will be a default one. If you manage to "kill" a laptop with uefi tests you should probably get a spi flashing device when it is not in warrenty. I needed a chip replacement already last year after some simple tests, so update firmware to latest version and hope that most bugs are already fixed (board used was Asus P8Z68-V with buggy firmware 3203). All Asus boards suffer from another cleanup UEFI entry error but there is a trick to avoid it when you use differnet upper/lowercase for the loader (efibootmgr -l) in case you really want to use different optionsets with one kernel as direct boot entry.

[TheLexMachine]

This is encouraging. What if any are the major issues to look out for with regard to UEFI support under Linux?

If you're using integrated graphics, nothing. If you are using an add-in card, which you intend to, disabling UEFI in the BIOS is the only way to use your add-in card unless it's UEFI compatible. There may be problems with other add-in cards as well but only GPUs have been reported at this time.

[Kristian Joensen]

This is encouraging. What if any are the major issues to look out for with regard to UEFI support under Linux?

[brosis]

I believe the appropriate wording you are looking for is "unfortunate". By quoting my post and saying "This is IMHO incorrect", you are saying in your post that what I said was incorrect when it is in fact correct and the rest of your post agrees with what I said. As for the drives, all new AF drive models - which are all they make now - now only work as AF drives and no longer have a fall-back mode where a jumper over a pair of pins makes the drive appear to be the same as the older drives by modifying the controller processes.

Hello, no, my wording is appropriate. I disagreed with

Remember that not all disk utilities in Linux properly recognize and format Advanced Format drives

by saying that

only parted/gparted are able to map the disks correctly and also create a GPT table. Fdisk and other tools have not been updated for years and their use is IMHO a legacy.

projected to msdos era, the claims would be
"not all disk utilities support ntfs"

with my counter claim
"all of them support it, just not the legacy stuff which no one should ever use"

Also, it has nothing to do with drive switch, they still report themselves as 512-sector size drive. There is eventually a flag in harddisk parameters "AF:true/false", but even there is false information, at least by all Seagate and Samsung drives.

I hope that clarifies everything.

[TheLexMachine]

Hello. This is IMHO incorrect, as only parted/gparted are able to map the disks correctly and also create a GPT table. Fdisk and other tools have not been updated for years and their use is IMHO a legacy.
Also, my experiences are such that Advanced Format harddisks are all falsely marking themselves as "normal 512-kb" drives, even in SMART. Possibly its a workaround for pre-Windows Vista compatibility.
Eitherway, the reliable way is to check if its AF drive is the disk manufacturer documentation or question on official forums.

I believe the appropriate wording you are looking for is "unfortunate". By quoting my post and saying "This is IMHO incorrect", you are saying in your post that what I said was incorrect when it is in fact correct and the rest of your post agrees with what I said. As for the drives, all new AF drive models - which are all they make now - now only work as AF drives and no longer have a fall-back mode where a jumper over a pair of pins makes the drive appear to be the same as the older drives by modifying the controller processes.