Quote from the article:

Why not offer the option to disable it? A user who does would end up with a machine that doesn't boot, and if they managed to figure that out they'd have worse power management. That increases support costs. For a consumer device, why would you want to? The number of people buying these laptops to run anything other than Windows is miniscule

Well, the minuscule group they just shat on is the same group of people who make purchasing decisions and recommendations for a ton of consumers. Even if someone doesn't run Linux, laptops that support it have often proven to be more reliable and have a higher resell value.

For everyone asking me what laptop they should buy, I'll be sure to let them know how disorganized and confused Lenovo support tends to be.

If 1% of your customers cause 10% of support tickets, it might be profitable to stop selling to those customers. Lenovo has no obligation to you or anyone else.

Firstly, companies spend a pretty penny to associate their brand with vague positive platitudes. In the first place being associated with something negative even if their populace doesn't understand the technicalities can be worth more than 1% of your business.

Second tech savvy users influence others.

Thirdly supposing you publish sufficient tech specs as to enable the kernel to support your hardware I'm guessing linux users actually require less not more tech support.

Windows users repeatedly trash their own machines with malware and pick up the phone to call their oem. Linux users google the error message and open a bug on the project that actually doesn't work's site.

I'm not even sure where your hostile attitude comes from. Lenovo just dropped the ball.

Magic windows bios raid modes have existed for the 13 years I've been using linux and I have never seen one that didn't include a switch in the bios to disable it.

What you are incorrectly framing as lenovo righteously dumping entitled linux fanboys is in fact just Lenovo not knowing how to design a computer which given repeated mistakes in recent years is hardly shocking.

Since when is it firmware's business to care about dumb users well-being and, more specifically, to disallow switching perfectly capable storage controller into standard AHCI mode?

Why doesn't Linux support it? Is "RAID" mode based on some spec that was only provided to Microsoft and kept secret from everyone else?

Windows only supports it if you use the Intel storage driver - even Microsoft don't ship a driver that supports it.

> even Microsoft don't ship a driver that supports it.

.....and why do you suppose that is? Fake RAID drivers and hardware are shit that's why. It's an option that is hardly used, if ever, and I never thought I'd see the day when a system had it as its default.

Yay, so supporting completely non-standard and inevitably more buggy drivers is the correct option?

Linux has its own methodology for providing raid and the raid modes provided by motherboard makers are in general terrible junk that nobody including windows users should use. Even if the information necessary to implement support were available which it generally is not it would be worth less than nothing.

Everyone's hardware supports communicating with drives in a standard fashion not providing a way to enable this given that the hardware already supports it is wholly and totally lenovos fault.

MS doesn't write a lot of drivers for hardware. Even the stuff they provide on disk has a lot of drivers written by a third party. Intel's RAID mode is something that's not supported by default on Windows 10 even, which means you have to have the special RAID driver for even Windows to work on it. Intel has the driver info and isn't even releasing it to any OS makers as far as I can tell.

> "RAID" mode is typically just changing the PCI IDs so that the normal drivers won't bind, ensuring that drivers that support the software RAID mode are used.

It sounds like the kernel just needs to learn about the new PCI ID.

I doubt Microsoft knows the spec to it, all they care about is that the manufacturer provides a working driver.

Lenovo may have set power saving standards for the laptop based on what they want to market. It's Intel that dropped the ball here. They need to submit patches for RAID to upstream Linux. Or it could just be a firmware bug.

Wait, does RAID actually save power over AHCI (single-drive RAID)? If so, how?

For one thing, no benchmarks I've seen of nvme SSD has shown them to be significantly faster if the BIOS is set to RAID, but I've not heard any thing for or against the power savings benefits of it.

You misread it. RAID forces the use of the Intel driver vs the Microsoft one. The Intel provided driver enables power saving.

So it's not the RAID that saves power, it's the [use of the] driver.

Okay, then make the default setting RAID in the BIOS, don't proactively prevent (or should I say, restrict) that setting from being changed at all.

Why? Why should Lenovo let 99% of their users use sub-par drivers when they reinstall Windows?

Where did I say that Lenovo should change the default setting?

To be clear, I'm perfectly fine with a default setting of RAID -- but right now, they also lock you out of changing it back to AHCI.

What I think is reasonable is allowing users to change the setting back to AHCI (at their own risk, of course, as is the case with MOST settings in the BIOS). They should have done that all along, but instead, they artificially restricted it, locking us Linux users out by their own volition.

I think the RAID setting can be the default. But they should provide the option to change if users want to install linux. Most linux users are BIOS friendly anyway.

Answering you and Hydraulix989:

Because Internet forums. That's why. Search for "I can't install Windows 10 on my Lenovo Whatsyourface" and you'll get "Change this super secret setting to AHCI". And who's going to be doing those searches? Everyone, but specially people less computer savvy. The ones that know enough are here discussing Lenovo's decision.

Again, just to clarify: I'm not trying to defend Lenovo. What they did was a bad business decision, but maybe it was the lesser of two evils. Maybe they should've kept Intel accountable on this one - but who can argue with the only serious CPU manufacturer for laptops?

Every other chipset and motherboard in existence has had that option and it hasn't affected anyone adversely. There is no less of two evils. There isn't any evils here.

That's not what the article claims. Apparently there are indeed two evils here:

Lower performance for everyone or no drivers for Linux.

I take you know more than the article's author, in which case I'd be interested in reading your analysis as I'm not an expert.

There is no way that using a fake RAID set of drivers and a completely incompatible and more complex on-disk RAID format is going to increase performance in any way.

There's a reason why that hasn't been done on any system in the past - because it's stupid and fake RAID drivers are notoriously buggy and require support from cloning and other low level software that gets used.

Yer, you're not an expert, or appear to know anything about hardware.

Because providing that option doesn't give anyone sub-par drivers?

Also, the power saving stuff is bollocks. I never thought I would see the day when anyone would recommend using a bullshit fake RAID driver for power saving reasons.

Please refer to my other response to you. It would be awesome to read your analysis of the drivers and the underlying hardware.

I base my opinion on what experts call facts as I'm not a hardware expert myself.

It would be awesome if you knew what you were talking about ;-).

We're all just waiting for you to enlighten us. Go ahead, show us what you've got :-)

There's actually some interesting OSS politics at hand (it appears that this new ultrabook BIOS may have been "copied" over from the Lenovo server BIOS):

http://www.gossamer-threads.com/lists/linux/kernel/2352338