www.makemkv.com

Hello,

Despite reading a ton of topics here about LibreDrive, I still haven't quite grasped the details essential for me.

From what I’ve understood so far, it’s a mode of operation for an optical disc drive that, as of now, only one program—MakeMKV—can make use of. Is that correct?

Here’s the practical issue I’m facing: I’d like to play an AACS v81 Blu-ray disc on my Linux machine. By "play," I mean the following process: https://web.archive.org/web/20250923123 ... wsmy.name/.

I’ve already successfully ripped the disc using MakeMKV. I’ve never flashed my drive, so the firmware is stock. Now, MakeMKV reports that the highest AACS version supported by the drive is 81. The disc title doesn’t seem to be present in the publicly available KEYDB file.

This raises two further questions:

1. If I flash the drive to MK firmware now, will it be able to decrypt the disc (i.e., allow my machine to play it)?
2. If not, would flashing to MK firmware before the drive saw the disc have made decryption possible? Does the answer depend on the drive’s highest AACS version? Unfortunately, I didn’t check the highest AACS version before inserting the disc.

A drive's highest AACS version is a record of the highest version of AACS that the drive has ever seen. It will change regardless of wether or not the drive's firmware has been flashed to support LibreDrive.

https://forum.makemkv.com/forum/viewtopic.php?p=184611#p184611

iskander wrote: ↑

Fri Oct 03, 2025 3:08 pm

I’ve already successfully ripped the disc using MakeMKV. I’ve never flashed my drive, so the firmware is stock. Now, MakeMKV reports that the highest AACS version supported by the drive is 81. The disc title doesn’t seem to be present in the publicly available KEYDB file.

Have you tried adding that disk's key to the KEYDB file, since it is currently missing? Since the highest MKB version that your drive has seen is MKBv81, you should be able to find an unrevoked host certificate to allow libaacs to obtain the disk's VID. Has the aacs_info tool already computed the disk's Media Key for you?

MrPenguin,

I would be happy to add the disk's key to the KEYDB file (I suppose it's the VUK we are talking about), but the latest certificates/keys available in the public KEYDB file are apparently revoked by my drive. To this day, I've never heard of aacs_info tool. I found out about the revocation with bd_info tool. I will later try aacs_info, since it's apparently part of libaacs.

Overly simplified, LibreDrive does two things:
1. Allows one to read VID from disc
2. Disables bus encryption

You can solve (1) with keydb. But you still need LibreDrive microcode(or a valid UHD host certificate) for (2).

mike admin wrote: ↑

Sat Nov 08, 2025 9:42 pm

Overly simplified, LibreDrive does two things:
1. Allows one to read VID from disc
2. Disables bus encryption

You can solve (1) with keydb. But you still need LibreDrive microcode(or a valid UHD host certificate) for (2).

Aside from microcode and valid host certs, it is also an option to use patched firmware with LG/Asus drives which disables bus encryption entirely.

But my preferred option is to use a microcode extension to calculate the unique drive seed value and then use the drive seed to calculate the RDK from the VID when required. This allows for playback/decryption of any disc as long as the keydb has at least the VID and UK.

keydb_helper wrote: ↑

Sun Nov 09, 2025 1:27 am

But my preferred option is to use a microcode extension to calculate the unique drive seed value and then use the drive seed to calculate the RDK from the VID when required. This allows for playback/decryption of any disc as long as the keydb has at least the VID and UK.

Interesting! My understanding is that each drive has its own unique "drive seed" (16 bytes), and that each RDK is actually just the disk's VID encrypted using this drive seed. Hence why each RDK depends on both disk and drive.

OTOH it has taken Mike 2 years to add LD support for the most recent batch of Pioneer drives. The probability therefore of "teaching" every UHD drive to provide its drive seed as you suggest would seem to be "very low" - unless Mike were also to help .

iskander wrote: ↑

Fri Oct 03, 2025 3:08 pm

I’d like to play an AACS v81 Blu-ray disc on my Linux machine. By "play," I mean the following process: https://web.archive.org/web/20250923123 ... wsmy.name/.

Outside what has been already discussed, since you'rr able to successfully rip the disc, you can also simply use MakeMKV to create an unencrypted copy. This has the benefit that the copy can be played without any key nonsense.

It has the added benefit that if the title is UHD, it can then played using any BDXL drive - even non-UHD ones, which essentially means all BluRay drives made since 2010. Or it can be played as a BDISO.

It does have the drawback that if its a UHD disc copied to BDXL, then most UHD standalone Blu-ray video players will refuse to play it. But it sounds like you want it to play on VLC, so an unencrypted copy will save you a lot of grief if you want to play from diac .

MrPenguin wrote: ↑

Sun Nov 09, 2025 12:15 pm

Interesting! My understanding is that each drive has its own unique "drive seed" (16 bytes), and that each RDK is actually just the disk's VID encrypted using this drive seed. Hence why each RDK depends on both disk and drive.

Yes, you are correct that the drive seed is unique to each drive. It is derived from the upper 128 bits of the drives private key. You can have a lot of fun with both AACS 1.0 and 2.x after extracting a drives private key and drive certificate...

MrPenguin wrote: ↑

Sun Nov 09, 2025 12:15 pm

OTOH it has taken Mike 2 years to add LD support for the most recent batch of Pioneer drives. The probability therefore of "teaching" every UHD drive to provide its drive seed as you suggest would seem to be "very low" - unless Mike were also to help .

I can't do much with Pioneer drives but I've implemented drive seed extraction for all available MT1959 DE/MK firmware versions. I'd be happy to provide my solution to Mike if he'd like to include it in LibreDrive.

keydb_helper wrote: ↑

Fri Nov 14, 2025 11:38 am

MrPenguin wrote: ↑

Sun Nov 09, 2025 12:15 pm

Interesting! My understanding is that each drive has its own unique "drive seed" (16 bytes), and that each RDK is actually just the disk's VID encrypted using this drive seed. Hence why each RDK depends on both disk and drive.

Yes, you are correct that the drive seed is unique to each drive. It is derived from the upper 128 bits of the drives private key. You can have a lot of fun with both AACS 1.0 and 2.x after extracting a drives private key and drive certificate...

MrPenguin wrote: ↑

Sun Nov 09, 2025 12:15 pm

OTOH it has taken Mike 2 years to add LD support for the most recent batch of Pioneer drives. The probability therefore of "teaching" every UHD drive to provide its drive seed as you suggest would seem to be "very low" - unless Mike were also to help .

I can't do much with Pioneer drives but I've implemented drive seed extraction for all available MT1959 DE/MK firmware versions. I'd be happy to provide my solution to Mike if he'd like to include it in LibreDrive.

What's the benefit of that?

Coopervid wrote: ↑

Fri Nov 14, 2025 12:38 pm

What's the benefit of that?

Direct playback or ripping of AACS 1.0/2.x titles without needing to worry about host certificate revocation or the ongoing usage of microcode blobs. You would only need the public keydb to contain at least the unit key (UK) and volume ID (VID) of your disc. If both are present and you know your drive seed, you could even playback/rip UHD titles on the latest stock, encrypted LG drive firmware versions. I wouldn't recommend that as you lose the ability to generate usable disc dump files but it is still pretty neat as the drive seed is consistent regardless of the specific firmware version.

An additional, even more niche, use case would be decrypting an inadvertently bus encrypted ISO by calculating the RDK after ripping.

keydb_helper wrote: ↑

Fri Nov 14, 2025 1:36 pm

An additional, even more niche, use case would be decrypting an inadvertently bus encrypted ISO by calculating the RDK after ripping.

That's an interesting thought. You'd be surprised how many people end up doing this only to find out they have a completely useless ISO image. Fascinating.

keydb_helper wrote: ↑

Fri Nov 14, 2025 1:36 pm

Coopervid wrote: ↑

Fri Nov 14, 2025 12:38 pm

What's the benefit of that?

Direct playback or ripping of AACS 1.0/2.x titles without needing to worry about host certificate revocation or the ongoing usage of microcode blobs. You would only need the public keydb to contain at least the unit key (UK) and volume ID (VID) of your disc. If both are present and you know your drive seed, you could even playback/rip UHD titles on the latest stock, encrypted LG drive firmware versions. I wouldn't recommend that as you lose the ability to generate usable disc dump files but it is still pretty neat as the drive seed is consistent regardless of the specific firmware version.

An additional, even more niche, use case would be decrypting an inadvertently bus encrypted ISO by calculating the RDK after ripping.

Fascinating! As a side note: It is really not easy to get Mike's attention. Before he updated Libredrive these days his last visit to the forum was in May. Have you thought about to release your own tools?

Coopervid wrote: ↑

Fri Nov 14, 2025 3:11 pm

Fascinating! As a side note: It is really not easy to get Mike's attention.

I have noticed... But I also appreciate his tacit approval of the keydb thread and the fact he hasn't revised the LibreDrive metadata obfuscation for LG/Asus drives so I'm not going to push my luck.

Coopervid wrote: ↑

Fri Nov 14, 2025 3:11 pm

Have you thought about to release your own tools?

Yes. No time frame at this point.