Forum dedicated to the new portable standalone MPC, the MPC Sample.
By J.O.BEATS Tue Mar 24, 2026 8:09 pm
Ultros wrote:And already, people want to make it a macro-mpc rather than respecting it for being a micro-sampler. if you want the big daddy sampler then why buy this?

It's the reduction of features and simplicity that makes it attractive. I hope they dont have any planned updates for it and it stays static in the way it comes outta the box. like any normal hardware sampler.



There’s already an update lol
User avatar
By Ultros Tue Mar 24, 2026 9:41 pm
J.O.BEATS wrote:There’s already an update lol


ffs.. lol thanks for the info. hmm maybe ill go download it and figure out how its constructed.
By Certified Beatz Tue Mar 24, 2026 10:05 pm
Ultros wrote:And already, people want to make it a macro-mpc rather than respecting it for being a micro-sampler. if you want the big daddy sampler then why buy this?

It's the reduction of features and simplicity that makes it attractive. I hope they dont have any planned updates for it and it stays static in the way it comes outta the box. like any normal hardware sampler.


People are never happy at all.. akai delivered an affordable sampler and now they want to tack on another thousand dollars requests :-D
Last edited by Certified Beatz on Tue Mar 24, 2026 10:05 pm, edited 1 time in total.
User avatar
By Ultros Tue Mar 24, 2026 10:17 pm
haha i'll try to see what i can do...

I am hoping all i need to do is extract the rootfs, patch it then rebuild the .exe

I'll be back in a little while with my findings.

**updates while snooping**

The device name is called the AC50 using rockchip,rk3568, and it appears to not be a classic raw rootfs image this time. instead this is a composite image with a bootimage and rootfs combined into one, i am going to have to figure out how to split the images from the file.

ah there we go... 7z x the .exe the 7z x the update.img derp de der... lol

ah shit they got a new image type.... now to learn how to extract this .fit stuffs.. (u-boot utilities)
Code: Select all##doesnt work but its a close representation of how to extract a kernel.fit
dumpimage -T fit -o 0 -d kernel.fit


Code: Select allultros@leonardo:~/Desktop/sample/mount$ ls -al
total 25322
drwxr-xr-x 3 root   root       1024 Apr 17  2025 .
drwxrwxr-x 3 ultros ultros     4096 Mar 24 18:39 ..
-rw-r--r-- 1 root   root   27776176 Apr 17  2025 kernel.fit
drwx------ 2 root   root      12288 Apr 17  2025 lost+found


**update i draw nearer**

Code: Select allUsage: dumpimage [-T type] -l image
          -l ==> list image header information
          -T ==> parse image file as 'type'
       dumpimage [-T type] [-p position] [-o outfile] image
          -T ==> declare image type as 'type'
          -p ==> 'position' (starting at 0) of the component to extract from image
          -o ==> extract component to file 'outfile'
       dumpimage -h ==> print usage information and exit
       dumpimage -V ==> print version information and exit
ultros@leonardo:~/Desktop/sample/kernel-fit$ dumpimage -l  kernel.fit
FIT description: Kernel image for rk3566-az07-ac50
Created:         Tue Apr  5 19:00:00 2011
 Image 0 (kernel)
  Description:  Linux kernel
  Created:      Tue Apr  5 19:00:00 2011
  Type:         Kernel Image
  Compression:  uncompressed
  Data Size:    21100552 Bytes = 20606.01 KiB = 20.12 MiB
  Architecture: AArch64
  OS:           Linux
  Load Address: 0x02000000
  Entry Point:  0x02000000
  Hash algo:    sha256
  Hash value:   84722800392ea2326b3019b3dad3398c3999c95fbe38587f16015c1992c0a51e
 Image 1 (ramdisk)
  Description:  Initrd
  Created:      Tue Apr  5 19:00:00 2011
  Type:         RAMDisk Image
  Compression:  uncompressed
  Data Size:    6612144 Bytes = 6457.17 KiB = 6.31 MiB
  Architecture: AArch64
  OS:           Linux
  Load Address: 0x12180000
  Entry Point:  unavailable
  Hash algo:    sha256
  Hash value:   c0044b021ce3700b47dd005e5f2b8a2fcd6b6853202ac9301b49128906499735
 Image 2 (fdt-rk3566-az07-ac50)
  Description:  rk3566-az07-ac50
  Created:      Tue Apr  5 19:00:00 2011
  Type:         Flat Device Tree
  Compression:  uncompressed
  Data Size:    60510 Bytes = 59.09 KiB = 0.06 MiB
  Architecture: AArch64
  Load Address: 0x12000000
  Hash algo:    sha256
  Hash value:   528cb279b6ebcc59b4d72de529bd025bbc40e93b6473f9d6b1a19857607dfab1
 Image 3 (fdt-cmdline)
  Description:  unavailable
  Created:      Tue Apr  5 19:00:00 2011
  Type:         Flat Device Tree
  Compression:  uncompressed
  Data Size:    189 Bytes = 0.18 KiB = 0.00 MiB
  Architecture: AArch64
  Load Address: 0x12100000
  Hash algo:    sha256
  Hash value:   9901cd3874dd84f0b5ce5e52d6c373dd5f1072c2ffea4a88e14a37a81e5142e1
 Default Configuration: 'conf-rk3566-az07-ac50'
 Configuration 0 (conf-rk3566-az07-ac50)
  Description:  rk3566-az07-ac50
  Kernel:       kernel
  Init Ramdisk: ramdisk
  FDT:          fdt-rk3566-az07-ac50
                fdt-cmdline
  Sign algo:    sha256,rsa2048:akai-1
  Sign value:   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
  Timestamp:    Tue Apr  5 19:00:00 2011
 Configuration 1 (cmdline)
  Description:  unavailable
  Kernel:       unavailable
  FDT:          fdt-cmdline
  Sign algo:    sha256,rsa2048:akai-1
  Sign value:   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
  Timestamp:    Tue Apr  5 19:00:00 2011



tools need thus far:
Code: Select all# Install build dependencies
sudo apt update
sudo apt install build-essential libncurses5-dev libssl-dev swig libfdt-dev

# Clone U-Boot source code
git clone https://github.com/u-boot/u-boot.git
cd u-boot

# Build the tools
make tools


I bet the akai-1 algo is some bullshit lame ass xor orrrr an adaptation of some common cryptographic library.. i doubt they are crafty enough to invent their own full-fledged crypto... i love that it shit out the sign value for me.. now its a matter of bruteforcing it and finding out what works.

Who knows maybe this akai-1 crap may and up opening doors on other systems as well. it's got my attention now.

now im taking a break lol... fkn uboot env is a lady's private parts to set up and im not motivated today. anyone else wanna join in? where's dustyslices... when we need him. he has the energy to go the distance.
Last edited by Ultros on Tue Mar 24, 2026 11:31 pm, edited 13 times in total.
By emdog Tue Mar 24, 2026 10:40 pm
Ultros wrote:And already, people want to make it a macro-mpc rather than respecting it for being a micro-sampler. if you want the big daddy sampler then why buy this?

It's the reduction of features and simplicity that makes it attractive. I hope they dont have any planned updates for it and it stays static in the way it comes outta the box. like any normal hardware sampler.


There is still room for them to do updates without making it too crazy. like simple sample modulations etc
User avatar
By Ultros Tue Mar 24, 2026 11:34 pm
emdog wrote:
Ultros wrote:And already, people want to make it a macro-mpc rather than respecting it for being a micro-sampler. if you want the big daddy sampler then why buy this?

It's the reduction of features and simplicity that makes it attractive. I hope they dont have any planned updates for it and it stays static in the way it comes outta the box. like any normal hardware sampler.


There is still room for them to do updates without making it too crazy. like simple sample modulations etc


true enough, and even if they do add a buncha shit i dont think it will get messy in the same regard as our current stuff.

i saw you were going to go pick one up @ emdog, did you order one or visit music world today or just gonna wait a bit and see what happens?
User avatar
By Lampdog Wed Mar 25, 2026 1:40 pm
I’ve always wanted “flagships”, never cared for micro-machines, I was expecting a normal to decent experience and was thinking I’d re-sell it. I’m keeping it. I like it more than I thought I would.
User avatar
By NearTao Wed Mar 25, 2026 2:16 pm
There's a handful of things that are fiddly here... but it's mostly a no nonsense experience where it either *does the thing* written on the tin or it doesn't.

* An xox/tr style sequencer might be nice... though for me it feels unnecessary
* The "step edit" needs some love to edit more than the velocity... slice/tune/length would all be nice
* "step edit" for note timing... struggling from the UI to figure out how to move a note...
* 16 levels feels like it needs a rework...
** tune needs to be able to specify the note... and adding root/scale options would be a welcome addition
** velocity/filter top half of the pads feel like they don't do anything...
** Adding in slice/start point would be a welcome addition (to me)
* Chops beyond 16 slices would open up a lot more workflows and make single pads have a lot more possibilities... though as is *does* work well enough
* Samples could use a *bit* more than just TRIM... some of the big brother MPC features like fade in/out, time stretch, normalize (though this *might* be in here), etc... would be welcome
* It'd be nice to have more than one Song slot... I think a lot of us will make multiple songs across the banks of samples... and we have plenty of sequences... I guess keep a notebook?

The list is mostly just little things... and we could see some of this land in a future update. Honestly a *lot* of the issues can be worked around by adjusting to the device's workflow and resampling... so I'm not even that worried about a lot of it.