{"id":41954,"date":"2026-05-22T17:00:00","date_gmt":"2026-05-22T16:00:00","guid":{"rendered":"https:\/\/3dwork.io\/?p=41954"},"modified":"2026-09-14T23:25:03","modified_gmt":"2026-09-14T21:25:03","slug":"manta-m8p-v2-ebb36-gen2-review","status":"publish","type":"post","link":"https:\/\/3dwork.io\/en\/manta-m8p-v2-ebb36-gen2-review\/","title":{"rendered":"Bigtreetech Manta M8P v2 + EBB36 Gen2: Review"},"content":{"rendered":"\n

The Bigtreetech Manta M8P v2, together with the EBB36 Gen2, offers an exceptional solution for 3D printer customization enthusiasts<\/strong>. Its ability to adapt to various configurations and technical requirements makes it a preferred choice among advanced users.<\/p>\n\n\n\n

At 3Dwork, we have spent weeks testing this combination on different machines<\/strong>: what we are going to tell you during the review is what the experience with these electronics really is, what things can be improved, and in which cases the investment makes sense.<\/p>\n\n\n\n

\"-<\/figure>\n\n\n\n

We are going to analyze these control electronics that let you create and improve your machines without relying on third-party proprietary firmware… maximum control, maximum use of your machine, and a lot of knowledge<\/strong> that we will gain.<\/p>\n\n\n\n

Do you have questions about these electronics? Join our community: Instagram<\/a>, YouTube<\/a>, TikTok<\/a>. And on Telegram: the general 3Dwork<\/a> group, the Bigtreetech<\/a> group, and the Klipper<\/a> group \u2014 all of them in Spanish.<\/strong><\/p>\n\n\n\n

Bigtreetech Manta M8P v2<\/h2>\n\n\n\n

What is the Manta M8P v2?<\/strong> It is a 3D printer controller board based on the STM32H723ZET6, a Cortex-M7 processor at 550 MHz with a floating point unit integrated. It is just the control electronics.<\/strong> The host module (CB1, CB2 or Raspberry Pi CM4) is purchased separately and integrates into the board’s BTB socket.<\/p>\n\n\n\n

\"-<\/figure>\n\n\n\n

The MCU, heart of the M8P v2, mounts an STM32H723ZET6 processor at 550 MHz:<\/strong> it features 376 KB of SRAM, 1024 KB of Flash, and integrated L1 cache memory. It also has native I2C at 400 kHz (previously this had to be done via GPIO, which was slow and less reliable). Dedicated servo port with precise PWM for BLTouch, Euclid, or similar systems.<\/p>\n\n\n\n

\"-<\/figure>\n\n\n\n

Regarding peripherals we can connect, the M8P v2 comes loaded with options, among which we can highlight:<\/strong><\/p>\n\n\n\n

    \n
  • up to 8 motor drivers with intelligent SPI\/UART communication<\/strong><\/li>\n\n\n\n
  • 4 PWM outputs<\/strong> for hotend, bed, fans<\/li>\n\n\n\n
  • 4 ADC inputs<\/strong> for thermistors<\/li>\n\n\n\n
  • CAN bus FD<\/strong> up to 1 Mbps for communication with toolboards<\/li>\n\n\n\n
  • BTB Socket<\/strong> integrated for host<\/li>\n<\/ul>\n\n\n\n
    \"-<\/figure>\n\n\n\n

    The Manta M8P v1 used STM32G0B1 at 64 MHz, while the v2 switched to STM32H723 at 550 MHz<\/strong>. That is 8.5 times more raw processing power; with the v1 you could be limited to standard kinematics, creating problems if we use complex algorithms simultaneously on more complex platforms.<\/p>\n\n\n\n

    With the v2, you run full Input Shaping, real-time collision detection, complex kinematics like vzBot without slowdowns, advanced sensor filtering with Kalman filters, and USB camera processing since the CPU processor does not create bottlenecks on the host.<\/p>\n\n\n\n

    \"-<\/figure>\n\n\n\n

    There is another less obvious but important change… the v1 has had some issues with reboots during CM4\/BTB startup. The culprit was the Manta’s 5V rail, which could only deliver 500mA. The CM4 during startup can need peaks of 1-1.5A. The v2 added high-capacity capacitors (4x 100\u00b5F + 4x 10\u00b5F) that completely stabilize this, preventing or reducing the likelihood of this problem occurring.<\/p>\n\n\n\n

    Host modules: CB1 v2, CB2, Raspberry Pi CM4<\/h2>\n\n\n\n

    Today, with the rise of firmwares like Klipper or the use of Marlin with Octoprint, the need for an external host to control the MCU is a necessity. The Bigtreetech series’s Manta solves this by integrating CM4\/BTB modules into the electronics themselves with a direct connection to the MCU, <\/strong>easily leaving a compact solution that can be perfectly integrated into any machine<\/strong>.<\/p>\n\n\n\n

    \"-<\/figure>\n\n\n\n

    Thanks to this, we have several options to use as a host:<\/p>\n\n\n\n

      \n
    • CB1 v2 (15-20\u20ac):<\/strong> Allwinner H616 at 1.2 GHz single-core ARMv7, 1 GB RAM, 100 Mbps Ethernet. It works well for basic Klipper without intensive use or heavy features. It is the ideal solution if you have a tight budget and a standard machine that does not require major extra features at the Klipper level.<\/li>\n\n\n\n
    • CB2 (35-50\u20ac):<\/strong> Rockchip RK3566 quad-core at 1.8 GHz, 2-4 GB RAM, Gigabit Ethernet, integrated eMMC (no fragile microSD), dual-band WiFi. It is usually the best price-performance ratio.<\/li>\n\n\n\n
    • Raspberry Pi CM4 (25-80\u20ac):<\/strong> BCM2711 quad-core at 1.5 GHz. If you already have one at home, using it is cost-effective. In pure Klipper, similar to the CB2. Better for heavy applications like: video analysis with OpenCV, IP camera streaming, AI for image analysis and problem detection, and control of multiple instances\/printers.<\/li>\n<\/ul>\n\n\n\n
      \"-<\/figure>\n\n\n\n