Uploaded September 2026 | Updated November 2026, 1 day ago
Start at minute - 3:37 - Sorry for the technical difficulties.
RISC-V Builders is a YouTube Live series where people from a project come together for a conversational chat about what they built, how they built it, and what they learned along the way.
Episode 1: Project Tartarus started with a simple question: can a student-built RISC-V processor run DOOM?
In this RISC-V Builders session, Tobias Scheipel (Asst. Professor at Graz University of Technology) and Bernd Stelzl (Master’s Student in Computer Science at Graz University of Technology) discuss how HaDes-V grew from a teaching core into a complete RV32IM system with DDR memory, graphics, keyboard input, audio, and enough performance to make DOOM playable. Along the way, they share the engineering challenges, design decisions, and lessons learned from pushing a small RISC-V platform far beyond its original purpose.
Short outline of the conversation:
- Introduction: HaDes-V, the teaching project, and how Tartarus started
- The challenge: What was missing before DOOM could even run
- Building Tartarus: CPU extensions, memory, graphics, input, and audio
- Making it playable: Profiling, bottlenecks, and hardware/software optimization
- What went wrong: Debugging challenges and lessons from the implementation
- What we learned: RISC-V, processor design, and using real workloads in education
Start at minute - 3:37 - Sorry for the technical difficulties.
RISC-V Builders is a YouTube Live series where people from a project come together for a conversational chat about what they built, how they built it, and what they learned along the way.
Episode 1: Project Tartarus started with a simple question: can a student-built RISC-V processor run DOOM?
In this RISC-V Builders session, Tobias Scheipel (Asst. Professor at Graz University of Technology) and Bernd Stelzl (Master’s Student in Computer Science at Graz University of Technology) discuss how HaDes-V grew from a teaching core into a complete RV32IM system with DDR memory, graphics, keyboard input, audio, and enough performance to make DOOM playable. Along the way, they share the engineering challenges, design decisions, and lessons learned from pushing a small RISC-V platform far beyond its original purpose.
Short outline of the conversation:
- Introduction: HaDes-V, the teaching project, and how Tartarus started
- The challenge: What was missing before DOOM could even run
- Building Tartarus: CPU extensions, memory, graphics, input, and audio
- Making it playable: Profiling, bottlenecks, and hardware/software optimization
- What went wrong: Debugging challenges and lessons from the implementation
- What we learned: RISC-V, processor design, and using real workloads in education









