H.264 vs H.265 vs AV1: Hardware Encoding Explained
Understand the differences between H.264, H.265, and AV1 codecs. Learn how hardware encoding works and which codec to choose for screen recording in 2026.
What Is Video Encoding?
When you record your screen, each frame is captured as a raw image. A 1080p frame at 32-bit color is about 8 MB. At 30 fps, that’s 240 MB per second — 14 GB per minute of raw video.
Obviously, you can’t store raw video. Encoding (also called compression) reduces this massive data stream to something manageable. A 1-minute 1080p recording typically compresses to 50–200 MB depending on the codec and quality settings.
Encoding works by finding redundancy in the video data. If the background doesn’t change between frames, there’s no need to store it again. If two adjacent pixels are the same color, they can be represented once. Codecs use sophisticated algorithms to find and exploit these patterns.
Software vs Hardware Encoding
Software Encoding (CPU)
Software encoding uses your CPU to compress video. The CPU runs the codec algorithm instruction by instruction. This produces excellent quality but consumes significant CPU resources during recording.
When your CPU is busy encoding video, everything else slows down. Games drop frames, applications stutter, and the system feels sluggish. This is why software encoding is generally not recommended for screen recording.
Hardware Encoding (GPU)
Modern GPUs include dedicated hardware blocks specifically designed for video encoding:
- NVIDIA NVENC — available on GTX 600 series and newer
- Intel Quick Sync (QSV) — available on most Intel CPUs with integrated graphics
- AMD AMF/VCE — available on AMD GPUs and APUs
These hardware encoders operate independently from the GPU’s rendering cores. They can encode video with near-zero impact on CPU performance and minimal GPU impact. This means you can record your screen at high quality without slowing down your computer.
The quality difference between hardware and software encoding has narrowed significantly. Modern hardware encoders like NVENC (Turing/Ampere/Ada) produce output very close in quality to software x264 at the same bitrate. For screen recording, hardware encoding is a practical choice that balances quality and performance.
The Three Codecs: H.264, H.265, and AV1
H.264 (AVC) — The Universal Standard
H.264 has been the dominant video codec since 2003. It’s supported everywhere — every browser, every media player, every device. If you need maximum compatibility, H.264 is the safe choice.
Pros:
- Universal playback support
- Fast encoding and decoding
- Mature and well-optimized
- Works on all hardware encoders
Cons:
- Larger file sizes compared to newer codecs
- Less efficient compression
Best for: Recordings you’ll share widely, upload to any platform, or play on older devices.
Typical file size: A 10-minute 1080p 30fps screen recording at good quality ≈ 150–300 MB.
H.265 (HEVC) — The Efficient Successor
H.265 was designed as H.264’s successor, offering 30–50% better compression at the same visual quality. A recording that takes 200 MB in H.264 might take 100–140 MB in H.265.
Pros:
- Significantly smaller files than H.264
- Excellent quality at lower bitrates
- Hardware encoding support on modern GPUs
- Good for archival and storage-conscious users
Cons:
- Not as universally supported (some older devices and browsers)
- Slightly slower encoding than H.264
- Patent/licensing complexity (though this doesn’t affect end users)
Best for: Keeping file sizes manageable when recording long sessions, archiving recordings.
Typical file size: A 10-minute 1080p 30fps screen recording ≈ 80–180 MB.
AV1 — The Future Standard
AV1 is the newest codec, developed by the Alliance for Open Media (including Google, Apple, Microsoft, Netflix, and others). It offers 30–50% better compression than H.265 — meaning roughly half the file size of H.264 for equivalent quality.
Pros:
- Smallest file sizes of all three codecs
- Royalty-free (no licensing costs)
- Backed by major tech companies
- Increasingly supported in browsers and platforms
Cons:
- Hardware encoding only available on newest GPUs (NVIDIA RTX 40 series, Intel Arc, AMD RX 7000 series)
- Slower encoding than H.264/H.265
- Not yet universally supported for playback
Best for: Users with modern GPUs who want the smallest possible file sizes and don’t need broad compatibility.
Typical file size: A 10-minute 1080p 30fps screen recording ≈ 50–120 MB.
Side-by-Side Comparison
| Feature | H.264 | H.265 | AV1 |
|---|---|---|---|
| Release year | 2003 | 2013 | 2018 |
| Compression efficiency | Baseline | +30–50% vs H.264 | +30–50% vs H.265 |
| File size (relative) | 1x | 0.5–0.7x | 0.3–0.5x |
| Encoding speed | Fastest | Medium | Slowest |
| Compatibility | Universal | Wide | Growing |
| NVIDIA support | GTX 600+ | GTX 950+ | RTX 4000+ |
| Intel support | All QSV | Skylake+ | Arc GPUs |
| AMD support | All VCE | Polaris+ | RX 7000+ |
| Royalty status | Licensed | Licensed | Royalty-free |
Quality Settings: CRF and Bitrate
When encoding video, you control quality through two main approaches:
Constant Rate Factor (CRF)
CRF lets the encoder automatically choose the bitrate needed to maintain a target quality level. Lower CRF = higher quality = larger files.
| CRF | Quality | Use case |
|---|---|---|
| 18 | Visually lossless | Archival, source material |
| 23 | High quality | General purpose (default) |
| 28 | Good quality | Smaller files, web sharing |
| 35+ | Noticeable compression | Minimum file size |
CRF is recommended for screen recording because it adapts to content complexity. Static desktop content gets low bitrate (small file), while motion-heavy content gets more bitrate (preserving quality).
Constant Bitrate (CBR)
CBR forces a fixed bitrate regardless of content. This is mainly useful for streaming where bandwidth is limited. For recording to disk, CRF is almost always the better choice.
Which Codec Should You Choose?
For everyday screen recording: Use H.264 with CRF 23. Maximum compatibility, good quality, reasonable file size. This is the safe default.
For long recording sessions: Use H.265 with CRF 23. Your files will be 30–50% smaller, saving significant storage on multi-hour recordings.
For modern setups with new GPUs: Use AV1 if your GPU supports hardware encoding. Smallest files with excellent quality. Check your GPU first — AV1 hardware encoding requires recent hardware.
If unsure: Start with H.264. You can always convert to a more efficient codec later using a tool like DalConvert without re-recording.
How DalVideo Handles Encoding
DalVideo supports all three codecs with automatic hardware encoder detection. When you select a codec, DalVideo checks if your GPU supports hardware encoding for that codec and uses it automatically. If hardware encoding isn’t available, it falls back to software encoding.
The default settings (H.264, CRF 23, hardware encoding) work well for most users. You can change the codec and quality in Settings without needing to understand the technical details — DalVideo picks the right encoder and parameters for your hardware.
To try it yourself, download DalVideo for free. The free version includes all codecs with no restrictions on quality or format.
The Bottom Line
- H.264 for compatibility — works everywhere, established standard
- H.265 for efficiency — same quality, much smaller files
- AV1 for the future — best compression, needs modern hardware
All three produce excellent quality for screen recording when using hardware encoding. The choice mostly comes down to file size vs. compatibility. And if you’re not sure, H.264 is never the wrong answer.
DalVideo
Try it free