Modern stills cameras that support video (e.g. mirrorless or DSLR bodies) often provide multiple codec/format options so that videographers can strike the right balance between image quality, file size, editing workflow demands, and hardware constraints. Among the commonly offered formats are Intra‑frame (All‑Intra / S‑I), Long‑GOP (e.g. “S” style), and High Efficiency variants (e.g. “HS” / HEVC). Below is a structured comparison and explanation.
1. Codec Types & Compression Schemes
Intra‑Frame Compression (All‑Intra / S‑I types)
- In this scheme, each frame is compressed independently, without reference to preceding or subsequent frames.
- Because each frame is self‑contained, editing, colour grading, effects or cuts are more forgiving, and CPU load during post is lower per frame.
- The downside is file size: to maintain quality, bitrates must be high, leading to heavier storage requirements.
- Many cameras support “XAVC S‑I” (or equivalent proprietary all‑intra modes) for high end video recording. For example, Sony’s guide notes that XAVC S‑I 4K / HD uses
- MPEG‑4 AVC / H.264 intra compression, and is better for editing compared to Long GOP alternatives. 
- Because the compression is simpler (no inter‑frame prediction), the decoding and random access is easier, reducing lag in edits.
Long‑GOP Compression (e.g. “S” / “XAVC S”)
- Long GOP stands for Group of Pictures: frames are compressed by referencing other frames (I‑frames, P‑frames, B‑frames), which can yield much better compression efficiency.
- This means much smaller file sizes for a given visual quality (or, equivalently, more footage stored for the same storage).
- The trade‑off is that editing is more complex: the system must reconstruct prediction frames, manage dependencies, and the editing software must handle the complexity.
- In practice, for many use cases—YouTube, casual production, corporate video—Long‑GOP offers a pragmatic balance.
- Sony’s documentation confirms that XAVC S uses Long GOP (H.264 / AVC) compression. 
High Efficiency / HEVC Variants (e.g. “HS”, H.265, XAVC HS)
- These codecs apply more advanced compression algorithms (HEVC / H.265) to achieve better compression efficiency than H.264, meaning smaller file sizes for comparable or slightly superior quality.
- In effect, they push the same image quality into fewer bits, which is attractive for long shoots or limited storage.
- However, the trade is heavier processing demand: the CPU/GPU burden for decoding, rendering etc. is higher, and some editing suites or hardware may lack smooth compatibility.
- Sony’s support documentation warns that when using XAVC HS 4K (HEVC), you should use editing software compatible with HEVC and have a powerful system to maintain smooth performance. 
- Another caveat: some playback or sharing devices (smartphones, older hardware) may not support HEVC natively, reducing compatibility. 
2. Comparative Table: Performance, Workflow, and Trade‑Offs
|
Codec / Format Type |
Strengths / Advantages |
Weaknesses / Trade‑Offs |
Ideal Use Cases |
|---|---|---|---|
|
All‑Intra (S‑I, e.g. XAVC S‑I) |
Simplest editing workflow, no dependency frames, better for colour grading & effects |
Very large file sizes, high sustained data rate requirement |
High-end production (commercials, narrative, music videos) |
|
Long‑GOP (S / XAVC S style) |
Good balance of quality vs file size, manageable load on system |
More complex editing, sometimes artifacts in fast motion scenes |
Social media, documentary, general production |
|
HEVC / HS / compressed high efficiency |
Best compression efficiency (lower file size for same quality) |
Highest decoding load, potential compatibility problems |
Long continuous recordings, storage-limited shoots |
From user forums and real-world experience, many note that HS (HEVC) modes pack more visual detail with fewer bits, but are more punishing on a slower computer.  Users also caution that heavy compression may hide noise, which becomes visible on heavier grading. 
In one community discussion:
“HS = H.265 = Smaller file sizes / better image quality, due to the more efficient compression, but takes more computer power to edit the files because H.265 is more intensive.” 
Thus, the decision often comes down to your workflow, hardware, and storage constraints.
3. Colour Depth, Chroma Subsampling & Bitrate Considerations
When choosing formats, aside from compression type, parameters like bit depth (8‑bit vs 10‑bit) and chrominance subsampling (4:2:0 vs 4:2:2 / 4:4:4) are crucial for image fidelity and colour grading flexibility.
- All‑Intra formats commonly support 10‑bit 4:2:2 sampling for higher dynamic range and smoother gradients. Sony’s help guide lists, for instance, 60p 4K All‑Intra at 600 Mbps, 10‑bit 4:2:2. 
- Long‑GOP or HS modes may default to 4:2:0 or 8-bit in lower settings, especially for broadcast compatibility or simpler workflows. 
- Always check sustained bitrate requirements—a format may have a “max” bitrate, but the continuous data rate your card/system must handle is what matters.
Memory Card Solutions
Given the varying demands of these codecs (particularly the sustained write speed needed for All‑Intra and high bitrates), choosing the proper memory card is a critical component of a robust video workflow.
1. Required Characteristics
- Sustained Write Speed: Peak or burst speeds are insufficient; the card must sustain high throughput throughout the recording.
- Video Speed Class / V Rating: Cards with V60, V90, or Video Performance Guarantee (VPG) ratings are preferable for video.
- Interface & Format: Many high-end cameras support CFexpress Type A or Type B, as well as SD UHS-II / UHS-III.
- Balance & Compatibility: Use cards with matching performance in dual-slot systems to prevent bottlenecks.
- Card Reader / Interface: The card reader or interface (USB, PCIe, etc.) must support the card’s full bandwidth for efficient offload.
2. SD V90 vs CFexpress: Which to Choose?
SD cards with V90 rating are well-suited for many video tasks. However, at the highest bitrates (e.g. 600 Mbps All‑Intra) many professionals favour CFexpress Type A (or appropriate variant) because of superior sustained performance.
- According to ProGrade Digital, while SD V90 is excellent for RAW burst and some video uses, CFexpress Type A allows capturing at the highest bit rates offered by many cameras. 
- In a detailed comparison, SD V90 cards may reach physical limits in write throughput, whereas CFexpress formats often offer much higher sustained speeds and headroom. 
- Some users in forums decide in favour of V90 SD when they don’t require extreme bitrates or heavy post: e.g. “I go with a V90 UHS‑II card over CFExpress type A … CF cards are faster but for my shooting style (…) I prefer V90.” 
- But note: some camera features, such as slow-motion or maximum All-Intra bitrates, may be disabled or limited when using SD instead of CFexpress. 
Thus, the pragmatic approach is:
- Use CFexpress Type A for the highest bitrate / highest quality / pro workflows.
- Use SD V90 (UHS-II) as a backup, overflow or less-demanding codec choice when hardware or budget constraints exist.
3. Sample Memory Card Models
Here are a few recommended cards (CFexpress Type A and high-end SD V90) that are well-suited for video recording:
Here are a few highlights:
- Sony CFexpress Tough Type A — a robust, camera‑grade Type A card built for video workloads.
- Angelbird AV PRO CFexpress A — designed specifically for high‑bitrate video workflows.
- Lexar Professional CFexpress GOLD — high performance, balanced option in Type A space.
- Angelbird AV PRO V90 SD — V90‑rated SD card suitable for lower but still demanding video setups.
- SanDisk Extreme PRO UHS-II SD — widely trusted high-speed SD card, though not always sufficient for top-tier all intra modes.
- Delkin Power SDXC UHS-II — performance SD card suitable for many video tasks.
These exemplars illustrate the trade space: CFexpress gives you headroom for ultra‑high bitrate and future formats, while high-end SD (V90) cards can still deliver excellent performance in many real-world video capture scenarios.
Strategic Guidance & Recommendations
1. Match card capability to codec choice
If you plan to regularly shoot in All‑Intra or maximum bitrate modes, lean toward CFexpress Type A. If your workflow tolerates Long‑GOP or lower data rates, V90 SD cards may suffice.
2. Test sustained write performance before critical shoots
Use tools or sample video clips to verify that the card doesn’t drop frames or heat up under load.
3. Maintain firmware and software compatibility
Update the camera’s firmware and editing software (with HEVC support) to ensure stability.
4. Dual-slot balancing
In cameras with dual card slots, use identical or performance‑matched cards to avoid bottlenecks, especially in relay or simultaneous recording modes.
5. Plan for backups and offload speed
Use fast card readers that support full bandwidth of your cards; delays in offloading footage can cost time on set.
For video-capable still cameras, balancing codec choice, storage constraints, and post-production workflow is essential. All-Intra (S‑I) modes offer the smoothest editing pipeline at the cost of heavier storage needs. Long-GOP (S) modes deliver a balanced middle ground. High-efficiency (HS / HEVC) modes push compression further but demand more from editing hardware and may introduce compatibility constraints.
To support these codecs, memory card choice is mission-critical. CFexpress Type A is often the preferred solution for high bitrate video workflows, while SD V90 (UHS-II) cards can serve for less intensive or more budget-constrained scenarios. Selecting reliable, sustained-performance cards and matching them to your codec and hardware context ensures smooth, reliable video capture and post workflow.
