zgaetano/rainbow-dragon
1
0
Fork 0
Code Issues 1 Pull requests Projects Releases Packages Wiki Activity Actions

Competitive research: Livegrade 7 / Silverstack Lab feature gaps vs. implementation plan #1

New issue
Open
opened 2026-07-10 22:30:28 -04:00 by zgaetano · 0 comments
zgaetano commented 2026-07-10 22:30:28 -04:00
Owner
Copy link

Full write-up: docs/research/livegrade-silverstack-research.md

Deep-dive on Pomfort's Livegrade 7 (on-set live grading) and Silverstack Lab (DIT offload/dailies) feature sets, checked against docs/plans/001-implementation-plan.md. Key gaps/decisions surfaced:

  1. Slot/routing abstraction — Livegrade's core architecture is input/output/library "slots" with multi-view grids and per-slot filter chains independent of the grade stack. Forge's plan only has a flat CameraDriver per camera. Decide if we need this generality or can stay single-camera-focused.
  2. Checksum algorithms — Silverstack Lab supports XXH64/MD5/SHA1/C4ID/XXH3/XXH128. Forge Task 7.1 only scopes xxHash64 + MD5. Decide on C4ID/SHA1/XXH3/128 scope.
  3. Copy/verify scheduling — Silverstack runs copy and verify as separate, resumable jobs with per-drive concurrency caps (small DAG scheduler). Forge's ForgeOffload (Phase 7) is currently a flat sequential pipeline. Likely the biggest architectural gap for "Lab"-tier parity.
  4. RAW development — Silverstack Lab debayers ARRIRAW/REDCODE/BRAW/etc. in-app for playback, stills, and transcoding. Forge's plan has no RAW decode path (only ProRes/DPX/PNG via ForgeVideo). Needs an explicit scope decision.
  5. Look-matching bridge — Pomfort's core cross-product value prop: grade live on set → same look auto-applied at ingest by clip-name/metadata match. Forge's shared ForgeGrade module makes this feasible; worth naming as an explicit feature rather than leaving implicit.
  6. Interchange breadth — Import: QTAKE/MovieSlate/Drylab/ZoeLog. Export: Resolve/.ccc, Avid/Scratch .ale, Premiere XML, Drylab, COPRA. Forge's Phase 3 only covers CDL/CCC/CUBE/ALE/CSV/EDL. ALE-with-color and FCP7 XML are the two most commonly requested by editorial.
  7. Hardware control — Both products support Tangent (Forge has this via Phase 5 ForgePanel). Neither supports Stream Deck or MIDI, which are cheap high-visibility additions since DITs often already own a Stream Deck.
  8. Feature prioritization signal — Livegrade gates LED wall control, internet streaming, chroma keyer, and 50-input I/O behind paid extension modules, implying Pomfort treats these as niche/advanced. Useful signal for sequencing anything beyond Phase 8.

No action taken on the plan yet — this issue is for triage/decision on which gaps to fold into future phases.

Full write-up: `docs/research/livegrade-silverstack-research.md` Deep-dive on Pomfort's Livegrade 7 (on-set live grading) and Silverstack Lab (DIT offload/dailies) feature sets, checked against `docs/plans/001-implementation-plan.md`. Key gaps/decisions surfaced: 1. **Slot/routing abstraction** — Livegrade's core architecture is input/output/library "slots" with multi-view grids and per-slot filter chains independent of the grade stack. Forge's plan only has a flat `CameraDriver` per camera. Decide if we need this generality or can stay single-camera-focused. 2. **Checksum algorithms** — Silverstack Lab supports XXH64/MD5/SHA1/C4ID/XXH3/XXH128. Forge Task 7.1 only scopes xxHash64 + MD5. Decide on C4ID/SHA1/XXH3/128 scope. 3. **Copy/verify scheduling** — Silverstack runs copy and verify as separate, resumable jobs with per-drive concurrency caps (small DAG scheduler). Forge's `ForgeOffload` (Phase 7) is currently a flat sequential pipeline. Likely the biggest architectural gap for "Lab"-tier parity. 4. **RAW development** — Silverstack Lab debayers ARRIRAW/REDCODE/BRAW/etc. in-app for playback, stills, and transcoding. Forge's plan has no RAW decode path (only ProRes/DPX/PNG via `ForgeVideo`). Needs an explicit scope decision. 5. **Look-matching bridge** — Pomfort's core cross-product value prop: grade live on set → same look auto-applied at ingest by clip-name/metadata match. Forge's shared `ForgeGrade` module makes this feasible; worth naming as an explicit feature rather than leaving implicit. 6. **Interchange breadth** — Import: QTAKE/MovieSlate/Drylab/ZoeLog. Export: Resolve/.ccc, Avid/Scratch .ale, Premiere XML, Drylab, COPRA. Forge's Phase 3 only covers CDL/CCC/CUBE/ALE/CSV/EDL. ALE-with-color and FCP7 XML are the two most commonly requested by editorial. 7. **Hardware control** — Both products support Tangent (Forge has this via Phase 5 `ForgePanel`). Neither supports Stream Deck or MIDI, which are cheap high-visibility additions since DITs often already own a Stream Deck. 8. **Feature prioritization signal** — Livegrade gates LED wall control, internet streaming, chroma keyer, and 50-input I/O behind paid extension modules, implying Pomfort treats these as niche/advanced. Useful signal for sequencing anything beyond Phase 8. No action taken on the plan yet — this issue is for triage/decision on which gaps to fold into future phases.
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
master
No results found.
Labels
Clear labels
No items
No labels
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: zgaetano/rainbow-dragon#1
No description provided.
Delete branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?