AVID VIDEO ENGINE PERFORMANCE TESTS IN PRO TOOLS 11 HD - PART 1
06-05-2014 11:00 AM
Aims and Objectives
Following on from my series of articles about problems with the Avid Video Engine in Pro-Tools 11.1.2, (especially with Blackmagic hardware on the Mac platform), Avid have been kind enough address most of the problems, but during my research and testing of the AJA video hardware, a number of questions came up. I have decided to make a series of videos of the tests so you can see and make some judgments as to the various performance parameters in the video chain.
The Post Production Video Chain
In the post production there are many places where variables can occur, and these are principally:
- The video codec itself (e.g. DnxHD, ProRes, or H264)
- The media wrapper or container (Quicktime or Avid MXF)
- The hardware used for playback i.e. AJA or Blackmagic hardware or desktop video
- Whether the video hardware and Pro-Tools are referenced to (the same) video references
- The video display technology itself (analogue tube, Plasma or LCD)
I am ignoring certain other potential variables such as disk drives (and assume that the AVE itself manages that) and the various settings in the displays themselves (i.e game mode etc).
My Video Post Set Up
I use Mac OSX and mostly PAL frame rates, so my tests will be principally confined to these, I will try and include a Blackmagic Mac Vs Blackmagic Windows comparison if I am able to organise this in time.
As previously described, my video set up uses composite video, due to my heritage with the Avid Mojo (vanilla version). I own a Blackmagic Decklinck Studio PCIe card, and I have a loan AJA IO XT thunderbolt box (which I want to buy). I will include some tests using the HD outputs of these cards with a Panasonic Viera LCD for comparison to the composite video.
Both the Blackmagic and the AJA equipment have video sync inputs and onboard hardware down conversion from HD to SD. I will be using this HD to SD down convert as it enables me to use a Sony analogue tube television, which has the least sync delay of most monitors (almost negligible) and which therefore allows me to remove one variable from the chain, the display.
The AVE Tests
These tests will not be exhaustively 100% scientific test with error calculations and a large sample size, covering of all variations of parameters etc as this would take days. The tests will be a quick and dirty indicator that will hopefully allow the reader to make some informed judgment as to what to look at themselves and the direction that they might want to go in, rather than just rely on industry hearsay (of which there is both way too much and of an ill informed nature).
I will also include some qualitative footage of performance, so that you can judge for yourself just how good the hardware solutions are at:
- Scrubbing video and audio together
- Jumping around the timeline randomly
- Using the nudge keys to frame advance
- Start and stop performance
The sync tests will be performed with a Syncheck 3 from Pharoah Editorial Inc. GET ONE - no post set up should be without one. See my review on the Syncheck 3.
Desktop video will be on my Slate Raven MTi – a state of the art 1920x1080 touch screen fed via DVI from my NVidia GeoForce 610 GFX card.
The Avid Video Engine Will Be Tested With The Following Solutions:
- Avid Mojo (composite not SDI) in Pro-tools 10 SD IO
- AJA IO XT a thunderbolt genlockable HD/SD IO in Pro tools 11
- Blackmagic Decklink Studio PCIe genlockable HD/SD IO in Pro tools 11
- Desktop video in Pro Tools 11
I will try and work through each solution in a logical manner, so that they can be compared in various common scenarios.
The tests will commence in part 2….
There are currently 1 users browsing this thread. (0 members and 1 guests)