I set up some tests using Sapphire Math_Ops set to subtract.
Test 1: XDCAM35 versus DNxHD 100 : very little difference, not easily seen on video monitor. Spikes on the scope are in the 1-2 IRE range.
Test 2 & 3: XDCAM35/ DNxHD 100 versus DNxHD 145 : edge detail differences are easily seen; 20+ IRE.
Test 4: DNxHD 145 versusDNxHD 220 : less difference than Test 1.
The primary content is talking head interviews in a home, and the broll is low-action household activity.
The quality differences are seen on edges of plates and jewelry. Since my client is a jewelry company, we'll be using the DNxHD145.
Thanks for everyone's input.
Curtis Nichols
Señor Editor
PCS Production Co.
Irving, Tx.
------------------
________________________________
From: Greg Huson
Subject: Re: [Avid-L2] 100 v 145
I'm not sure I agree that 100 is 'good enough.' XDCam35 is NOT 'decimated raster, (1440x1080), but rather full raster (1920x1080.) Seems like a pretty serious transcode.
Why not just use XDCam35? Unless you're doing lots of layering, or your CPU is a dog, MXF-wrapped XDCam35 should work well, and you're not really transcoding at all. Yes, XDCam is theoretically not necessarily a good editing format, but we've found it works quite well when re-wrapped and managed as an Avid-native file. I would, definitely, NOT leave it in AMA, but either 'transcode' or import to XDCam35 MXF.
Try it- if you don't like it blame me privately and refuse to ever work with me again.
Probably best NOT to set your renders to XDCam35, though, but rather use 145 (or better.)
[Non-text portions of this message have been removed]
No comments:
Post a Comment