Blog

Which is best?   Depends on the camera, scene and shot dynamics I guess.  The same point is queried at the following thread:

• http://www.dvinfo.net/forum/sony-xdcam-ex-cinealta/238593-steady-shot-smoothcam-both-what-2.html#post1551809

Some general advice from a computer-post-savvy author: definitely use the camera’s SteadyShot:

• http://www.gcvcc.com/March%202008%20%20Camera%20Techniques%20No2.pdf

Limitations of post

• Stabilization necessitates motion estimation and image reconstruction, which are extremely CPU-heavy, hence really slow to execute.
• Most stabilization apps (in post) can’t currently cope with motion-blurred edges or parallax effects (though both should be possible in principle, by deconvolution and 3D modelling both informed from multiple frames).
• For rolling-shutter-ed footage (e.g. CMOS sensors as in Sony Exmor as in Sony XDCAM-EX e.g. EX1 & EX3), there exist options to reduce the effect (don’t expect perfection, but may suffice):
  • Sony’s ClipBrowser 2.6 can remove this http://www.dvinfo.net/forum/sony-xdcam-ex-cinealta/467511-clipbrowser-takes-out-rolling-shutter.html.
  • Gunnar Thalin’s Deshaker plugin for VirtualDub can compensate for this.

My experiences:

• Stabilizing Tools:
  • Gunnar Thalin’s Deshaker works really well.  And it is multi-threaded, really speeds up the process.  The author says it is more intended for handheld pans etc. than fast-shaking shots from vehicles etc. (but has nevertheless seen good results in such situations).
    • The author says [http://forum.doom9.org/showthread.php?p=1418923] to try “to stabilize only on the most distant parts in the frames, since the moving inwards-effect is less there”.  And “you should probably increase the value for [discard motion of blocks that move > X pixels in wrong direction]. That’s to allow the blocks to move “freely” a little, since Deshaker can’t handle the “moving inwards”-effect.
    • Possibly equally applicable to other smooth/stabilize/deshake tools ?
  • Boris’s Optical/Motion Stabilizer (in Boris Red 4.3.3 on XP) is only single-threaded and I find it slower, clunkier and less intuitive than Deshaker.  Has a Smooth mode, which is like the others here, as well as a Stabilize mode (try to keep frame static, no good for motion then).  The other tools can be configured to do the same thing.
  • Mercalli in Sony Vegas has no mode for 720p50 but otherwise is pretty good and very intuitive and configurable.
  • FCP’s SmoothCam Effect worked best for a challenging clip for wobbly-hand-held camera tracking close past an object (a Formula-1 car) hence huge degree of moving-inwards effect.  The default settings worked straight away.  The result quality was way above that of the other tools.  On the other hand sometimes it’s not the best (sorry, forgot the exact situation).
• Cameras & Shots:
  • Historically, using a TRV33 DV HandyCam indoors (hence low-light hence long shutter time):
    • Way back in the past, using a (now ancient) TRV33 DV handy-cam (which has huge sensor margin i.e. spare pixels), when I shot big zooms to lecture audience individuals (e.g. question-time) I had the camera’s steady-shot (digital, not mirror) enabled  and used Gunnar Thalin’s Deshaker (VirtualDub plugin) also.  The result was astoundingly steady.
    • The same arrangement worked OK with hand or shoulder mounted cam for walk-throughs past nearby objects (e.g. walls, people, furniture).
    • An attempt to do the same thing without steady-shot enabled on the camera resulted in seriously motion-blurred edges.
  • Now, using a Sony EX3:
    • With camera Steady-Shot set to Medium, hand-held pans and motion past nearby objects seem to acquire a positional instability, as if the camera feedback mechanism needs greater damping. Maybe the camera’s internal mirror “suspension” has to be tighter (than the TRV33 digital equivalent) because it lacks the generous pixels margin of the TRV33?  or maybe something to do with the mirror’s inertia?  Or (real-time-constrained) processing-power?
      • Experimentation is needed with the camera’s other SteadyShot modes (High, Low).
      • In the absence of more generous sensor pixel margins, I wish it could be loosened-up e.g. to allow black borders (to crop in post) so as to permit smoother rides overall.

Problem:

• In Windows I export from Sony Vegas to AVI (CineForm).  In OS X I read the file into FCP and apply the SmoothCam effect, then export to ProRes.  In Windows, Sony Vegas, I replace the original file with the smoothed one.  The levels/gamma are wrong.

Solution (Search):

• Sony Vegas forum http://www.sonycreativesoftware.com/forums/ShowMessage.asp?ForumID=4&MessageID=718371
  • Use DNxHD
    •  Couple of tips re DNxHD:  709 color level assumes 16-235, and RGB assumes 0-255.
  • Force it back again: www.kenstone.net/fcp_homepage/gamma_mac_pc.html
    • But this presumably implies getting re-quantized twice (the roundtrip issue and the forcing), which for 8-bit footage I imagine could reduce the quality (banding).
• Uncertainties
  • Where and how does this gamma get applied?  In FCP I didn’t (knowingly) alter the levels (eg until it looked right), I just applied the SmoothCam filter.  So I guess it would look wrong on the (pre-SnowLeopard) Mac but I wouldn’t care.  Wouldn’t FCP then export back whatever it got but smoothed?  This one is really confusing.    Experiments needed (when I get time…) I guess.

Reading book “Final Cut Pro Workflows” by Osder & Carman, 2008.  On page 284 it relays advice that it is best to put Project Files [.fcp] on a separate drive to the Media Drive (e.g. Media Drive= XSAN), due to:

• Safety - not all on one drive
• Avoid fragmenting the media drive (project files, cache and to a lesser extent render files) are written often (transient files?)

I’m not immediately convinced by these arguments:

• Media drive should be backed-up in any case.  I use (slower) USB drives and file-synch.
• Fragmentation may be an old-school issue, less of an issue on Mac’s HFS +.
  • Example: article at [http://www.macmousecalls.com/files/1f6efbb25f58d77390499af1f984d2c7-33.html] linking-on to [http://www.kernelthread.com/mac/apme/fragmentation]
  • Example: Larry Jordan, “Final Cut Pro Power Skills” 2010, page 34, Chapter 2, “Defragmenting Your Hard Drive”: Once popular, this technique is no longer recommended.

How to view degree of fragmentation on an HFS volume:

• [http://osxbook.com/book/bonus/chapter12/hfsdebug/fragmentation.html]
  • Command-line app to report a variety of storage-volume statistics, including fragmentation.
  • After download, can check the sha1 checksum, but this is of the executable, not the download itself ([.dmg] file).  The ’sha1′ command is inbuilt to Mac OS, as: [/usr/bin/openssl sha1].  Note the last character of ‘openssl’ is a small ‘L’ niot a ‘1′.

• General
  • Show advanced tools [No] -> Yes
• Video
  • Import HD video as: [Large 960×540] -> Full - Original Size

No way I can see to define the iMovie work-area folder, where media gets imported to for example.  Maybe there’s no choice over this, which would be a shame (don’t want movies filling up my system drive).  In my case the iMovie work-area was:

• /Users/davidesp/Movies
  • iMovie Events.localized/
    • New Event/
      • iMovie Data
      • iMovie Cache/
      • iMovie Thumbnails/
      • <a source media file>
  • iMovie Projects/
    • My First Project.rcproject

Great article by Studio Daily at http://www.studiodaily.com/main/work/In-Three-on-the-Workflow-Behind-3D-Conversions_12083.html, on the 2D⇒3D conversion company In-Three http://www.in-three.com/.  from this article I learnt that:

• Some new movies, not just legacy ones, are being converted from 2D to 3D (stereo).  This step is being planned as part of production.  Don’t know why they can’t just shoot it in stereo (cost? maturity? conservatism?) but that’s how it is.
• The method: a tech & manual rotoscoping pipeline (production line) where images are masked to create layers and artistic judgement is applied to the appearances of individual objects.  As one would imagine, no simple “magic solution”.  However beyond those basics they have their own patented 2D⇒3D inference algorithms operating on individual objects even at sub-pixel level.
• Not quick or cheap: “for a 100-minute or 120-minute 2D-to-3D conversion, you would need about 300 to 400 artists phasing and out of production over about four to six months.”.  Clash of the Titans was so-processed in under half that time - possibly explaining some negative press (mentioned in the article) regarding the quality of its 3D.

The interviewee in the article was from In-Three.  Their website http://www.in-three.com/ explains:

• Dimensionalization is a method developed by In-Three of converting 2D content to stereoscopic 3D content.
• There are various approaches to creating 3D content: capturing 3D using dual camera rigs, rendering 3D using dual “virtual” camera rigs within a computer graphics environment, and creating 3D by converting 2D content with processes such as Dimensionalization.
• Dimensionalization is trademarked because it describes a patented process which gives the unique, depth, shape and perspective to each individual object on a pixel or even sub-pixel level. Throughout our process, there are a multitude of “special and unique techniques” our experienced stereo team has and continues to develop, so that you can be confident that we bring the tools and the skill to any conversion project.
• The Dimensionalization process is covered by a number of U.S. patents. These patents make In-Three a leader in the development of intellectual property surrounding the conversion of two-dimensional films to stereoscopic experiences.

The following has a nice explanation and diagram of the arrangement, showing Final Cut Server being used to interface to both online and archive material.

• http://studiodaily.firstlightera.com/ERA/ResourceHandler.ashx?73070b2c-c5ba-4db8-a4a1-616a7da709ef;SB00016.pdf

For Mac OS:

• Disk Inventory X

For Windows:

• WinDirStat
• FolderSize

They are both pretty similar, in each case displaying filespace usage via a tree map looking like a patchwork of multicoloured PVC, each colour representing a different type of file (audio, video, application, document etc.).  Their advantage over traditional browser trees is you can see all the largest files and folders simultaneously (as a plan-view).  Tree maps (treemaps) are explained at http://w3.win.tue.nl/nl/onderzoek/onderzoek_informatica/visualization/sequoiaview/about_treemaps/ - they are formed by subdividing in alternate dimensions (horizontal/vertical), each time in proportion to relative size of item, be it folder or file.  A variation on this, employed by the above tools, is a cushion treemap [http://www.win.tue.nl/~vanwijk/ctm.pdf], where shading reveals the directory structure.  A further variation is the squarified treemap [http://www.win.tue.nl/~vanwijk/stm.pdf], where subdivision and grouping attempt (no guarantee of success) to make the rectangles as square as possible. (more…)

Problem:

• MacBook Pro, when in Boot Camp Windows 7, gets uncomfortably hot (in its heatsink areas, the bar above keyboard and also the chassis underside).  In comparison, it runs pretty cool (temperature) under Mac OS.

Reason:

• No fan control when in Boot Camp Windows mode.

Popular Solution (does not imply any recommendation):

• When in Boot Camp windows, use Lubbo’s Fan Control [http://sourceforge.net/projects/lubbofancontrol]

Some good links (as of 2010-03-27):

• http://www.almico.com/sfdownload.php
• http://sourceforge.net/projects/lubbofancontrol
• http://www.sevenforums.com/general-discussion/17177-fan-control-macbook-pro-15-inch-under-windows-7-a.html
• http://www.lokiman.com/blog/2009/12/13/fan-control-for-macbook-pro-under-snow-leopard-or-windows-7-and-more

Experiences:

• Just copied the contents of the zip file to  [C:\Program Files (x86)\Lubbo’s Fan Control] and ran it from there.  It has files [inpout32.dll] and [inpoutx64.dll].
• Ran it but it gave error messages:
  • (The) Following process(es) is/are using SMC:
    • kbdmgr
  • It’s not an error, but Lubbo’s Fan Control cannot share apple SMC access.
  • Do you want to kill it/them?
  • NO = the system may freeze.  Try only if you are running BootCamp 3.1
  • YES = the incompatible process(es) will bekilled and the program will start.
    • But I have read elsewhere that this means function keys won’t then work (for that session).
    • Didn’t work - it said “It was not possible to load IO driver.  Retry?” and “(May be better to press CANCEL and reopen the program)”
  • CANCEL = this program will exit
• Someone else had the same problem but found a solution that appeared to work for them:
  • “…upon updating to .NET 4 and running with “administrator control” or whatever, it appears to be working. [http://sourceforge.net/projects/lubbofancontrol/forums/forum/970570/topic/3507393]
  • Further confirmation that (for Windows 7),  .NET v4 is needed [http://sourceforge.net/projects/lubbofancontrol/forums/forum/970570/topic/3444415]
    • “…I have the same problem with the error about the IO drivers. When I try to install .NET framework 3.5 using the setup file, the screen just flashes and nothing happens. I have a feeling win7 did not install it.  This is in win7 64 bit professional”
    • “The problem was fixed after I installed .NET framework 4.0 Beta for windows “
• So it seems that to get Lubbo’s utility working I have to:
  • Kill a system process
  • Install a version of .NET that is not yet officially supported.
    • I’ll write a separate blog post about .NET
• Life on the edge, huh?

The Gibb Effect is an MPEG compression artefact. It is a blurring of the outline of sharp objects, with inappropriately-coloured pixels appearing around the outline of the object. The commonest area in which the Gibb Effect is seen is during end credits, where insufficient bits have been allocated to compressing the data. Another name for this artefact is mosquito noise.[http://www.michaeldvd.com/Articles/VideoArtefacts/VideoArtefactsGibbsEffect.html]

 Mobile eSATA (via ExpressCard) for MacBook Pro.Sonnet Fusion F2.  Up to 1GB (when configured as Striped i.e. RAID0).  Sustained Read/Write of 126MBps=1008Mbps.

• FireWire 800 bitrates (for comparison):
  • Sustained 100 MBps (per bus) = 800 Mbps for a RAID0 drive [http://www.lacie.com/download/more/WhitePaper_FireWire_800.pdf].
  • Not much less than the eSATA drive.
• Video Bitrates [http://www.march21.nl/learningzone/files/7a3bdb9c2f07c7e76be79ba6c3b34bc5-43.html]
  • ProRes HD 1920×1080 @ 50p: Just under 35 MB/s.  126/35 = 3.6 such streams.
  • Prores HD 1920×1080 @ 50i: 26MB/s.  126/26 = 4.8 such streams.

Links:

• http://www.macworld.co.uk/digitallifestyle/reviews/index.cfm?RSS&ReviewID=3203
• http://www.macvideo.tv/editing/reviews/index.cfm?reviewid=102494 
  • Includes a video of usage.

It connects via two eSATA cables to ExpressCard adaptor and also via a FireWire connector purely for the power (no bandwidth).  The intention is that the FW bandwidth is still free for use by other devices e.g. “AJA’s Io external capture and effects box – which requires all of the FireWire bandwidth to itself”. [http://www.macworld.co.uk/digitallifestyle/reviews/index.cfm?RSS&ReviewID=3203]

• Note - for Sony EX1 and EX3 users the Fusion F2 uses th Express 34 slot on the MacBook Pro, meaning one would need to transfer SxS data to either a FireWire or USB drive and then across to the Fusion 2.

Thinks: It works as Software RAID for the Mac.  Is there any practical way to also use it from Windows?