FAQ
Whether or not to enable deinterlacing
Using Persistent streaming to automatically start streaming when the system is booted
Using Tags to create unique file names and to generate metadata for reports
User defined, or custom Tags
Unicast vs Multicast
Help setting up a Windows Media Multicast
MPEG1 and MPEG2 Advanced Settings
Transcoding files created on AVID or Media 100 NLE systems
Renaming files after the encode process using the post-encode Command function
Using Microsoft Windows’ Remote Desktop to control Stream
Changing your settings so that file name extensions are shown
How to Calculate the drive space required for an uncompressed video capture
Whether or not to enable deinterlacing
In order to decide if deinterlacing is required or not you need to determine where the result video will be played back. If the playback is intended for computer screens, you will need to deinterlace. If the playback is a television or an NTSC/PAL monitor, then you need to leave the interlaced fields intact.
Interlaced video played back on a computer screen will reveal the interlacing in high motion areas - it will appear as a stepping of horizontal lines where ever there is fast motion content.
Deinterlaced video played back on a TV or studio monitor will appear as stuttering where there is high motion content.
Depending on the type of encode you are doing, there may be two places where deinterlacing can be enabled:
The first type of deinterlacing is the hardware deinterlacer available on the DRC capture board. This should be used for deinterlacing whenever possible to achieve the highest quality result. Hardware deinterlacing is only available when capturing from a live source. For the DRC-500 to DRC-2600 boards, deinterlacing is available from the Video Process tab. For the DRC-5550 or DRC-5650 board (the HD board), deinterlacing is available by using the “format conversion” option in the source profile and choosing “progressive” instead of “interlaced”.
The second type of deinterlacing is found in the profile creation for many wrapper/codec types. These tools are generally provided by the codec supplier. They are a software deinterlacing and this is typically used when transcoding interlaced content from a file source. Software deinterlacing is not as good as hardware deinterlacing and requires more CPU cycles to accomplish.
In either case, do not enable both hardware and software deinterlacers. Doing so will add interlace artifacts and use unnecessary CPU resources. There are some examples of software and hardware deinterlacing in the demo room on our web page if your interested in pursuing the visual difference in quality of the results.
Using Persistent streaming to automatically start streaming when the system is booted
Stream supports a mode that allows for persistent encoding from live video to streaming (broadcast) output. In this mode Stream will always begin streaming as soon as the system is booted. This is important to allow for recovery from power loss during critical encoding sessions. This will allow Stream to return to live streaming mode almost as fast as the system can boot (usually about 1 minute).
To use Persistent Streaming mode you must have previously saved the project you wish to use.
To configure Persistent streaming mode launch Stream and from the main menu choose Tools > Persistent Streaming.
You can have one persistent streaming project for each real-time video processing channel in your Stream system. Stream 500, 1000 and 1500 models support one channel of video processing, Stream 2000 and 2500 support two channels of video processing.
Using Tags to create unique file names and to generate metadata for reports
Tags are quite useful for creating unique, useful filenames and for automatically entering information into the metadata fields of each codec.
Filename example: d:\output\%BN%-%R%-%T_%.wmv might generate the filename d:\output\MyProject-320x240-4_10_51_PM.wmv. Encoding a second file without changing any settings within the Stream app would generate a unique filename encoded with the current time. By using a combination tags such as %R% (resolution), %BR% (bitrate), and %T_% (time), it is possible to create a default output filename, which will always generate unique filenames.
User defined, or custom Tags
Custom tags can be set manually from the Stream interface. From the main menu choose Options > Preferences and go to the Tags tab. There is an Add button to the right of the window. This will open a window that allows you to enter the name and value for a new custom tag.
As well, custom tags are stored in C:\Program Files\Digital Rapids\Stream\tags.xml
A sample tags.xml file is below with two custom tags added for a hypothetical situation where both "Policy number" and "Claim Number" are the required custom tags.
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<KLAMATH type="tags" version="1">
<TAGS>
<TAG name="policy_num" value="67890" />
<TAG name="claim_num" value="12345" />
</TAGS>
</KLAMATH>
In some situtations altering the xml file programatically prior to an encode may give you the workflow you need for a job.
Unicast vs Multicast
Unicast is a point to point protocol requiring an arbitration between the broadcast point (streaming server) and the endpoint (the user). It allows the user to be more interactive with the stream as there is a 1 to 1 correlation (connection) between the user and the broadcast point.
Unicast requires that every user gets a unique stream of video data with the overall constraining factor being total available bandwidth. If 10 users want to watch a 1Mb stream, it will require 10Mb of bandwidth.
Multicast in contrast is a 1 to many protocol requiring cooperative negotiation all along the path between the broadcast point and the users. Because it is a protocol unto itself (as it pertains to routing) it requires the intermediate hops (routers) to be aware and configured to allow the data to pass through right up to the end user.
Multicasting allows a much larger number of clients than Unicast because of its almost pure broadcast nature. The same stream can be sent to many users, the limitation being the number of users, not the overall bandwidth.
Digital Rapids products are media encoders. It can put the compressed, multiplexed data out to file or to a UDP or RTP stream. This encoding step is prior to the unicast/multicast decision.
A typical unicast workflow wuold be:
Unicast: StreamZ (unicast)->Media Server (unicast)->Clients
The limit when using unicast is the size of the data pipe between the Media Server and the clients.
A typical multicast workflow would be:
Multicast: StreamZ (unicast)->Media Server (multicast)->Clients
The limit when using multicast is the number of client IP address the Media Sercer can process in real-time.
You can run a Digital Rapids board in a multicast server. However, this will limit the total number of connections that the multicast server can service (nothing is free).
Help setting up a Windows Media Multicast
For those having difficulty setting up for Windows Media Multicast, click on the following link to read a step by step tutorial: How_to_WMV_Multicast.pdf
MPEG1 and MPEG2 Advanced Settings
For those that require more information on the Main Concept MPEG codec advanced settings, a document on this is provided within Stream software. Acrobat reader needs to be installed. Once done, you can get to the document at the top menu pulldowns, Codec Profile ->Create ->MPEG Main Concept ->MPEG2 ->Advanced Settings ->Help
If you want to access that document and all the others that were provided with Stream 2.4 installation, browse to C:\Program Files\Digital Rapids\Stream\documents\
Transcoding files created on AVID or Media 100 NLE systems
In order to transcode a file that was created on an AVID or Media 100 NLE system it is first necessary to download and install the AVID Codec or the Media 100 codec.
Once installed, you should be able to play the file in the Quicktime player. If the plays in the player, then the file can be used as a Quicktime input source and transcoded to the format of choice.
Avid/M100 files can be made by creating a new Quicktime profile and selecting the appropriate codec.
Note that many of these types of codecs were not intended to provide for real time playback of the encoded file. Check with the codec vendor regarding any playback issues using this codec.
Renaming files after the encode process using the post-encode Command function
In some cases the default output file name extensions may not be what is required. For example, if you are encoding files for use in DVD Impression, this program will not accept a *.mpg or *.mp2 file extension, it requires .mpv or .mpa.
These files can automatically be renamed after the encode process using the Post Process External Command function in StreamPro software.
First, create a RenameForImpression.bat file in your output media folder and edit the contents to read:
rename *.mpg to *.mpv
rename *.mp2 to *.mpa
On the output tab of StreamPro software select Command and in the command box, paste the following including quotations:
"C:\Program Files\Digital Rapids\Stream\OutputMedia\RenameForImpression.bat"
Using Microsoft Windows’ Remote Desktop to control Stream
To obtain control of the DRC hardware you must be the first person logged in to the system. To obtain control of the console, logging off any other previous users:
Select Start > Run and in the Open box type: mstsc –v:server.ip.address –console
(where server.ip.address is replaced by the one used by the Stream system).
Changing your settings so that file name extensions are shown
To change your system settings so that they do not hide file name extensions:
From the desktop, double click on My Computer to open the Windows Explorer.
From the menu select Tools > Folder Options.
Choose the “View” tab. In the Advanced Settings section scroll down the list to “Hidden files and Folders” and choose the “Show hidden files and folders option”. Under that option you will see a check box for “Hide extensions for known file types”. Make sure that check box is clear (not selected).
Click the Apply button to save the settings. Click Ok to close the window.
How to Calculate the drive space required for an uncompressed video capture
To access a web page that will do the calculation for you, for any frame size or frame rate, click on bandwidth_calculator.html
For uncompressed 8-bit video, the number of bits/second is:
width x height x 2 samples (one for luma, one for chroma) x 8 bits per sample x frame rate
For NTSC: 720 x 480 x 2 x 8 x 29.97 = 165,722,112 bits/second
To convert from bits/second to MegaBytes/Second (MB/s):
165,722,112 bits / 8 bits/Byte = 20,715,265.25 Bytes/s
20,715,265.25 Bytes/s / (1024 kB/B x 1024 MB/kB) = 19.76 MB/s
So for one hour of uncompressed NTSC video:
19.76 MB/s x 3600 seconds/hour = 71,136 MB/hour = 69.5 GB/hour
|