MLP Manual.book

Downmix Parameters25 7 Channel Assignment 26 Why Use Groups?<br><br> 26 Channel Assignments 27 Assignment Options 27 8 Encode Section 28 Encode Start and End Time 29 iii The Encode Button 29 The Stop Button 29 10 Playback Section 30 Source/Destination Select 30 Transport Controls 31 Current Location 31 11 Verification/Playback Section 32 Verify Select 32 Write Decoded File Select 33 Playback Select 33 MLP Logs 33 12 ReBit (Bit-Depth Reduction) 34 Why Is Bit-Depth Reduction Needed? 34 Why Is SurCode MLP 9s ReBit Needed? 35 How ReBit Works 36 The ReBit Section 37 Manual Bit-Depth Reduction 37 Automatic Bit-Depth Reduction 38 ReBit Step Order 38 Time Savings 39 Worst-Case Scenario 40 13 Appendix A - Tables41 Compression Amounts 41 DVD-A Playing Time 41 Channel Assignments 42 iv TECHNICAL SUPPORT: Please Read This Manual First!<br><br> tech@minnetonkaaudio.com Minnetonka Audio Software, Inc. 17113 Minnetonka Blvd. Suite 300 Minnetonka, MN 55345 PH 952-449-6481 Hours: 10:00AM - 6:00PM CST Mon.-Fri.<br><br> FAX 952-449-0318 www.minnetonkaaudio.com Introduction to SurCode MLP 1 1 Introduction to SurCode MLP Welcome to SurCode MLP 3 the best way to encode high- resolution audio files for the DVD-Audio format. SurCode MLP utilizes licensed encoding algorithms from Meridian Inc. 3 MLP stands for Meridian Lossless Packing, which is an encoding system designed to compress high-resolution audio files with bit-for-bit accuracy.<br><br> Unlike the perceptual, or lossy audio data reduction systems used in DVD-Video authoring, MLP guarantees not to alter the final decoded signal in any way. Instead, it simply packs the audio data more efficiently into a smaller data rate. SurCode MLP is an intuitive Windows-based program that simplifies the steps involved in encoding a set of one or more audio files, while automatically verifying the accuracy of the resulting MLP file.<br><br> While this manual covers the wide number of choices and options built into SurCode MLP, please note that simply entering soundfiles into their respective channels and hitting the cEncode d button is all that is neccessary for most Introduction to SurCode MLP 2 productions, as the default settings in SurCode MLP are based on industry standards and norms (see the cGetting Started d chapter). 1.1 Why Encode? MLP encoding has two primary benefits: 1.<br><br> It minimizes the size of the compressed file, allowing a larger amount of audio to be stored on the DVD-A disc. 2. It can reduce the maximum instantaneous peak data rate.<br><br> This second benefit is very important for DVD-Audio, which has an upper limit of 9.6Mbps (megabits per second). For example, six channels of 24-bit, 96kHz Linear PCM (Pulse Code Modulation) audio has a data rate of 13.82Mbps, which is much greater than the data throughput of the DVD-Audio system. This high data rate would also limit the playing time to approximately 45 minutes.<br><br> MLP encoding can reduce the worst-case data rate to 9.6 Mbps, extending the playing time to the industry standard of 74 minutes, as well as allowing bit-accurate high-resolution six-channel audio. 1.2 How It Works Because MLP was originally designed with consumer applications in mind, one of its design aims was that any complexity must be in the encoder instead of the decoder. The design of the decoder ensures that it will remain lossless regardless of hardware platform or processor implementation.<br><br> In addition, the decoder is driven by Introduction to SurCode MLP 3 information in the bitstream, allowing improved versions of the encoder to be developed without becoming incompatible with the installed base of decoders. MLP encoding uses a combination of three methods to reduce the data rate: 1. Lossless matrixing is used to reduce the correlation between channels.<br><br> 2. Lossless waveform prediction is used to reduce the inter-sample correlation. 3.<br><br> Entropy coding is used to reduce the data rate by effi- ciently encoding the most likely occurring values in the audio data. SurCode MLP does not make any assumptions about the assignment of channels, or the correlation between channels, but takes advantage of whatever redundancy is in the overall signal to encode the data using the smallest possible bandwidth. In addition to these procedures for reducing the data rate, MLP uses stream buffering to reduce the variations in the transmitted data rates, and absorb transients that are hard to compress, in order to assure a maximum instantaneous peak data rate.<br><br> The buffer allows the peak data rate to be minimized for virtually all practical audio data. If the audio data cannot be compressed within the specified peak data rate, SurCode MLP will signal an error. The pro- ducer can then use one or more options for reducing the data rate, or reducing the total space used by the recording, such as Minnetonka 9s cReBit d, described next.<br><br> Introduction to SurCode MLP 4 1.3 ReBit Minnetonka Audio Software has implimented a unique time-saving feature in SurCode MLP called cReBit d that enables users, in exteme cases, to avoid the necessity of re- mastering their material in order to pass through the MLP encoding process (see the cReBit d chapter). 1.4 Compression Amounts See Table 1 in cAppendix A - Tables d. 1.5 DVD-A Playing Time See Table 2 in cAppendix A- Tables d.<br><br> 1.6 Two-channel Downmix Content providers will often want to make a two-channel version of a multi-channel audio stream available on a DVD-A disc, for consumers who only have two-channel playback capability. SurCode MLP provides an elegant and simple solution to providing a two-channel downmix. The encoder includes lossless matrixing, which can encode a two-channel downmix as a linear combination of the multi-channel mix and encode this alongside the multi-channel version on the DVD-A disc (see the cDownmix d chapter).<br><br> 1.7 Error Detection MLP also has powerful built-in error detection that allows rapid recovery from bit-stream errors, and prevents any erroneous noises, clicks, or pops following data errors. Introduction to SurCode MLP 5 1.8 Channel Assignment DVD-Audio supports up to six channels in one of 21 combinations of channel assignments (see Table 3 in cAppendix A- Tables d). The cChannel Assignment d combinations place the channels into two groups, referred to as Group 1 and Group 2.<br><br> The significance of the group is that the sample-rates can be specified independently for each group. However, the sample-rates within each group must be the same, and the Group 2 sample-rate must be an even multiple of Group 1 (see the cChannel Assignment d chapter). The DVD-Audio channel assignments 8 to 12 are almost identical to the channel assignments 13 to 17, and differ only in the way the channels are grouped.<br><br> 1.9 File Types SurCode MLP processes data from one or more input files, with WAV and AIFF audio file formats directly supported. Note: SurCode MLP does NOT support the Multi-Channel WAV file format (Multiple mono WAV files must be used as input to the encoder). 1.10 MLP Logs A feature of SurCode MLP is the Encode Log (see Figure 1- 1) generated by each encode pass.<br><br> It gives you specific information about file types and rates, durations, and whether the encode passed (including verification). When ReBit (Bit-Depth Reduction) is active, the Log includes the Introduction to SurCode MLP 6 steps taken in ReBit (see the cReBit d chapter). The log can be saved for further reference.<br><br> Note: There is also a cVerifier Log d generated by SurCode MLP after the cVerify d stage (see the cVerification/ Playback Section d chapter). Figure 1-1 Getting Started 7 2 Getting Started 2.1 Open SurCode MLP When you open SurCode MLP, you will see the Main Screen, shown in Figure 2-1 (see the cMain Screen d chapter also). .<br><br> Figure 2-1 2.2 Name The File You will need to name the MLP file (the file that will contain the encoded MLP stream). The MLP file type is not selectable and has a default in the cSave as type: d window Getting Started 8 to cMLP Encoded Files (*.mlp) d. This is because SurCode MLP only produces MLP files.<br><br> SurCode MLP will automatically add an c.mlp d extension to your file name, unless it sees a previously entered c.___ d-type extension. Do not to change this unless you specifically want a different extension. Many programs (including Minnetonka Audio 9s discWelder CHROME) look for the .mlp extension when authoring.<br><br> 2.3 Select A Destination Click on the cDestination d button (under the SurCode Logo) to bring up the screen for cSelect A File d (see Figure 2-2) . Figure 2-2 Browse in the cSave in d window (by clicking on the arrow button on the right side of the window) to select the folder or volume where you wish to save your MLP file, then enter Getting Started 9 a suitable name for the MLP file in the cFile name d window. After clicking the cSave d button, the MLP file name will be displayed in the SurCode MLP title bar.<br><br> 2.4 Select Files To Encode Clicking on the Channel Name button ( cLeft Front, Right Front d, etc.) opens up a browse window, where you can select the soundfiles you wish to enter in the selected channel (enter the cleft d soundfile into the cLeft Front d channel, etc.) Continue entering soundfiles into each channel that you have source soundfiles for. 2.5 Press The Encode Button When all the source soundfiles have been entered into their respective channels, press the cEncode d button in the lower- left side of the screen. SurCode MLP will encode your files, automatically setting the options and assignments to the default values based on industry standards.<br><br> 2.6 Listen To The MLP File You can check the encoded file by playing it back using the transport controls in the lower right side of the screen (it will have been verified for accuracy automatically). Select the cDestination d button under the controls, and press cPlay d. 2.7 Other Options The remainder of this manual describes encoding options and features that may be used to customize the encoding process.<br><br> Main Screen 10 3 Main Screen The SurCode MLP Main Screen is where your soundfiles will go to be encoded or played (see Figure 3-1). Figure 3-1 3.1 Menu Bar This is a standard Windows menu bar, with the following components: Main Screen 11 3.1.1 Setup This is where you will find the standard Windows commands cNew d, cOpen d, cSave d, cSave As d, cRecent Projects d, and cExit d. If these commands are unfamiliar to you, please consult the Windows Manual.<br><br> 3.1.2 Options The cEncoder Options d section found in this menu bar selection is important in setting the parameters for encoding if you decide to use settings other than the default parameters (see Figure 3-2). Figure 3-2 Also found in this menu bar selection is cDevice Options d, which allows you to set the devices for audio playback (see the cOptions d chapter). 3.1.3 Help This is a standard Windows menu bar selection, with cSur- Code Help d (which opens the SurCode MLP Manual in HTML form), cSurCode Web Site d (which connects you to the Minnetonka Audio Software web site via your Internet connection), and cAbout SurCode MLP d (which has ver- sion, copyright, and contact information).<br><br> Main Screen 12 3.2 Destination The cDestination d button allows you to set up the encoded file destination folder or volume, as well as the encoded file name (see the cGetting Started d chapter). 3.3 Channel Assignment This drop-down list allows you to specify the type of chan- nel assignment combination that fits your project. There are 21 combinations of channel assignment/group assignment, and you may choose the type of combination for your project (see Table 3 in cAppendix A - Tables d and the cChannel Assignment d chapter), or SurCode MLP will chose the combination automatically based on the channel soundfiles you enter (automatic combination selection is the default setting).<br><br> 3.4 Soundfiles This set of browsable windows with corresponding channel buttons (Lf, Rf, Center, LFE, Ls, and Rs) is where you will enter the soundfiles for these channels (see the cSoundfiles d chapter). Main Screen 13 3.5 Clear These six buttons clear the soundfile from each of the six channel windows (see Figure 3-3). Figure 3-3 3.6 Group Number This small window (see Figure 3-3) next to each of the six channels indicates whether the channel is assigned to Group 1 or Group 2 (see Table 3 in cAppendix A -Tables d and the cChannel Assignment d chapter) 3.7 Channel Bit/Sample Rate These windows (see Figure 3-3) indicate the bit- and sam- ple-rates of each of the channel soundfiles.<br><br> Main Screen 14 3.8 Mute/Solo These buttons Mute (M) or Solo (S) the six individual chan- nels during playback (see Figure 3-4). Figure 3-4 3.9 Channel Name These are the cBrowse for Soundfile d buttons for the 5.1 Surround channels (see Figure 3-4). When a Channel Name button is pressed, a browse window opens for soundfile selection for that channel.<br><br> Double-clicking on a soundfile places it into that channel 9s window (see the cSoundfiles d chapter) 3.10 Encode Section This section contains the cEncode d button, the cStop d but- ton , the Start Time Set and Reset buttons, and the End Time Main Screen 15 Set and Reset buttons, as well as the Start and End Time windows (see the cEncode Section d chapter). 3.11 Playback Section This section contains the playback transport controls, the playback location (time) indicator, and the Source/Destina- tion selector (see the cPlayback Section d chapter). Options 16 4 Options The cOptions d selection on the Menu Bar is where the set- tings for playback ( cDevice Options d) and encoding ( cEncoder Options d) are found.<br><br> 4.1 Device Options This option sets the computer 9s sound playback devices (internal sound cards or USB/Firewire interfaces) for the four pairs of audio outputs (see Figure 4-1): Lf/Rf; Ls/Rs; Center/LFE; and Downmix (see the cDownmix d chapter). Note: If the Playback Device(s) are newly-installed, make sure you have their current software drivers installed before selecting them. Figure 4-1 4.2 Encoder Options This section is central to setting up the SurCode MLP encoder parameters if you do not wish to encode using the default settings (see Figure 4-2).<br><br> Options 17 Figure 4-2 Options 18 4.2.1 Downmix Section This section allows you to choose whether or not to include a downmix in your MLP encoded stream, and what the individual channel parameters within that downmix will be (see the cDownmix d chapter). 4.2.2 ReBit Section This section allows you to reduce the bit-depth of any or all of the six channels in extreme cases where encoding within the 9.6Mbps DVD data rate is not possible without altering the source files (see the cReBit d chapter). 4.2.3 Verification/Playback Section This section allows you to specify two optional modes of verification (standard verification is automatic and non- defeatable), and one of two types of soundfile playback (see the cVerification/Playback d chapter).<br><br> Soundfiles 19 5 Soundfiles 5.1 Entering Soundfiles Entering your source soundfiles in SurCode MLP is very easy - simply click on the cChannel Name d button (see Fig- ure 5-1) and browse through the available volumes and files for the soundfiles you wish to encode (see Figure 5-2). Then double-click on the soundfile that corresponds to the selected channel name (i.e., select the Left soundfile for the Left Front window). Figure 5-1 Enter all the rest of your soundfiles in their respective win- dows by clicking on the Channel Name buttons and double- clicking the appropriate soundfiles to enter them.<br><br> If you do not have soundfiles for some of the channels, simply leave them blank. SurCode MLP will automatically re-configure the Channel Assignment choices to reflect the channels that are filled (see Section 5.4, cChannel Assignment d). Soundfiles 20 Figure 5-2 5.2 Clearing Soundfiles To clear a soundfile from a channel window, click on the cChannel Clear d button next to that channel 9s window.<br><br> 5.3 Channel Bit- and Sample-Rates SurCode MLP shows you the bit-rate and sample-rate infor- mation for the channel soundfiles in the cBit/Sample Rates d windows. Standard digital audio numbering is followed here, which is cbit-rate number/sample-rate number d. 5.4 Channel Assignment There are 21 combinations of Channel Assignments (see Table 3 in cAppendix A -Tables d and the cChannel Assign- ment d chapter) involving two different Groups.<br><br> This allows you to place channels with lower sample-rates in a second group, as long as two conditions are met: Soundfiles 21 1. The Group 2 channels must all have the same sample- rate (this is also true for Group 1). 2.<br><br> The second group 9s sample-rate must be an even mul- tiple of the first group 9s; for example, if Group 1 is at 96kHz and Group 2 is at 48kHz, then Group 2 is exactly one-half the sample-rate (Fs) of Group 1. SurCode MLP automatically limits the choices of channel- to-group assignments to those applicable to the number and designation of channels that have soundfiles entered. For example, if you place a soundfile in the cCenter d channel, SurCode MLP will allow only the Group combinations with a cC d in the available choices.<br><br> Note: If you are using ONLY a Mono soundfile in a track, it MUST be assigned to the Center ( cC d) channel (Channel Assignment Option 0), not to the Left (or Right) channel. 5.5 Channel Group Indicator These windows indicate which Group (1 or 2) the associ- ated channel is assigned to (see the cChannel Assignment d chapter). Downmix 22 6 Downmix 6.1 What Is Downmixing?<br><br> The purpose of downmixing is to allow consumers without surround audio capabilities to listen to your surround mixes in stereo. Some producers include a separately-mixed stereo track in the DVD-Audio disc, but this uses more disc space. SurCode MLP provides a simple solution to providing a 2- channel (stereo) downmix.<br><br> The encoder includes lossless matrixing, which can encode a two-channel downmix as a linear combination of the multi-channel mix and encode this alongside the multi-channel version on the DVD-A disc. The advantage of this approach is that the producer can listen to the downmix at the encoding stage with the knowledge that it will be delivered bit-for-bit to the end user at the decoding stage. Another advantage is that a two- channel-only playback device does not need to decode the multi-channel stream, and needs only to decode the stereo.<br><br> 6.2 Downmix Defaults SurCode MLP defaults to cDownmix On d, and to industry standard downmix coefficient values (shown in Figure 6-1). To encode a downmix using values other than the defaults, you will need to specify the new values for each channel in the Channel Coefficient windows. These coefficient values can be determined by listening to the surround mix on an audio workstation or mixing console, blending the tracks into a stereo mix, and determining the desired balance.<br><br> The resulting channel fader settings can then be applied to the respective Downmix Channel Coefficient settings directly Downmix 23 in dB, as well as any phase (polarity) reversal and/or channel mutes. Figure 6-1 6.3 Left/Right Downmix This section, which is found in the cOptions d Menu Bar section (Menu Bar > Options > Encoder Options) allows you to choose whether or not to include a downmix in your MLP encoded stream by clicking on the Downmix Select checkbox (the default is cOn d), and what the individual channel coefficients within that downmix will be (see Figure 6-1). Downmix 24 6.3.1 Channel Coefficients The Channel Coefficient values are entered into the Left Mix and Right Mix Channel Coefficient windows via the up/down buttons (located next to the coefficient windows).<br><br> These coefficients determine the relative levels (that is, relative to the levels of the source surround mix soundfiles) of each channel in the encoded stereo downmix, and they are entered in decibels (dB). Decibel entry is effective because the relative dB levels can be read directly from audio console or workstation channel fader settings. The range of values for the channel coefficients is from (plus) 6dB (increased gain) to minus infinity (maximum attenuation, or cOff d).<br><br> 6.3.2 Channel Phase The Channel Phase In/Out button (actually a 0/180 degree polarity button) reverses the phase (polarity) of the individual channels. The c0 d setting is in-phase, the cout d setting is 180 degrees out-of-phase (reverse polarity). 6.3.3 Channel Mute The Channel Mute (On/Off) button is useful for quickly reducing the coefficient values down to minus infinity ( cOff d) for alternate channel stereo assignment (i.e., Lf to Left Mix, Rf to Right Mix; and Ls to Left Mix, Rs to Right Mix).<br><br> The On/Off button toggles between cOff d and the cOn d value shown in the Channel Coefficient window. Downmix 25 6.3.4 Restore Defaults This button resets all downmix parameters to their default values (which are shown in Figure 6-1). 6.3.5 Downmix Clipping The downmix process does NOT prevent clipping of the stereo output signal, because that may be the artistic intent.<br><br> When clipping occurs, the signal is not clipped in the full multi-channel MLP stream (which continues to operate losslessly) but in the DVD-Audio player. If the downmix is clipped, the file WILL encode (i.e. it is not a Fatal Error, which would cause the encoder to stop), but the Encode Log will indicate that the downmix has cSaturated d.<br><br> Verification of this encoded file will fail, with the reason for failure in the Verifier Log noted as cDownmix Clipped d. cSaturation d and cClipped d are terms for the same thing. Note: Gain added in any of the Channel Coefficient windows (the available gain is 6dB) will possibly clip the downmix playback - use caution when adding gain, and always listen to the encoded MLP file to make sure the downmix is free of clipping (unless that is the artistic intention).<br><br> 6.4 Downmix Parameters The Downmix output is at the same sample-rate as the soundfiles in Group 1, and it is always at 24-bit, even if the source soundfiles are at 16- or 20-bit. The Downmix output is also always in the stereo format. Channel Assignment 26 7 Channel Assignment SurCode MLP allows the user the choice of defining the channel-to-group assignments, or letting the program auto- matically determine the assignments based on the user 9s soundfile entries (see the cSoundfiles d chapter).<br><br> The avail- able combinations are shown in the Channel Assignment Options Window, which is activated by the Channel Assign- ment Options Pull-down button on the right side of the Channel Assignment Window (see Figure 7-1). Figure 7-1 7.1 Why Use Groups? The reason for the assignment of channels into two groups is the capability of mixing two different sample-rates (the sample-rate within a group must be the same for all chan- nels).<br><br> For example, the Lf, Rf, and C channels may be at 96kHz resolution in Group 1, and the Ls, Rs, and LFE chan- nels may be at 48kHz resolution in Group 2. However, the Channel Assignment 27 sample-rate for Group 2 must be an even multiple of the sample-rate in Group 1, for example Group 1=96kHz, Group 2=48kHz.. Note: Channel bit-depths may be different within the two groups.<br><br> 7.2 Channel Assignments DVD-Audio supports up to six channels in one of 21 combinations of channel assignments based on number and types of channels (see Table 3 in cAppendix A - Tables d) 7.3 Assignment Options SurCode MLP will make the assignment option choice for you based on the channels you enter into the channel win- dows. This process is automatic. However, you may wish to set up the Group assignment combination manually, either after entering a single sound- file, or after all your soundfile entries are done.<br><br> If you wait until all your soundfiles are entered, the combination options will be limited to the applicable choices. For exam- ple, if the source soundfiles you have entered include Lf, Rf, LFE, and C channels, SurCode MLP will limit the option choices to those with these channels in the combination. This lessens the possibility of error in channel assignment group combinations.<br><br> Note that there will be channels in the available combination options that you have NOT entered (to allow further channel entry, if desired), but each of the options will include those channels already entered. Note: If you are using ONLY a Mono soundfile in a track, it MUST be assigned to the Center ( cC d) channel (Channel Assignment Option 0), not to the Left (or Right) channel. Encode Section 28 8 Encode Section Once the soundfiles have been placed (see the cSoundfile chapter) and the encoded file destination specified (see the cGetting Started d chapter), you are ready to encode the soundfiles into an MLP stream, unless you wish to change the defaults of any of the following: 1.<br><br> The encoder options (see the cOptions d chapter) 2. The downmix coefficients (see the cDownmix d chap- ter) 3. The channel assignment (see the cChannel Assign- ment d chapter) If your project is ready to encode, simply press the cStart Encode d button (see Figure 8-1) and wait for the finished MLP file.<br><br> Figure 8-1 Encode Section 29 8.1 Encode Start and End Time You can encode the entire soundfile length (which is the default setting), or encode only a section of the soundfile by specifying a different start and end time in the cStart Time d and cEnd Time d windows (see the cCaution d below). Additionally, you can set the encode Start or End times at the current playback location with the Set buttons (see Fig- ure 8-1). If you want the soundfile encode to start at 00:00:00.00 (which is the Start Time default), you can press the Reset button next to the Start Time window.<br><br> Pressing the Reset button next to the End Time window resets the End Time to the length of the soundfiles (which is the End Time default). Note: Time location in SurCode MLP is numbered as Hours: Minutes: Seconds. Fractions (hundredths).<br><br> CAUTION: You can directly enter the time values in the Start Time and End Time windows; however, if you do so you will lose the cthousandths d decimal place, which is not shown in the windows (00:00:00.000 ). This may result in a loss of sam- ples at the start or end of the current project, depending on which window you edited by direct entry. 8.2 The Encode Button This button starts the encoding process.<br><br> 8.3 The Stop Button This button stops the encoding process. This will result in an incomplete but playable MLP stream, which can be used to test various settings in the encoder before committing to an entire soundfile encode (which can take a long time with large soundfiles and slow computers). Playback Section 30 9 Playback Section SurCode MLP allows you to playback your soundfiles from the Source, or from the Destination.<br><br> (see Figure 9-1). Figure 9-1 9.1 Source/Destination Select If you chose cSource d with the cSource/Destination Select d radio-style buttons, the imported source files are played through the output devices selected in the cDevice Options d window (Menu Bar > Options > Device Options), following the selected track assignments (see the cOptions d chapter). If you chose the cDestination d button, the cDestination d MLP file is decoded and played.<br><br> In addition, while playing back the decoded MLP file, SurCode MLP will verify that the file is a valid MLP stream (see the cVerification/Play- back Section d chapter). If the MLP stream is not valid, an error message is generated. Playback Section 31 9.2 Transport Controls The control buttons, from left to right, are: Return To Zero (the start of the soundfile), Rewind, Fast Forward, Stop, and Play.<br><br> These function similarly to tape transport controls. 9.3 Current Location This windows indicates the current location of the playback in Hours: Minutes: Seconds only (no fractions of a second). Note: Time location in SurCode MLP is numbered as Hours: Minutes: Seconds.<br><br> Fractions (hundredths). Verification/Playback Section 32 10 Verification/Playback Section 10.1 Verify Select SurCode MLP automatically verifies that the encoded MLP stream is identical to the source files. In the cEncoder Options d section (Menu Bar > Options > Encoder Options), there are two available modes in the cVerify Select d section, selectable with radio-style buttons (see Figure 10-1): 1.<br><br> cVerify after encoding is complete d (the default set- ting), or: 2. cVerify on-the-fly while encoding d. The cVerify-on-the-fly d mode is faster than the cVerify after encoding is complete d, but is useful mainly for test puposes, becuase it verifies only the data in the buffer, not the actual data written to disc.<br><br> The cVerify after encoding is complete d mode is the only true bit-for-bit verification between the source soundfiles and the disc data. Figure 10-1 Verification/Playback Section 33 10.2 Write Decoded File Select Additionally, SurCode MLP can write decoded Wave files while verifying, for later quality reference. These Wave files are written as decoded stereo pairs (Lf/Rf, C/LFE, Ls/Rs, Downmix) that are placed in the same folder as the MLP Destination (encoded) file.<br><br> These files can be imported by a sound editing program, including the program that created the source soundfiles, for waveform quality comparison if desired. Simply check the cWrite decoded wave files while verifying d box (see Figure 10-1). 10.3 Playback Select Surcode MLP allows you to playback the decoded MLP stream in one of two modes (also see the cPlayback Sec- tion d chapter): 1.<br><br> cPlayback individual channels d mode (the multi- channel MLP stream), or: 2. cPlayback downmix d mode (the stereo MLP down- mix). The selection is made by pressing the appropriate radio- style button (shown in figure 10-1).<br><br> Note: Playback of the MLP stream is possible only after encoding. 10.4 MLP Logs A feature of SurCode MLP is the Encode Log (see the cIntroduction d chapter) generated by each encode pass. There is also a cVerifier Log d generated by SurCode MLP after the cVerify d stage.<br><br> ReBit (Bit-Depth Reduction) 34 11 ReBit (Bit-Depth Reduction) 11.1 Why Is Bit-Depth Reduction Needed? In certain rare cases of highly un-correlated source material or very large source bandwidth requirements (for example, six channels of completely random noise at 24-bit, 96kHz resolution), the MLP process cannot pack the original chan- nel data into the 9.6Mbps maximum data bandwidth of the DVD-Audio specification and remain Lossless. In these cases, SurCode MLP will generate an error message and abort the encoding process at the point the material failed to encode properly (see Figure 11-1).<br><br> Figure 11-1 The result is that the original tracks themselves must be altered BEFORE encoding, to enable SurCode MLP to pack the material into an MLP stream. There are several ways to do this alteration: 1. Reduce the bit-depth of one or more channels, such as from 24-bit to 22-bit.<br><br> 2. Filter one channel 9s frequency bandwidth to low-pass only, or LFE (Low Frequency Effect, commonly called the Subwoofer channel) if this has not been done already as part of the 5.1 Surround mix process. Note ReBit (Bit-Depth Reduction) 35 that the LFE channel is frequency-limited in the 5.1 Sur- round mode, but other surround modes may use six full- range channels (6.0 Surround, for example).<br><br> 3. Reduce the audio bandwidth; for example, by filter- ing information above an arbitrary frequency, such as 40kHz when sampling at 96kHz. 4.<br><br> Use Minnetonka Audio 9s exclusive feature called cReBit d, which can manually or automatically reduce the effective bit-depth on a per-channel basis from within SurCode MLP. All of these options will increase the amount of compres- sion that SurCode MLP can achieve, thus increasing the playing time or reducing the peak data rate. However, only one of these choices works within SurCode MLP: ReBit.<br><br> 11.2 Why Is SurCode MLP 9s ReBit Needed? Minnetonka Audio designed SurCode MLP to be able to process all cases of highly un-correlated source material or very large source bandwidth requirements from within the program using ReBit, for three reasons: 1. To save the time necessary to export the original material to a separate program for reducing bit-depth, frequency bandwith, or both.<br><br> 2. To eliminate the need for requesting the source mate- rial 9s producers to do this process themselves. 3.<br><br> To ensure that the changes made to the material are sufficient to allow the MLP process to encode the mate- rial without further changes (and to ensure that excess bit-depth reduction is not done). ReBit (Bit-Depth Reduction) 36 11.3 How ReBit Works The ReBit process reduces the effective bit-depth of selected channels by resetting the Least-Significant Bits (or LSB - the 24th bit, the 23rd bit, etc.) to zero, either manu- ally or automatically, in 2-bit steps. This reduces the amount of data compression necessary to encode the channel, with- out reducing the actual number of bits (which is why this is known as ceffective bit-depth reduction d).<br><br> The significance of this is that if the DVD-A player has a c24-bit d indicator, it will stay lit. ReBit is a clossy d process, because it reduces the actual bit- depth resolution of one or more channels before encoding (the MLP encoding process itself remains Lossless). If the source material has failed the encode process, the original soundfiles must be altered or be removed from the DVD- Audio disc.<br><br> In these rare cases, ReBit is an effective solu- tion. Another consideration is that the ReBit process (in the cAutomatic d mode) affects the LFE channel first, and then occasionally the Ls/Rs channels, and typically by only 2-4 bits. The LFE channel is often frequency-limited in 5.1 sur- round playback to 120Hz low-pass, and the reduction of effective bit-depth on this channel is very nearly inaudible.<br><br> Likewise, the Surround channels are frequently reproduced by less-than-full-bandwidth speakers, and the reduction of effective bit-depth on these channels is also nearly inaudi- ble. ReBit (Bit-Depth Reduction) 37 11.4 The ReBit Section To enable the ReBit process, check the cReduction Select d box. (see Figure 11-1).<br><br> Figure 11-2 11.4.1 Manual Bit-Depth Reduction There are two modes in ReBit - manual and automatic. If you wish to set the bit-depth reduction parameters manually, use the up-down buttons next to each of the individual chan- nel cReduction Amount d windows. The most common prac- tice is to start by reducing the LFE channel bit-depth first, because this change is usually inaudible due to LFE/Sub- woofer channel low-pass filtering during playback (in 5.1 surround mode).<br><br> Subsequent bit-depth reduction can be per- formed on the Surround channels (Ls and Rs) in 2-bit incre- ments if the LFE channel bit-depth reduction fails to reduce the data sufficiently to pass the encode process. Usually the Lf, Rf, and C channels will not need to be reduced. ReBit (Bit-Depth Reduction) 38 11.4.2 Automatic Bit-Depth Reduction The strength of ReBit is the ability to automatically reduce the data rate by following pre-set steps calculated to cause the least audible change to the source material.<br><br> ReBit cycles through effective bit-depth reduction and re-encoding steps until the material passes the encode process - the bit-depth reduction goes no farther than necessary. To apply Automatic Bit-Depth Reduction, check the cAuto- matic Reduction Select d box, then click on the cOK d but- ton, and then the cEncode d button on the Main Screen. Note: The cAutomatic d setting also re-configures each cChannel Reduction d window to the cMaximum d bit-depth reduction amount.<br><br> You must enter the maximum number of bits you want to reduce each channel by - ReBit will do nothing if the cReduction Amount Maximum d windows all remain at zero. ReBit will fully reduce each channel by the entered amount only if it is necessary. This allows you to set the effective bit-depth from a 24-bit file down to a minimum of 16-bits (a setting of c8 d in the cReduction Amount Maxi- mum d window, i.e.<br><br> 24 minus 8=16). For example, you could set the Lf/Rf and Center Maximums to 2 and the rest of the channel Maximums to 8, ensuring that the front and center channels would not be reduced below a minimum of 22-bit effective resolution, while the other 3 channels could be reduced down to a minimum of 16-bit effective resolution. 11.4.2.1 ReBit Step Order In ReBit, there are 16 successive steps in the automatic pro- cess of resetting the Least Significant Bits (LSB) to zero, two bits at a time, with an encoding restart between each step - a failed encode will initiate the next LSB reset step.<br><br> ReBit (Bit-Depth Reduction) 39 The sequence of automatic bit-depth reduction steps is shown in the table below. Note: Effective bit-depth minimum is16-bits. 11.4.2.2 Time Savings The time saved by the ReBit cAutomatic d mode can be sig- nificant, because the MLP encode process itself can be Table 1: Effective Bit-Depth per Channel (24-bit files) Step LfRfCLsRsLFE 12424242424 22 22424242424 20 3242424 22 22 20 4242424 20 20 20 5242424 20 20 18 6242424 18 18 18 7242424 18 18 16 8242424 16 16 16 92424 22 16 16 16 102424 20 16 16 16 112424 18 16 16 16 122424 16 16 16 16 13 22 22 16 16 16 16 14 20 20 16 16 16 16 15 18 18 16 16 16 16 16 16 16 16 16 16 16 Step Shaded Channels have been Bit-Depth Reduced ReBit (Bit-Depth Reduction) 40 time-consuming, depending on the amount of material and the host computer 9s speed and RAM.<br><br> The ability to solve the problem of highly un-correlated source material or very large source bandwidth requirements from within SurCode MLP can save time as well as extended re-mastering. ReBit cAutomatic d eliminates the need to manually alter the mate- rial, re-encode, then repeat the alter/re-encode cycle until it passes the encode process. 11.4.3 Worst-Case Scenario In encode-failure testing, worst-case scenario source sound- files - consisting of six channels of 24-bit, 96kHz random noise, which failed the encode process immediately - were used to step ReBit cAutomatic d through the LSB reset sequence.With the cMaximum d channel coefficients set to c8 d (or a possible effective minimum of 16-bits), ReBit went through eight steps before the random noise files passed the encode process.<br><br> That translates into no change on the Lf, Rf, and C channels, and an effective bit-depth in the LFE and Ls/Rs channels of 16-bits. The front and center channels retained their 24/96 resolution and the surround and LFE channels were reduced to 16/96 resolution - not ideal, but the random noise passed. An cEncode Log d is generated by Surcode MLP during encoding, and the successive ReBit steps are shown in this log.<br><br> The log can be saved for further reference (see the cIntroduction d chapter). Most cases of music, or speech with music, will not proceed through any significant number of processing steps, if any. SurCode MLP 9s ReBit is really a worst-case scenario prob- lem-solver - you very likely will never need it, but it 9s good to have if you do.<br><br> Appendix A - Tables 41 12 Appendix A - Tables 12.1 Compression Amounts The following table gives examples of the typical compres- sion that can be realized with different audio bit- and sam- ple-rates, 12.2 DVD-A Playing Time The following table gives examples of playing times on DVD-Audio that can be obtained with different channel, bit-rate, and sample-rate combinations. Bit-RateSample-RateMinimumTypical 16-bit48kHz0%50% 20-bit96kHz40%55% 24-bit96kHz38%52% 24-bit192kHz43%50% Table 1: Compression Amounts Number Of Channels Bit-RateSample-RatePlay Time 6 Chan.24-bit96kHz86 min. 5.1 Chan.24-bit96kHz100 min.<br><br> 2 Chan.24-bit192kHz2 hours 2 Chan.24-bit96kHz4 hours 2 Chan.16-bit44.1kHz12 hours 1 Chan.16-bit44.1kHz25 hours Table 2: DVD-A Playing Time Appendix A - Tables 42 12.3 Channel Assignments DVD-Audio supports up to six channels in one of 21 combi- nations of channel assignments. #Ch. 0Ch.<br><br> 1Ch. 2Ch. 3Ch.<br><br> 4Ch. 5 # In Grp. 1 # In Grp.<br><br> 2 0 M10 1 L R20 2 Lf Rf S21 3 Lf Rf Ls Rs22 4 Lf Rf LFE21 5 Lf Rf LFE S22 6 Lf Rf LFE Ls Rs23 7 Lf Rf C21 8 Lf Rf C S22 9 Lf Rf C Ls Rs23 10 Lf Rf C LFE22 11 Lf Rf C LFE S23 12 Lf Rf C LFE Ls Rs24 13 Lf Rf C S31 14 Lf Rf C Ls Rs32 15 Lf Rf C LFE31 16 Lf Rf C LFE S32 17 Lf Rf C LFE Ls Rs33 18 Lf Rf Ls Rs LFE41 19 Lf Rf Ls Rs C41 20 Lf Rf Ls Rs C LFE42 Group 1 Group 2 Table 3: Channel Assignments Appendix A - Tables 43 Table 3 Key: M= Mono L= stereo Left R= stereo Right Lf= Left front Rf= Right front C= Center LFE= Low Frequency Effect (commonly called Subwoofer) S= Surround Ls= Left surround Rs= Right surround The DVD-Audio channel assignments 8 to 12 are almost identical to the channel assignments 13 to 17, and differ only in the way the channels are grouped (the Center Chan- nel changes Groups). Note: The sample-rates of the channels in the two Groups must be even multiples, i.e. 96kHz/48kHz, or 88.2kHz/44.1kHz.<br><br>

less