- From: Robin Raymond <robin@hookflash.com>
- Date: Thu, 08 May 2014 00:13:06 -0400
- To: "public-ortc@w3.org" <public-ortc@w3.org>
- Message-ID: <536B0452.9080505@hookflash.com>
I am contributing a proposal on how to resolve an issue discovered in
the usage of "parameters". While the details can always be tweaked, I
think it successfully resolves much of the concern around the level and
knowledge required to configure a "parameters" object for anything other
than the basic use cases.
In response to this posting:
http://lists.w3.org/Archives/Public/public-ortc/2014May/0007.html
This also addresses the issue of exchanging detailed parameters over the
wire and instead base parameters based on capabilities.
I am going to copy the entire proposal below to official contribute the
proposal but for the sake of readability I am also including a link to
the google doc(s).
Proposal-ORTC Sender / Receiver Capabilities Based Model
https://docs.google.com/document/d/1htyRaNjXTE_O1GhD8TcLCNXFvVsgszpE8Lqgp3OCHlU/edit?usp=sharing
Proposal-ORTC Sender / Receiver Use Case [Usage Comparison Analysis]
https://docs.google.com/document/d/1hdhCHj-gpwv06vIbAftxMG3oZtz7A-nuYsuwQEkTat4/edit?usp=sharing
Proposal-ORTC Sender / Receiver Capabilities Based Model
*
Introduction
After attempting to write out some use cases using the existing
RTCRtpSender and RTCRtpReciever objects and parameters for ORTC, some
issues were discovered. Specifically, the application developer would
need to have a fair amount of knowledge on exactly how to tweak low
level parameters for anything beyond very simple use cases. For example,
setting up an SVC (Scalable Video Codec) would have required knowing
about what codecs support SVC, how the layering is setup for particular
codecs, and finally setting up specific geometric (or temporal)
attributes and layering relationship details by an application developer.
As a result of the lack of easily configuration of RTP features, the
idea came out to give the application developer "preferences" where the
developer could choose what they want desire with high level knobs and
dials and let the engine (which has explicit knowledge of each codec)
configure the low level "parameters" details according to a developer's
wishes. The engine could then return the closest set of preferences that
could be achieved given the capabilities of the engine and the developer
can then choose to proceed or not setting up media flows using these
preferences and constructed parameters.
Another important discovery was made in the process of defining
"preferences". If two ORTC engines were given the same set of
preferences and the capabilities of both sender and receiver, each
engine could be made to construct "compatible" sender and receiver
"parameters" details without ever exchanging the parameter details over
the wire. This small realization about generating "parameters" from
capabilities for local consumption by an engine has a huge impact. This
generation removes the need for an engine to understand and filter
settings that it may not understand created by another engine of unknown
origin, which may use proprietary and/or custom settings. A simple
"ignore capabilities you don't understand" rule could replace complex
and cumbersome rules that would be otherwise required if "parameters"
were to be sent over the wire and later filtered using a set of
capabilities.
Parameters can be generated based on the union of sender and receiver
capabilities along with application developer preferences being used as
a guideline on how to create the parameters. The engine will do it's
best to fulfill the preferences and it will return the parameters that
are possible given the union of the capabilities.
Two different engines must be able to compute compatible parameters
given all the same preferences and capabilities. Fortunately, any two
engines that understand the same capabilities can easily follow the same
rules to generate compatible parameters. While the parameters created on
the sender and receiver are required to be "compatible", they need not
be identical. The application developer should call
"createParameters(...)" on sender to create parameters suitable for the
sender. The application developer should call "createParameters(...)" on
the receiver to create params suitable for a receiver. The calculated
"parameters" for both sender and receiver have to be compatible only to
the extent that whatever a sender produces a receiver must be capable of
decoding.
The application developer has the option to tweak the detailed
parameters output by "createParameters(...)" but should only do so with
extreme caution. The resultant parameters output by
"createParameters(...)" are only meant for local consumption by the
local sender / receiver "start" methods. Sending these created
parameters over the wire is discouraged because implementations may
produce objects which may not be entirely understable by the remote
party, even though the media sent on the wire will be compatible.
Differences from Current Sender / Receiver API
Both models and APIs are more similar than they are different. The
subtle differences make important behavioural usage implications.
Both models send and receive based upon "parameter" settings. The
difference is in how the "parameters" are generated. The new model
generates the "parameters" based on an exchange of capabilities and the
application developer is given convenient 'knobs' called "preferences"
to perform most common use cases. The "parameters" in the new model are
intended for local consumption only and the application developer is not
required (and actively discouraged) from marshalling these "parameters"
over the wire. The new model proposes marshaling and exchanging
"capabilities" and optionally "preferences" and then generating
compatible "parameters" based on those exchanges.
In both models, the application developer may choose to tweak low level
parameters should specific compatibilities be required. But the
"preferences" model allows most application developers to completely
ignore the low level parameters.
Advantages of the New Capabilities Model
Overall the proposed capabilities based API has strong advantages. Main
advantages are:
1.
Simplicity in setup based on "preferences" for the application developer
2.
Less brittle designs/implementations since low level parameters are
not exchanged, filtered, and interpreted by different browser engines
3.
Much less knowledge (and often no pre-knowledge) is required for the
application developer to take full advantage of a browser's capabilities
RTCRtpSender / RTCRtpReceiver
interface RTCRtpSender{
// ...
static RTCRtpParameters createParameters(
MediaStreamTrack track,
CapabilitiesreceiverCaps,
optional (RTCRtpAudioPreferences or
RTCRtpVideoPreferencesor
RTCRtpSimulcastPreferences) prefs,
optional CapabilitiessenderCaps // optional as system can obtain
this information
);
void start(RTCRtpParametersparams);
// ...
);
interface RTCRtpReceiver{
// ...
static RTCRtpParameters createParameters(
DOMString kind,
CapabilitiessenderCaps,
optional (RTCRtpAudioPreferences or
RTCRtpVideoPreferencesor
RTCRtpSimulcastPreferences) prefs,
optional CapabilitiesreceiverCaps // optional as system can obtain
this information
);
void start(RTCRtpParametersparams);
//...
);
RTCRtpMediaPreferences
// This is the base dictionary used for both audio and video preferences
and represents
// the set of common preferences that are available for both media types.
dictionary RTCRtpMediaPreferences{
// If not specified, system will choose value. If specified, this
receiverIdwill
// be applied to primary SSRC "as is". If more than one SSRC is
needed to encode
// the stream (e.g. FEC, RTX, MST, simulcast), where the meaning of
the RTP packet
// with that alternative SSRC cannot be determined by the media flow
itself, the
// alternative SSRCs will construct a receiverIdvalue based upon
this receiverId
// value.
DOMString receiverId;
// This is the primary SSRC to use. Should alternative SSRCs be
required (e.g. FEC,
// RTX, MST, simulcast), all other SSRCs should be assigned
sequentially starting
// from the chosen SSRC value.
unsigned int ssrc;
// For a sender, force the chosen codec to be the codec within the
RTCRtpCapabilities
// with this name. If possible to choose this codec, the system will
confirm by
// choosing this codec in the result from "createParameters(...)".
// This value has no meaning for a receiver since a receiver must be
capable
// of receiving any of the compatible codecs within the union
RTCRtpCapabilities.
// A non specified value indicates the system will choose the
preferred sending
// codec.
DOMString codecName;
// This value indicates the relative importance of the media being
sent with a
// sender versus other media being sent. The logic is that all sent
media with
// the same priority will be treated as having an equal priority.
Those with
// a greater value will be given a greater priority and those with a
lower value
// will be given a lower priority. The value is relative meaning a
value of 2.0
// should be given roughly 2 times the priority vs a 1.0 value and a
value of 4.0
// should be given roughly 4 times the priority vs a 1.0 value.
double relativePriority = 1.0;
// This value indicates the maximum bit rate the media is allowed to
output as
// a combined whole (including all layers, FEC, RTX, etc). The system
will filter
// out codecs that are not capable of delivering below this bit rate
unless no
// codec is possible in which case the system will chose the minimal
codec bit rate
// possible and will override with a different maximum bit rate in
the result of
// "createParameters(...)".
double maxBitrate; // engine, keep under
this rate
// These values indicates the preferred treatment of FEC/RTX for the
RTP packets. For
// audio, some audio codecs have built in FEC/RTX mechanisms in which
case if the
// codec is capable, the codec should enable its FEC/RTX mode if
value is set to all
// for that codec rather than creating an additional RTP flow.
RTCRtpRecoveryOptionsfec = "none";
RTCRtpRecoveryOptionsrtx = "none";
};
enum RTCRtpRecoveryOptions{
"all", // apply to all layers
"base", // only apply for base (audio will treat "base" as
equivalent to "all")
"none" // do not apply to any layer
};
RTCRtpAudioPreferences
dictionary RTCRtpAudioPreferences : RTCRtpMediaPreferences{
// If not 0, tells the engine to pick and configure codecs that are
capable of
// the minimum of channels (if possible). If not possible, the
minimum number of
// channels will be returned in the result of "createParameters(...)".
unsigned int minChannels = 0;
// If not 0, tells the engine to pick a codec and configure codecs
which are
// capable of delivering the minimum Hz rate as indicated. If not
possible, the
// minimum Hz rate will be returned in the result of
"createParameters(...)"
unsigned int minHzRate = 0;
// The engine will choose and configure the codecs best able to
deliver the level
// of fidelity requested.
RTCRtpAudioFidelity fidelity = "speech";
};
enum RTCRtpAudioFidelity{
"speech", // speech only is expected so Hz range only need to
support the vocal range
"music", // music is expected, choose stereo compatible and minimal
32000 Hz
"movie" // music / sound effects expected, choose surround and
highest Hz available
};
RTCRtpVideoPreferences (and related)
dictionary RTCRtpVideoPreferences: RTCRtpMediaPreferences {
// minFrameRate, minScale, and minQualityeach indicate that the
engine must do
// it's best effort to keep the frame rate, scale or quality above a
certain minimal
// level. When using SVC, these values will hint at the requirements
typically needed
// for the base layer.
//
// minFrameRateis specified in frames per second.
double minFrameRate = 0; // please engine, keep equal or above
this rate
// minScaleis a relative value from 0.0 to 1.0 where 1.0 represents
full input stream
// width/height is requested and 0.0 represents no minimize size is
requested.
// The value of minScaleis multiplied by the source video window
width and height
// to calculate a minimal width and height that is relative to source
size.
double minScale = 0; // please engine, keep equal or above
this scale
// Alternatively, a specific fixed minimal width and height can be
requested.
double minWidth = 0; // please engine, keep above X pixels
wide
double minHeight = 0; // please engine, keep above Y pixels
high
// minQualityis a relative value from 0.0 to 1.0 where 1.0 means
maximum output
// quality is requested for a given codec and 0.0 allows any minimal
codec quality
// output is deemed acceptable.
double minQuality = 0; // please engine, keep equal or above
this quality
// The engine needs values to help decide what to sacrifice when
network conditions
// are not ideal. The frameRatePriority, scalePriority, and
qualityPriorityindicate
// the relative importance of each aspect of the video relative to
the other (or
// 0.0 which means the video aspect has no significance (with
exclusion to the minimum
// above). The values are relative to each other thus a value of 2.0
vs 1.0 has
// roughly 2 times the importance and a value of 4.0 vs 1.0 has
roughly 4 times the
// importance (relatively speaking).
double frameRatePriority = 1.0; // priority of frame rate
double scalePriority = 1.0; // priority of scale
double qualityPriority = 1.0; // priority of quality
// If a type of SVC layering is desired, the frameRateScalabilityOptions,
// scalingScalabilityOptions, and qualityScalabilityOptionsshould be
set to a
// non-null value for each SCV type desired. The details of the
// RTCRtpScalabilityOptionsdictionary will indicate the desired
details for
// each individual SVC type requested.
//
// Default of nullindicates no SVC of specific type is requested.
RTCRtpScalabilityOptions? frameRateScalabilityOptions = null;
RTCRtpScalabilityOptions? scalingScalabilityOptions = null;
RTCRtpScalabilityOptions? qualityScalabilityOptions = null;
};
dictionary RTCRtpScalabilityOptions{
// If the alternative value other than the default value of nullis
specified, this
// indicates to the engine the precise number of layers desired (if
possible for a
// given codec to deliver these layers). If null, the engine is free
to choose
// the default layering statically or dynamically dependent upon the
codec
// capabilities.
unsigned int? layers = null;
};
RTCRtpSimulcastPreferences
dictionary RTCRtpSimulcastPreferences{
// This value indicates the maximum bit rate all media is allowed to
output as
// a combined for all simulcast streams.
double? maxBitrate = null; // engine, keep
under this rate
sequence<RTCRtpVideoPreferences> simulcastStreams;
};
RTCRtpParameters
// Typically this object is constructed by the RTCRtpSenderfor local
consumption by
// the RTCRtpSenderand by the RTCRtpReceiverfor local consumption by a
RTCRtpReceiver.
// This is a "shotgun" object, meaning the developer is given the power
of a "shotgun"
// pointed at their feet and they can mess with this object at their own
peril should
// they need to modify it for unusual compatibility reasons. Normal use
cases should not
// require modifying the values within this structure and marshalling
this structure for
// remote consumption by another browser engine is highly discouraged.
dictionary RTCRtpParameters{
// When returned as a result, the system will express the actual
chosen preferences
// possible to best fulfill the preferences given the capabilities.
In other words,
// the developer can't always get what they want; but if they try
sometimes, they will
// get what they need.
(RTCRtpAudioPreferences or
RTCRtpVideoPreferencesor
RTCRtpSimulcastPreferences) preferences;
// the capabilities of both sender and receiver [value "as is" when
passed
// "createParameters(...)]"
RTCRtcCapabilitiessenderCapabilities;
RTCRtcCapabilitiesreceiverCapabilities;
// This value contains all the particularly low level details of how
the engine
// will encode the media on the wire.
(RTCRtpParameterAudioDetailsor
RTCRtpParameterVideoDetailsor
RTCRtpParameterSimulcastDetails) details;
// The chosen RTP features based upon the union of the capabilities.
SettingsrtpFeatures;
// The chosen RTP extensions and configurations based upon the union of
// the capabilities.
sequence<RTCRtpHeaderExtensionParameters>? headerExtensions = null;
};
RTCRtpParameterDetails
// This is the base dictionary of common parameters needed for both
audio and video media
// types. Audio and video will each have their own set of specific
parameters depending
// upon the media type.
dictionary RTCRtpParameterDetails {
DOMString receiverId = ""; // use this receiver ID for RTP
stream ("" = N/A)
unsigned int ssrc = null; // using this SSRC for RTP stream
DOMString fecReceiverId = ""; // use this receiver ID for FEC
RTP ("" = N/A)
unsigned int? fecSsrc = null; // using this SSRC for FEC (null
= N/A)
Settings fec; // modes of operation related to FEC
DOMString rtxReceiverId = ""; // use this receiver ID for RTX
RTP ("" = N/A)
unsigned int? rtxSsrc = null; // using this SSRC for FEC (null
= N/A)
Settings rtx; // modes of operation related to RTX
// nullfor a sender. For a receiver, this must contain the source SSRC to
// use for RTCP Receiver Reports (RRs).
unsigned int? rtcpSsrc = null;
// If true, the engine will mux RTCP with RTP on the same
RTCIceTransport. If false,
// the engine will send RTCP reports on the associated RTCP
RTCIceTransportcomponent.
boolean rtcpMux = true;
};
RTCRtpParameterAudioDetails (and related)
dictionary RTCRtpParameterAudioDetails: RTCRtpParameterDetails{
// Contains a list of audio codec options per possible to use codecs.
The order
// of the codecs is in preferred order.
sequence<RTCRtpParameterAudioCodecDetails>codecDetails;
};
dictionary RTCRtpParameterCodecDetails {
// The name of the codec as related to the codec name(s) contained
within the codecs
// listed within the RTCRtpCapabilitiesdictionaries.
DOMString codecName;
unsigned byte payloadType; // actual payload type sent on wire
Settings formatsParameters; // detailed settings chosen for
related codec
};
dictionary RTCRtpParameterAudioCodecDetails : RTCRtpParameterCodecDetails {
// nothing anything required at this time?
};
RTCRtpParameterVideoDetails (and related)
dictionary RTCRtpParameterVideoDetails: RTCRtpParameterDetails {
double scale = 1.0; // 0..1 relative scale from source
double frameRate = 1.0; // 0..1 relative frame rate from
source
double quality = 1.0; // 0..1 relative quality from source
// Contains a list of video codec options per possible to use codecs.
The order
// of the codecs is in preferred order.
sequence<RTCRtpParameterVideoCodecDetails> codecDetails;
};
dictionary RTCRtpParameterVideoCodecDetails : RTCRtpParameterCodecDetails {
// When layering is used, this value contains a sequence containing
the layer
// information as needed for the related codec.
sequence<RTCRtpParameterVideoLayerDetails>? layers = null;
};
dictionary RTCRtpParameterVideoLayerDetails{
// Value is set if required for describing the dependency tree
information for
// the codec's layers.
DOMString layerId = "";
// Value is nullfor the base layer or if dependencies are not needed
to be
// described (as may be the case for dynamic SCV codecs). If set, the
value
// contains a list of layers this layer is dependent upon (thus
allowing a
// dependency tree/graph to be created).
sequence<DOMString>? layerIdDependencies = null;
RTCRtpScalabilityType? layerScalabilityType = null; // null would be
for base
DOMString receiverId = ""; // use this receiver ID in
layer ("" = N/A)
unsigned int? ssrc = null; // if layer uses its own
SSRC (null = N/A)
double? frameRate = null; // framerate for layer (for
temporal SVC)
double? scale = null; // scale applied to layer
(for spatial SVC)
double? quality = null; // quality applied to layer
(for quality SVC)
DOMString fecReceiverId = ""; // receiver ID for FEC RTP
("" = N/A)
unsigned int? fecSsrc = null; // using this SSRC for FEC
(null = N/A)
Settings fec; // modes of operation
related to FEC
DOMString rtxReceiverId = ""; // receiver ID for RTX RTP
("" = N/A)
unsigned int? rtxSsrc = null; // using this SSRC for FEC
(null = N/A)
Settings rtx; // modes of operation
related to RTX
};
enum RTCRtpScalabilityType{
"temporal",
"spatial",
"quality"
};
RTCRtpParameterSimulcastDetails (and related)
dictionary RTCRtpParameterSimulcastDetails {
// This sequence contains the details of each simulcasted stream when
simulcasting
// is used or will contain exactly 1 video stream details when not
simulcasting.
sequence<RTCRtpParameterVideoDetails>? simulcastStreams;
};
RTCRtpCodec Dictionary Tweak
dictionary RTCRtpCodec{
DOMString name = "";
// Added to be able to pick payload type based upon sender or
receiver so they match
// when creating both the sender and receiver parameters.
unsigned byte preferredPayloadType;
unsigned int? clockRate = null;
unsigned int? numChannels = 1;
Capabilities formats;
};
*Proposal-ORTC Sender / Receiver Use Case [Usage Comparison Analysis]
Introduction *
After attempting to work through examples of code usage using the
current ORTC sender/receiver API, some issues, concerns and deficiencies
were discovered. A retuning of the current model was made to attempt to
address those findings. The differences are illustrated below in code
examples based on various use cases.
In the first set of use cases for simple application usages, there are
no advantages to a capabilities model (aside from the reduction of
complexity an engine might need to implement). As the use cases become
more involved, advantages begin to show. In the final example which
illustrates using SVC, the clear advantage of capabilities and
preferences can be demonstrated.
Use Cases
Alice wishes to send media to Bob
Current Parameter Based API
Step 1: (Alice)
var track = myObtainMediaTrack();
var senderCaps = RTCRtpSender.getCapabilities();
var senderParams = RTCRtpSender.createParameters(track, senderCaps);
mysignal(senderParams);
Step 2: (Bob)
var senderParams = mysignal();
var receiverCaps = RTPRtcReceiver.getCapabilities();
var receiverParams = RTPRtcReceiver.filterParameters(senderParams,
receiverCaps);
var receiver = new RTCRtpReceiver(...);
receiver.start(receiverParams);
mysignal(receiverParams);
Step 3: (Alice)
var receiverParams = mysignal();
var senderParams = RTPRtcSender.filterParameters(receiverParams,
senderCaps);
var sender = new RTCRtpSender(...);
sender.start(senderParams);
Comments
Because sender (i.e. Alice) sent her parameters that contained specific
SSRC (and possibly receiver ID) information in the her sender
parameters, the receiver will latch based upon exact SSRC matching.
Proposed Capabilities Based API
Step 1: (Alice)
var senderCaps = RTCRtpSender.getCapbilities();
mysignal(senderCaps);
Step 2: (Bob)
var senderCaps = signal();
var receiverParams = RTCRtpReceiver.createParameters("video", senderCaps);
var receiver = new RTCRtpReceiver(...);
receiver.start(receiverParams);
mysignal(receiverParams.receiverCapabilities);
Step 3: (Alice)
var track = myObtainMediaTrack();
var receiverCaps = mysignal();
var senderParams = RTCRtpSender.createParameters(track, receiverCaps);
var sender = new RTCRtpSender(...);
sender.start(senderParams);
Comments
Receiver (Bob) can match an incoming stream because the payload types
will match and therefore the incoming stream will latch to the receiver
based on payload type alone.
Alice wishes to send media to Bob Using Unhandled Eventing
Current Parameter Based API
Step 1: (Alice)
var track = myObtainMediaTrack();
var senderCaps = RTCRtpSender.getCapabilities();
var senderParams = RTCRtpSender.createParameters(track, senderCaps);
mysignal(senderParams);
Step 2: (Bob)
var senderParams = mysignal();
var receiverCaps = RTPRtcReceiver.getCapabilities();
var templateReceiverParams =
RTPRtcReceiver.filterParameters(senderParams, receiverCaps);
templateReceiverParams.encodings[0].receiverId = "";
templateReceiverParams.encodings[0].ssrc = null;
var listener = RTCRtpListener(...);
listener.onunhandledrtp = function(event) {
var receiver = new RTCRtpReceiver(...);
receiver.start(templateReceiverParams);
}
mysignal(receiverParams);
Step 3: (Alice)
var receiverParams = mysignal();
var senderParams = RTPRtcSender.filterParameters(receiverParams,
senderCaps);
var sender = new RTCRtpSender(...);
sender.start(senderParams);
Comments
Because sender (i.e. Alice) sent her parameters that contained specific
SSRC (and possibly receiver ID) information in the her sender
parameters, the receiver must override the template receiver params and
remove the exact SSRC to attach the incoming stream by payload type.
Proposed Capabilities Based API
Step 1: (Alice)
var senderCaps = RTCRtpSender.getCapbilities();
mysignal(senderCaps);
Step 2: (Bob)
var senderCaps = signal();
var listener = RTCRtpListener(...);
listener.onunhandledrtp = function(event) {
var receiverParams = RTCRtpReceiver.createParameters("video", senderCaps);
var receiver = new RTCRtpReceiver(...);
receiver.start(receiverParams);
}
mysignal(receiverParams.receiverCapabilities);
Step 3: (Alice)
var track = myObtainMediaTrack();
var receiverCaps = mysignal();
var senderParams = RTCRtpSender.createParameters(track, receiverCaps);
var sender = new RTCRtpSender(...);
sender.start(senderParams);
Comments
Receiver (Bob) can match an incoming stream because the payload types
will match and therefore the incoming stream will latch to the receiver
based on payload type alone.
Alice / Bob simultaneously exchange information in parallel
To avoid requiring a sequential offer / answer exchange, Alice and Bob
wish to simultaneously exchange their RTC information to receiver media
from the other party.
Current Parameter Based API
Step 1: (Alice / Bob)
// [Alice]
var aliceTrack = myObtainMediaTrack();
var aliceSenderCaps = RTCRtpSender.getCapabilities();
var aliceSenderParams = RTCRtpSender.createParameters(aliceTrack,
aliceSenderCaps);
var aliceReceiverCaps = RTCRtpReceiver.getCapabilities();
var aliceReceiverParams = RTCRtpReceiver.createParameters("video",
aliceReceiverCaps);
mysignal(aliceSenderParams);
mysignal(aliceReceiverParams);
// [Bob]
var bobTrack = myObtainMediaTrack();
var bobSenderCaps = RTCRtpSender.getCapabilities();
var bobSenderParams = RTCRtpSender.createParameters(bobTrack,
bobSenderCaps);
var bobReceiverCaps = RTCRtpReceiver.getCapabilities();
var bobReceiverParams = RTCRtpReceiver.createParameters("video",
bobReceiverCaps);
mysignal(bobSenderParams);
mysignal(bobReceiverParams);
Step 2: (Alice / Bob)
// [Alice]
var bobSenderParams = mysignal();
var bobReceiverParams = mysignal();
bobSenderParmas = RTCRtpReceiver.filterParams(bobSenderParams,
aliceReceiverCaps);
bobSenderParmas.encodings[0].receiverId = "";
bobSenderParmas.encodings[0].ssrc = null;
bobSenderParams = myFixPayloadTypes(bobSenderParmas, aliceReceiverParams);
var aliceReceiver = new RTCRtpReceiver(...);
aliceReceiver.receive(bobSenderParams);
bobReceiverParams = RTCRtpSender.filterParams(bobReceiverParams,
aliceSenderCaps);
var aliceSender = new RTCRtpSender(...);
aliceSender.send(bobReceiverParams);
// [Bob]
var aliceSenderParams = mysignal();
var aliceReceiverParams = mysignal();
aliceSenderParmas = RTCRtpReceiver.filterParams(aliceSenderParams,
bobReceiverCaps);
aliceSenderParmas.encodings[0].receiverId = "";
aliceSenderParmas.encodings[0].ssrc = null;
aliceSenderParams = myFixPayloadTypes(aliceSenderParmas, bobReceiverParams);
var bobReceiver = new RTCRtpReceiver(...);
bobReceiver.receive(aliceSenderParams);
aliceReceiverParams = RTCRtpSender.filterParams(aliceReceiverParams,
aliceSenderCaps);
var bobSender = new RTCRtpSender(...);
bobSender.send(aliceReceiverParams);
//---------------------------------
// [Alice and Bob need this method]
function myFixPayloadTypes(senderParams, originalReceiverParams) {
// TODO: loop through sender params and then secondarily loop through
// original receiver params and set the sender payload type based upon
// what is found in the receiver params.
// ...
return myFixedSenderParams;
}
Comments
The sender includes exact SSRC information and signals that to the
remote receiver. The issue is the actual sender is going to base it's
sending params upon the receiver params of the remote party which do not
contain a specific SSRC (or contains a different SSRC). Thus the SSRC
has to be stripped from the received sender params or they will not
match and the receiver won't latch onto the incoming stream as the
latching must occur by payload type instead.
The secondary problem is that the sender is actually using the payload
types as defined by the remote party's receiver but the receiver is
basing the payload types based upon the remote party's sender. This
means the payload types might mismatch and the latching based on payload
types may not occur. To fix this problem the web developer has to fix
either the sender's payload types or the receiver's payload type.
Proposed Capabilities Based API
Step 1: (Alice / Bob)
// [Alice]
var aliceSenderCaps = RTCRtpSender.getCapbilities();
var aliceReceiverCaps = RTCRtpReceiver.getCapabilities();
mysignal(aliceSenderCaps);
mysignal(aliceReceiverCaps);
// [Bob]
var bobSenderCaps = RTCRtpSender.getCapbilities();
var bobReceiverCaps = RTCRtpReceiver.getCapabilities();
mysignal(bobSenderCaps);
mysignal(bobReceiverCaps);
Step 2: (Alice / Bob)
// [Alice]
var bobSenderCaps = mysignal();
var bobReceiverCaps = mysignal();
var aliceTrack = myObtainMediaTrack();
var aliceReceiverParams = RTCRtpReceiver.createParameters("video",
bobSenderCaps);
var aliceReceiver = new RTCRtpReceiver(...);
aliceReceiver.receiver(aliceReceiverParams);
var aliceSenderParams = RTCRtpSender.createParameters(aliceTrack,
bobReceiverCaps);
var aliceSender = new RTCRtpSender(...);
aliceSender.send(aliceSenderParams);
// [Bob]
var aliceSenderCaps = mysignal();
var aliceReceiverCaps = mysignal();
var bobTrack = myObtainMediaTrack();
var bobReceiverParams = RTCRtpReceiver.createParameters("video",
aliceSenderCaps);
var bobReceiver = new RTCRtpReceiver(...);
bobReceiver.receiver(bobReceiverParams);
var bobSenderParams = RTCRtpSender.createParameters(bobTrack,
aliceReceiverCaps);
var bobSender = new RTCRtpSender(...);
bobSender.send(bobSenderParams);
Comments
The receiver is able to latch onto the sender based on payload type
alone. Unlike the current API, there's no need to strip SSRCs and no
need to fiddle and fix the payload type. The code is cleaner and clearer
as to what's going on and does not presume the application level
programmer has to know why payload types need to match or why SSRCs need
to be stripped.
Alice wants to use a SVC (Scalable Video Codec) to send to Bob
This is for illustration purposes only. Typical benefits of SVC are
greater in conference scenarios rather than traditional point to point
scenarios. However, this scenario can presume that an intermedia
conferencing bridge would be between Alice and Bob.
Current Parameter Based API
Step 1: (Alice)
var senderCaps = RTCRtpSender.getCapabilities();
mySignal(senderCaps);
Step 2: (Bob)
var senderCaps = mysignal();
var receiverCaps = RTPRtcReceiver.getCapabilities();
var receiverParams = RTPRtcReceiver.createParameters("video", receiverCaps);
var receiverParams = RTPRtcReceiver.filterParams(senderCaps);
var receiverParams = mySetupSVC(receiverParams);
var receiver = new RTCRtpReceiver(...);
receiver.start(receiverParams);
mysignal(receiverParams);
function mySetupSVC(receiverParams) {
// 1. search the receiver params for a codec capable of SVC based on
pre-knowledge
// of the codec types
// 2. setup SVC params based on codec's capabilities
// TODO - step 1 - code needs to be added here to do this logic
var chosenCodec = "h264svc"; // hard code for now
// TODO: Not sure this code is even right. How does this layer scale
even work?
// How is temporal and spatial layering defined together? Don't see a
knob for
// setting up temporal SVC...
receiverParams.receiverId = "foo";
receiverParams.encodings[0] = {
"codecName": chosenCodec,
"scale": 0.125,
"encodingId": "0"
};
receiverParams.encodings[1] = {
"scale": 0.25,
"dependencyEncodingIds": {"0"}
};
receiverParams.encodings[2] = {
"scale": 0.5,
"dependencyEncodingIds": {"0", "1"}
};
}
Step 3: (Alice)
var receiverParams = mysignal();
var senderParams = RTPRtcSender.filterParameters(receiverParams,
senderCaps);
var track = myObtainMediaTrack();
var sender = new RTCRtpSender(track, ...);
sender.start(senderParams);
Comments
The application developer has to have a ton of presumed knowledge about
available codecs, codec capabilities and needs to have a deep
understanding of how the engine interprets the layering information. The
sender cannot setup the SVC parameters desired because it doesn't know
the receiver capabilities.
The sample above may not work for SVC codecs which put each layer on a
unique SSRC because the receiver did not necessarily pre-dictate the
expected SSRCs on each layer so the application developer would have to
handle this situation too and assign SSRCs for each layer manually based
on knowledge that the codec behaves in this manner.
The method to setup temporal or quality SVC is unclear. Appropriate
parameter knobs for the application developer appear to be missing.
Proposed Capabilities Based API
Step 1: (Alice)
var senderCaps = RTCRtpSender.getCapbilities();
var senderPrefs = {
"receiverId": "foo",
"frameRateScalabilityOptions": {"layers": 2},
"scalingScalabilityOptions": {"layers": 2},
};
mysignal(senderCaps);
mysignal(senderPrefs);
Step 2: (Bob)
var senderCaps = signal();
var senderPrefs = signal();
var receiverParams = RTCRtpReceiver.createParameters("video",
senderCaps, senderPrefs);
var receiver = new RTCRtpReceiver(...);
receiver.start(receiverParams);
mysignal(receiverParams.receiverCapabilities);
Step 3: (Alice)
var track = myObtainMediaTrack();
var receiverCaps = mysignal();
var senderParams = RTCRtpSender.createParameters(track, receiverCaps,
senderPrefs);
var sender = new RTCRtpSender(track, ...);
sender.start(senderParams);
Comments
The application developer doesn't require pre-knowledge of the codecs.
The developer can quickly and easily specify the types of SVC properties
desired with much simpler knobs. The developer doesn't have to worry if
a codec is assigning each layer a unique SSRC or not of if the layering
ends up being dynamic or not.
Conclusion
Overall the proposed capabilities based API has strong advantages. Main
advantages are:
1.
Simplicity in setup based on "preferences" for the application developer
2.
Less brittle designs/implementations since low level parameters are
not exchanged, filtered, and interpreted by different browser engines
3.
Much less knowledge (and often no pre-knowledge) is required for the
application developer to take full advantage of a browser's capabilities
There's no strong reason to maintain the current API. The biggest
difference will be that browsers will need to generate compatible
parameters based on capabilities but that also comes at a big advantage
of the browser engines not needing to interpreting and filtering low
level parameters from other browser engines. Both new and current use
low level parameters to receive or send information so that design
aspect remains unchanged.
Advantages of Current Parameter Based API
1.
Browser engines do not need to generate parameter from capabilities
in a "compatible" manner (although low level parameters do need to
be filtered in a "compatible" manner so this is not a strong advantage).
Disadvantages of Current Parameter Based API
1.
Application developer needs pre-knowledge of SVC codecs to be able
to chose and setup their properties based upon pre-knowledge of
codec capabilities
2.
Application developer needs deep understanding of how layering works
to setup the layering properties correctly
3.
Browser engines need to agree on how to filter low level parameters
based upon capabilities in a consistent manner across browsers to
ensure compatibility
4.
Browser engines need to agree how to interpret low level parameter
objects that were generated by other browsers (or other applications)
5.
Low level parameter based exchanges introduce greater brittleness
between browsers since extending the parameters details could mean
breaking existing implementations (instead of capabilities which are
typically ignored when not understood)
6.
Less innovation / greater brittleness for anything that requires
parameter object extensions since many browsers as well as
applications will be fiddling, exchanging, and filtering these low
level parameter objects.
7.
Simulcasting with layering doesn't appear to be supported or it's
not obvious how to set up those scenarios.
8.
Unclear how to mix and match different SVC modes (e.g. temporal,
spatial, and quality)
9.
The application developer is uncertain based upon their preferences
what the browser engine is capable of delivering (without deep
understanding of all codecs and their properties).
10.
Header extensions will need manual setup by the application
developer despite not knowing that codecs or the engines might need
certain extensions to take advantage of codec features or browser
engine features.
Advantages Proposed Capabilities Based API
1.
Application developer can easily setup SVC without needing detailed
understanding
2.
Typical and even advanced use cases do not require a deep understand
of RTC to be able to take advantages of capabilities
3.
Less brittle implementations as low level parameter objects are only
consumed local by the browsers that generate them or only in
situations where specific compatibilities with legacy systems are
required which the default generated low level properties read would
not be compatible.
4.
Simulcast with layering is supported
5.
Easy for application developer to mix and match different SVC modes
(e.g. temporal, spatial, and quality)
6.
Easy to extend support for alternative SVC scalability modes (e.g.
colour depth, sharpness, ROI)
7.
Application developer knows what the browser engine is capable of
delivering given a set of preferences (from resultant preferences as
returned from "createParameters(...)"
8.
Header extensions can be automatically set up based on needs and
capabilities of the browser's RTP engines and codecs.
Disadvantages Proposed Capabilities Based API
1.
Browser engines need to agree on how to compute "compatible"
parameters for a given codec and media preferences. The rules for
generation of parameters must be clear.
Equal Capabilities of Current and Proposed Based API
1.
Application developer can always tweak low level properties on an
"as needed" basis for compatibility
2.
Both new and current proposals send and receive based on lower level
parameters (this does not change).
*
*
*
Received on Thursday, 8 May 2014 04:13:48 UTC