I have read the section on The AnalyserNode Interface on the W3C docs, which states that the AnalyserNode will pass the input audio to the output untouched. It also describes the process of computing its "current frequency data".
I am wondering whether this processing of the input audio is done continuously or on-demand, f.i. when getFloatFrequencyData() is called.
Does anyone know? Is it browser specific?
It might depend on the browser, but as a performance optimization, it could be done only on demand. Of course, the node needs to keep enough information around to do the computation on demand, but that should be much cheaper than continuously computing the frequency data.
I asked Paul Adenot who was kind enough to link me to the source of Chrome and Firefox respectively:
Chrome: Updated when requested. https://cs.chromium.org/chromium/src/third_party/WebKit/Source/modules/webaudio/RealtimeAnalyser.cpp?cl=GROK&gsn=discreteTimeConstantForSampleRate&rcl=1469603944&l=209
Firefox: Updated when requested. http://searchfox.org/mozilla-central/source/dom/media/webaudio/AnalyserNode.cpp#197
Related
In firefox, I'm able to request a video stream of a window with
navigator.mediaDevices.getUserMedia({
video: {
mediaSource: 'window',
},
})
This produces a dialog like this:
I only care about the current window. Is there a way to specify in my call to getUserMedia that I would like the current tab (or window) only?
I don't think so no...
What FF implements here is not really specified yet, but w3c is working on a new API which will take care of Screen Capture: MediaDevices.getDisplayMedia.
While it's not what Firefox did implement, there is a clear paragraph in this paper about why deviceId can't and will not work on such requests:
Each potential source of capture is treated by this API as a discrete media source. However, display capture sources MUST NOT be enumerated by enumerateDevices, since this would reveal too much information about the host system.
Display capture sources therefore cannot be selected with the deviceId constraint, since this would allow applications to influence selection; setting deviceId constraint can only cause the resulting MediaStreamTrack to become overconstrained.
So, once again even if FF does not implement this API yet, we can assume they do follow this same rule for their current implementation, for the same reasons.
However, what will apparently be possible to do when this API will come to life, is to use the "browser" constraint, instead of "window". While the specs are not really clear as to what it is exactly ("a browser display surface, or single browser window"), I guess it will be closer to what you want than "window", and someone even asked 2 days ago for a "tab" constraint, could we even hope for a "current-tab" constraint? That might need someone to open an issue on w3c github's page.
I've written a script that runs on a domain and fetches images from various other domains (3rd party resources).
I'm trying to use window.performance.getEntriesByType('resource') to get a general health check of things. It seems, since these resources are on other domains that the responses would need to have Timing-Allow-Origin set in the response header to get timing data via window.performance.getEntriesByType().
Is this true?
Further, when I run my script, Chrome Browser does return useful information. Indeed, I could use this data if I could get at it programmatically. But the data Chrome displays and the data returned in the window.performance.getEntriesByType() differ.
I've attached a screenshot, which shows Chromes useful timing breakdown of loading the resource. By the performance entry object's data doesn't match.
For example, see the DNS Lookup time in the timing graph on the right, and then look at the domainLookupStart and domainLookupEnd values in the performance entry object. These values don't match up with each other.
Why is there a discrepancy and how can I get at Chrome's data? How do I derive what Chrome is displaying from the performance entry object?
Thanks!
You've probably figured this out by now, but I had a similar question and found this.
Most detailed fields in a PerformanceResourceTiming object report zero for cross-origin resources that do not set the Timing-Allow-Origin header, as per the spec:
connectStart must return zero unless timing allow check algorithm
passes.
Similarly for other fields, such as the DNS lookup fields.
As for why the developer console allows you to see this information even if you can't access it programmatically, it's just a feature of Chrome that allows you to see this information. Hiding it is more of a courtesy than a security feature; the spec dictates what can be shared via the Resource Timing API, but the browser still has access to the information and may decide to share it with the user in other ways, as you have seen.
First of all, I've looked around the internet and found it quite badly documented.
Somewhere in my code I have a big memory leak that I'm trying to track and after using:
window.performance.memory.usedJSHeapSize
it looks like the value remains at the same level of 10MB, which is not true because when we compare to the values either visible here:
chrome://memory-internals/
or if we look at the Timeline in devTools we can see a big difference. Does anyone encountered a similar issue? Do I need to manually update these values (to run a command "update", "measure" etc?)
Following this topic:
Information heap size
it looks like this value is increased by a certain step, can we somehow see what is it or modify it? In my case from what I can see now the page has about 10MB, 30 minutes later there will be about 400MB, and half an hour after the page will crash..
Any ideas guys?
(Why the code is leaking it's a different issue, please treat this one as I was trying to use this variable to create some kind of test).
There's a section of the WebPlatform.org docs that explains this:
The values are quantized as to not expose private information to attackers. If Chrome is run with the flag --enable-precise-memory-info the values are not quantized.
http://docs.webplatform.org/wiki/apis/timing/properties/memory
So, by default, the number is not precise, and it only updates every 20 minutes! This should explain why your number doesn't change. If you use the flag, the number will be precise and current.
The WebKit commit message explains:
This patch adds an option to expose quantized and rate-limited memory
information to web pages. Web pages can only learn new data every 20
minutes, which helps mitigate attacks where the attacker compares two
readings to extract side-channel information. The patch also only
reports 100 distinct memory values, which (combined with the rate
limits) makes it difficult for attackers to learn about small changes in
memory use.
I am working on web workers and I am passing large amount of data to web worker, which takes a lot of time. I want to know the efficient way to send the data.
I have tried the following code:
var worker = new Worker('js2.js');
worker.postMessage( buffer,[ buffer]);
worker.postMessage(obj,[obj.mat2]);
if (buffer.byteLength) {
alert('Transferables are not supported in your browser!');
}
UPDATE
Modern versions of Chrome, Edge, and Firefox now support SharedArrayBuffers (though not safari at the time of this writing see SharedArrayBuffers on MDN), so that would be another possibility for a fast transfer of data with a different set of trade offs compared to a transferrable (you can see MDN for all the trade offs and requirements of SharedArrayBuffers).
UPDATE:
According to Mozilla the SharedArrayBuffer has been disabled in all major browsers, thus the option described in the following EDIT does no longer apply.
Note that SharedArrayBuffer was disabled by default in all major
browsers on 5 January, 2018 in response to Spectre.
EDIT: There is now another option and it is sending a sharedArray buffer. This is part of ES2017 under shared memory and atomics and is now supported in FireFox 54 Nightly. If you want to read about it you can look here. I will probably write up something some time and add it to my answer. I will try and add to the performance benchmark as well.
To answer the original question:
I am working on web workers and I am passing large amount of data to
web worker, which takes a lot of time. I want to know the efficient
way to send the data.
The alternative to #MichaelDibbets answer, his sends a copy of the object to the webworker, is using a transferrable object which is zero-copy.
It shows that you were intending to make your data transferrable, but I'm guessing it didn't work out. So I will explain what it means for some data to be transferrable for you and future readers.
Transferring objects "by reference" (although that isn't the perfect term for it as explained in the next quote) doesn't just work on any JavaScript Object. It has to be a transferrable data-type.
[With Web Workers] Most browsers implement the structured cloning
algorithm, which allows you to pass more complex types in/out of
Workers such as File, Blob, ArrayBuffer, and JSON objects. However,
when passing these types of data using postMessage(), a copy is still
made. Therefore, if you're passing a large 50MB file (for example),
there's a noticeable overhead in getting that file between the worker
and the main thread.
Structured cloning is great, but a copy can take hundreds of
milliseconds. To combat the perf hit, you can use Transferable
Objects.
With Transferable Objects, data is transferred from one context to
another. It is zero-copy, which vastly improves the performance of
sending data to a Worker. Think of it as pass-by-reference if you're
from the C/C++ world. However, unlike pass-by-reference, the 'version'
from the calling context is no longer available once transferred to
the new context. For example, when transferring an ArrayBuffer from
your main app to Worker, the original ArrayBuffer is cleared and no
longer usable. Its contents are (quiet literally) transferred to the
Worker context.
- Eric Bidelman Developer at Google, source: html5rocks
The only problem is there are only two things that are transferrable as of now. ArrayBuffer, and MessagePort. (Canvas Proxies are hopefully coming later). ArrayBuffers cannot be manipulated directly through their API and should be used to create a typed array object or a DataView to give a particular view into the buffer and be able to read and write to it.
From the html5rocks link
To use transferrable objects, use a slightly different signature of
postMessage():
worker.postMessage(arrayBuffer, [arrayBuffer]);
window.postMessage(arrayBuffer, targetOrigin, [arrayBuffer]);
The worker case, the first argument is the data and the second is the
list of items that should be transferred. The first argument doesn't
have to be an ArrayBuffer by the way. For example, it can be a JSON
object:
worker.postMessage({data: int8View, moreData: anotherBuffer}, [int8View.buffer, anotherBuffer]);
So according to that your
var worker = new Worker('js2.js');
worker.postMessage(buffer, [ buffer]);
worker.postMessage(obj, [obj.mat2]);
should be performing at great speeds and should be being transferred zero-copy. The only problem would be if your buffer or obj.mat2 is not an ArrayBuffer or transferrable. You may be confusing ArrayBuffers with a view of a typed array instead of what you should be using its buffer.
So if you have this ArrayBuffer and it's Int32 representation. (though the variable is titled view it is not a DataView, but DataView's do have a property buffer just as typed arrays do. Also at the time this was written the MDN use the name 'view' for the result of calling a typed arrays constructor so I assumed it was a good way to define it.)
var buffer = new ArrayBuffer(90000000);
var view = new Int32Array(buffer);
for(var c=0;c<view.length;c++) {
view[c]=42;
}
This is what you should not do (send the view)
worker.postMessage(view);
This is what you should do (send the ArrayBuffer)
worker.postMessage(buffer, [buffer]);
These are the results after running this test on plnkr.
Average for sending views is 144.12690000608563
Average for sending ArrayBuffers is 0.3522000042721629
EDIT: As stated by #Bergi in the comments you don't need the buffer variable at all if you have the view, because you can just send view.buffer like so
worker.postMessage(view.buffer, [view.buffer]);
Just as a side note to future readers just sending an ArrayBuffer without the last argument specifying what the ArrayBuffers are you will not send the ArrayBuffer transferrably
In other words when sending transferrables you want this:
worker.postMessage(buffer, [buffer]);
Not this:
worker.postMessage(buffer);
EDIT: And one last note since you are sending a buffer don't forget to turn your buffer back into a view once it's received by the webworker. Once it's a view you can manipulate it (read and write from it) again.
And for the bounty:
I am also interested in official size limits for firefox/chrome (not
only time limit). However answer the original question qualifies for
the bounty (;
As to a webbrowsers limit to send something of a certain size I am not completeley sure, but from that quote that entry on html5rocks by Eric Bidelman when talking about workers he did bring up a 50 mb file being transferred without using a transferrable data-type in hundreds of milliseconds and as shown through my test in a only around a millisecond using a transferrable data-type. Which 50 mb is honestly pretty large.
Purely my own opinion, but I don't believe there to be a limit on the size of the file you send on a transferrable or non-transferrable data-type other than the limits of the data type itself. Of course your biggest worry would probably be for the browser stopping long running scripts if it has to copy the whole thing and is not zero-copy and transferrable.
Hope this post helps. Honestly I knew nothing about transferrables before this, but it was fun figuring out them through some tests and through that blog post by Eric Bidelman.
I had issues with webworkers too, until I just passed a single argument to the webworker.
So instead of
worker.postMessage( buffer,[ buffer]);
worker.postMessage(obj,[obj.mat2]);
Try
var myobj = {buffer:buffer,obj:obj};
worker.postMessage(myobj);
This way I found it gets passed by reference and its insanely fast. I post back and forth over 20.000 dataelements in a single push per 5 seconds without me noticing the datatransfer.
I've been exclusively working with chrome though, so I don't know how it'll hold up in other browsers.
Update
I've done some testing for some stats.
tmp = new ArrayBuffer(90000000);
test = new Int32Array(tmp);
for(c=0;c<test.length;c++) {
test[c]=42;
}
for(c=0;c<4;c++) {
window.setTimeout(function(){
// Cloning the Array. "We" will have lost the array once its sent to the webworker.
// This is to make sure we dont have to repopulate it.
testsend = new Int32Array(test);
// marking time. sister mark is in webworker
console.log("sending at at "+window.performance.now());
// post the clone to the thread.
FieldValueCommunicator.worker.postMessage(testsend);
},1000*c);
}
results of the tests. I don't know if this falls in your category of slow or not since you did not define "slow"
sending at at 28837.418999988586
recieved at 28923.06199995801
86 ms
sending at at 212387.9840001464
recieved at 212504.72499988973
117 ms
sending at at 247635.6210000813
recieved at 247760.1259998046
125 ms
sending at at 288194.15999995545
recieved at 288304.4079998508
110 ms
It depends on how large the data is
I found this article that says, the better strategy is to pass large data to a web worker and back in small bits. In addition, it also discourages the use of ArrayBuffers.
Please have a look: https://developers.redhat.com/blog/2014/05/20/communicating-large-objects-with-web-workers-in-javascript
I have a server that generates pngs very rapidly and I need to make this into a poor-man's video feed. Actually creating a video feed is not an option.
What I have working right now is a recursive loop that looks a little like this (in pseudo-code):
function update() {
image.src = imagepath + '?' + timestamp; // ensures the image will update
image.onload = function () {update()};
}
This works, however after a while, it crashes the browser (Google Chrome, after more than 10 minutes or so). These images are being updated very frequently (several times a second). It seems the images are caching, which causes the browser to run out of memory.
Which of these solutions would solve the problem while maintaining fast refresh:
HTML5 canvas with drawImage
HTML5 canvas with CanvasPixelArray (raw pixel manipulation)
I have access to the raw binary as a Uint8Array, and the image isn't too large (less than 50 kb or so, 720 x 480 pixels).
Alternatively, is there anyway to clear old images from the cache or to avoid caching altogether?
EDIT:
Note, this is not a tool for regular users. It's a tool for diagnosing analog hardware problems for engineers. The reason for the browser is platform independence (should work on Linux, Windows, Mac, iPad, etc without any software changes).
The crashing is due to http://code.google.com/p/chromium/issues/detail?id=36142. Try creating object URLs (use XHR2 responseType = "arraybuffer" along with BlobBuilder) and revoking (using URL.revokeObjectURL) the previous frame after the next frame is loaded.
Edit: You really should be processing these into a live low-fps video stream on the server side, which will end up giving you greatly decreased latency and faster load times.
#Eli Grey seems to have identified the source of your crashing. It looks like they have a fix in the works, so if you don't want to modify your approach hopefully that will be resolved soon.
With regard to your other question, you should definitely stick with drawImage() if you can. If I understand your intention of using the CanvasPixelArray, you are considering iterating over each pixel in the canvas and updating it with your new pixel information? If so, this will be nowhere near as efficient as drawImage(). Furthermore, this approach is completely unnecessary for you because you (presumably) do not need to reference the data in the previous frame.
Whether fortunately or not, you cannot directly swap out the internal CanvasPixelArray object stored within an HTML5 canvas. If you have a properly-formatted array of pixel data, the only way you can update a canvas element is by calling either drawImage() or putImageData(). Right now, putImageData() is much slower than drawImage(), as you can see here: http://jsperf.com/canvas-drawimage-vs-putimagedata. If you have any sort of transparency in the frames of your video, you will likely want to clear the canvas and then use drawImage() (otherwise you could see through to previous frames).
Having said all that, I don't know that you really even need to use a canvas for this. Was your motivation for looking into using a canvas so that you could avoid caching (which now doesn't seem to be the underlying issue for you)?
If the "movie" is data-driven (ie. based on numbers and calculations), you may be able to push MUCH more data to the browser as text and then have javascript render it client-side into a movie. The "player" in the client can then request the data in batches as it needs it.
If not, one thing you could do is simply limit the frames-per-second (fps) of the script, possibly a hard-coded value, or a slider / setable value. Assuming this doesn't limit the utility of the tool, at the very least it would let the browser run longer w/o crashing.
Lastly, there are lots of things that can be done with headers (eg. in the .htaccess file) to indicate to browsers to cache or not cache content.
iPad, you say ?.. Nevertheless, i would advice using Flash/video or HTML5/video.
Because WebKit is very easily crashed with even moderate influx of images, either just big images or just a huge number of small ones..
From the other side, XHR with base64 image data or pixel array MIGHT work. I have had short polling app, which was able to run for 10-12 hours with XHR polling server every 10 seconds.
Also, consider delta compression, - like, if its histogram with abscissa being time scale - you can only send a little slice from the rigth, - of course, for things like heat-maps, you cannot do that.
These images are being updated very frequently (several times a
second).
.. if its critical to update at such a high rate - you MUST use long polling.