A little late, but I am told that the Apple USB C dongle is a standard USB Audio Class device and works in Windows. The Lightning version, probably not since that port is proprietary to iPhones. I would guess that the internals are identical.
I would have much higher confidence in the performance of the Apple adapter for the extra $3 since it's been measured by more than one person and isn't some random eBay PCB.
Last edited:
In that context, I beseech you to see that they get reserved for those (noble) uses, where replacements are no longer produced. Otherwise some dangerous clown will want to use them for a Toob amplifier. Where they perform poorly anyway.
Much thanks, as always,
Chris
Pretty hard to screw up implementation of PCM2704 I should think....ok ceramic caps is probably the main offender, layout should be easy.
Dan.
Thanks Demian, I am aware of the issues that is why I asked for a ready download.....or Scott W, you have mentioned a couple of times, perhaps you know of a ready download source.
Dan.
Probably true, but I would guess the unknown chip in the Apple adapter outperforms PCM2704.
Is the Apple headphone adapter an Apple approved product ?....for $8.00 I expect not so no such inferences can be made.
Dan.
Sir, you are a free style philosopher and, if I may add, the depth of your thinking approaches the dimensions of said atom. But here, we are discussing engineering and human perception. If you want to contribute, I humbly suggest you study some of the finer points being discussed.
Sorry,but no; the former two statements are correct, yours neglects the conditions needed - for example waiting for the system to settle after starting the excitation with the sine wave - but, as you are now realizing how ridiculous your "FUD shouting" was, it is still kind of progress.
Btw, this was your recent statement about the meaning of "steady state" :
Not quite the same meaning that it really has and that you now - after reading my quote about it - remember.
Scott Wurcer's hint was correct; although it doesn't always need "1000s of pages", quotes from the relevant literature often help to remind of the engineering basics. 😉
And, as said before, these things happen to all of us from time to time, and if you simply would realize that, by dropping your usual "FUD shouting", the insults and by restraining the constant grumbling, all of that could be settled (no pun intended 🙂 ) in short time and would lead to discussions that are more fun and more fruitful than they are today.
You use a fairly good set of software tools, iirc - can you not create one in that?
It is a useful measure, but does not help that much if it comes to conclusions about the perceptual detectibility. Besides the obvious ones, like in case of zero distortions and way to much distortion.
As said before, forum discussions are more difficult compared to "face to face" events; try to find out what your discussion partner means - and btw, it really helps to avoid "fourier denial" "audiophools" " magic" and other phrases/terms that are at least implicitely insulting - and don't take it as a given that he must be wrong about everything.
The "I suspect" is the crucial point; I'd say most, if not all, would not object.
And it is quite similar to the point Putzeys made in the the "audiophilestyle forum interview" . (And surely previously at other occasions)
That is directly related to "try to understand what your discussion partner means" , as it could be just a simple talking past each other.
Although I can't read mmerrill99's mind, but could it be that he just means - with bringing up the "it's a model" line - the conclusions from the Fourier approach?
Leaving aside the constraints for signals (Gibb's phenomenon for example) for the moment, what you've said above about the two or three sinewaves that might tell it all, is an example for a somewhat crude description of a model for applying conclusions from "Fourier" to the generalized question of perceptable differences.
Anything is possible, though his apparent misunderstanding seems deeper than that, why don't you ask him if he thinks music is made up of sine waves and it isn't a model? It is important that he understands this is the case IMHO
Nobody said that FFT is reality; maybe that is your confusion: whenever we say "Fourier" you think 'FFT'? FFT is a popular method for calculating an approximate Fourier transform. It is approximate because it is essentially a sampled method in the frequency domain. If you are purely in the frequency domain then temporal spread has no meaning. Just like in the pure time domain frequency spread has no meaning. That is why various wavelet options have been invented, so time and frequency can be considered together. However, as I keep saying, all these different domains are merely different ways of viewing the same reality. None of them are models, although they can be used to construct models.mmerrill99 said:
There is an excellent paper on limiting crest factor on multi-tone signals in the IEEE collection. I wrote a simple procedure to set crest factor on a 256 QUAM multi-tone by iteration and phase rotation of the tones without modifying their amplitude. IIRC there is not a closed form solution for minimum crest factor but the IEEE paper gave an approximate formula for the phases on equal amplitude tones for minimum crest factor.
Last edited:
Perhaps if knowing all the rules, however it is done already, why not use an accepted error free resource/test methodology ?,.
Dan.
I posted a set of multi-tones on the Linear Audio site, IIRC 48K is there corrected for unwindowed 64K FFT's for those using common tools. Windowing will only thicken the bins obscuring the low end, but multi-tones ONLY work at the target sample rate.
Ok, thanks for reinforcing what I already know, I am interested in rough amplitudes of such sidebands.
Get that pencil sharpened then! 🙂
edit: Fourier transform of damped sine wave returns a lot of google results....
edit: Fourier transform of damped sine wave returns a lot of google results....
Last edited:
You can buy it directly from Apple for 1 dollar more... it is their product. It’s an iPad accessory since the new one only has a USB C port.
When we say that the frequency of something is 1kHz we have to say over what time period for this to have any meaning. Unless, of course, we are in the pure frequency domain in which case the 1kHz lasts for ever but it is probably only one component of the total signal. There is a form of uncertainty principle in operation: the closer we specify the frequency (i.e. narrower bandwidth of either the signal or our apparatus) the less we know about the duration of the signal or when it happened. Most of the time we work quite happily with a mix of frequency and time and don't even think about it. This can lead to problems, like the debates last century about whether the sidebands of AM were real or 'mathematical'; they are, of course, both in the sense that the maths says they are present and careful measurement confirms this. If you worked only in the time domain you would see the carrier amplitude varying and would know nothing about sidebands; if you worked only in the frequency domain you would see the sidebands and know nothing about the modulation - yet both views contain a complete description of what is happening. Both are real (not models).
Just to confuse some people you can tell them that in AM the carrier does not vary in amplitude (but sidebands appear) whereas in FM the carrier amplitude (but not its frequency) depends on the modulation.
Just to confuse some people you can tell them that in AM the carrier does not vary in amplitude (but sidebands appear) whereas in FM the carrier amplitude (but not its frequency) depends on the modulation.
- Status
- Not open for further replies.