Separate names with a comma.
Couldn't agree more.
Or to sum it up in one word: Chesky.
You changed the topic. The original post was about transmission, comparing music streaming sources.
No, there is no point to transfer or backup music in any format beyond 16/44.1, there is no musical information there that human can hear. Yes, it makes sense to up-sample the 16/44 to 32/88 between CD and DAC to get around filtering as you described. You do this upsampling locally in your music system without adding any new information to the source. I hope the difference is understood.
I can't remember a post from Jason or Baldr giving any hint about it, but the other day I was thinking about Sol and to me it sounds pretty logical to take Sol as the perfect moment to introduce a Mani+
I beg to diiffer: Yes there is definately a difference between a 24/96 and a 16/44.1 sample rare and the transmission rates are also that high to get around filtering artifacts.
Yes it is possible to hear beyond the 16bits depht and the human system senses frequencies beyond 20 kHz as well. Ever listened to a high end analogue recording on record or master tape? They go well beyond 20kHz and this can be noticed.
By the way this is the Schiit thread and this is off topic indeed.
Ok, offtopic. I will not say anything except this link: https://people.xiph.org/~xiphmont/demo/neil-young.html . It explains everything with math.
I am a MSc in electronics so I know all that.
Maybe you should try listening.
I'm a little confused. "Don't backup beyond 16/44.1" but then you say:
So what's the difference between "upsampling locally" and upsampling the CD rip? The one thing I do understand is that upsampling the rip doesn't add any new information to the source.
Based on what you're saying, though, it seems that if I upsample the CD rip it will present the DAC with a 96/24 (or 88/32, whatever) file that would require less/no filtering. Wouldn't my DAC (Modi Multibit) "prefer" a 96/24 rip vs. 44/16? An added bonus would be requiring less of the hardware/software of the device serving up the files. No real-time encoding required. My point is that it wouldn't seem to matter where the upsampling takes place.
I've got a massive amount of hard drive space available so the increased file size isn't a concern.
I'm not trying to bait an argument, I just have a massive knowledge gap in this area.
These might be interesting to everyone.
High-Resolution Audio Demystified
And High-End Audio: Production And Playback Methods.
@Pietro Cozzi Tinin - could Constanza provide an interlude please?
@bosiemoncrieff - have you sorted your EK43 calibration yet? Visited my favourite coffee shop yesterday and then had two of those beasts on the counter. One dialled for espresso, one for filter (3 varieties of each on the menu, as per the norm).
Yes, please. Let's get this thread back on track with Cat pics and Coffee talk.
The Modi Multibit is a 16 bit DAC. It can play higher, but it rounds off or truncates anything beyond 16 bits.
Not my case. I have a few other things to buy before then and have no problem running Windows 7 as long as i need to
Good question. I would say I'm making progress. I made a variety of assumptions about espresso when I had the Breville smart grinder pro that were predicated on that grinder's inferior ability. First, I assumed you needed more espresso grounds in the portafilter than in fact you do — I got up to 23g and above, with some of the staler beans. I also assumed you needed to cut off shots at 30g in the cup.
With EK, I first needed to calibrate it to a 0 point that was in fact the 0 point of the burrs—just a hair from touching. Previously I ground far too coarsely and, surprise of surprises, the resistance was nowhere near sufficient, and my espresso machine registered 5 bars of pressure. After calibrating, I took the grind down to .5, and the 22g at that fineness choked the BDB. I went down to 21, 20.5, and now 20.3, and have taken the grind up from .75 to .9 and the grind is still causing the BDB a bit of constipation in the first phase of the shot. I think I will hold the grind where it is, but dose down to 19.5 and see if we get a shot that's better.
Certainly, the dosing tool I got (see Mike's thread) is helping get a very good grind distribution within the portafilter.
With respect to shot volume, I found with smart grinder that anything past 30 or 33 grams became astringent very quickly. However, with EK, I have found that longer shots are much more palatable. I had 40g this morning, and will experiment with 45 and 50g in the future.
Although EK43 is an excessive purchase for most people, the niche zero sells for $613 in the US (though this number jumps around, albeit slightly), and people have said it's 90-95% as good—a lot smaller, a lot easier to work with, a lot more forgiving, a lot quieter, and just a more rational home barista purchase.
I still need to align (or have aligned) my EK43 burrs, as I'm quite sure they're out of alignment, and will produce even finer espresso when they are aligned. I'm also still frustrated by phantom chirping - the machine's tendency to make the burrs produce the sound of touching when there is still room to go before they actually hit their 0 point. Apparently the machine is afflicted by a degree of drift, that is, the zero point drifts over time and with use. I had hoped that given my use of the machine is extremely low, this would not happen as readily as in a coffee shop, but alas it appears not.
After grinding for French press yesterday, bringing it back close to the 0 point caused some chirping on the way, and I am of course very sensitive to harming the machine by having the burrs grind against each other, but I pressed on and got the burrs to the proper point, even if it did come with the sensation of grinding ice in my teeth. I will say that with a true 0 point, even the coarsest setting is too fine for French press, so I'm taking the French press and the smart grinder pro to work, and will use them as my "work rig."
I mean for the Internet music service it doesn't make any sense to transfer audio over expensive Internet with quality beyond 16/44, because that's sufficient information that human can hear. Transferring faster than this will cost more but add nothing. Then, if your equipment likes it, you can upsample the 16/41 original and store it as 32/82, which obviously doesn't add any new information to the music. I don't know which DACs work better with 32/82, some do. Only those who tried can answer this. I have Khadas board, it does sound a bit different after upsampling. I think it's better, but I'm not sure, I do listen it in 32 bits mode.