iFi Hip-DAC Impressions Thread
Feb 20, 2020 at 3:58 PM Post #16 of 768
Just an FYI for those looking to use sensitive IEMs with this to look elsewhere. Both the Campfire Audio Andromeda and Solaris are unusable due to the noise floor of the unit. Also encountered hiss on a pair of UE 18+ too.

Yeah, that sounds about right. No issues with the Audeze iSine 10, though it's not particularly sensitive. But maybe someone can loan me the Andromeda, Solaris, and UE 18+ just to make sure. :ksc75smile:

numbers are fun to talk about, but I will let my ears be the judge

Numbers can help determine whether it'll meet your raw volume requirements. I listen well below 106dB, and if I can get somewhere close to my max tolerance (probably in the mid 90dB, I expect) for peak power, I'm golden. That number won't tell me whether I like the product, but that it can properly drive the gear I'll use with it.

And I hope you like the Hip-DAC!
 
Feb 20, 2020 at 4:02 PM Post #17 of 768
Yeah, that sounds about right. No issues with the Audeze iSine 10, though it's not particularly sensitive. But maybe someone can loan me the Andromeda, Solaris, and UE 18+ just to make sure. :ksc75smile:



Numbers can help determine whether it'll meet your raw volume requirements. I listen well below 106dB, and if I can get somewhere close to my max tolerance (probably in the mid 90dB, I expect) for peak power, I'm golden. That number won't tell me whether I like the product, but that it can properly drive the gear I'll use with it.

And I hope you like the Hip-DAC!

i have it charging right now and later today I will run it with a few different sets I own. P1, S8, and if I can dig them out a 2nd gen 18+ pro.

like I stated earlier that numbers are a good topic of discussion, but 35 years in the game taught me the ears rule my wallet
 
Feb 21, 2020 at 2:15 PM Post #18 of 768
Pleasant surprise today... spent yesterday with IEMs and today i went for some full sized cans. right now DT1990 Pro is responding nicely to a bit of Power Match although Xbass is a bit too much. with Power Match off i have to max out the volume, but with it on there is a bit of headroom to play with. First 24 hours shows promise. Fingers crossed that the battery run time is close to advertised and not what a few reviewers mentioned. i did get a bit concerned when the light went from white to green in an hour, but i have another 4.5 hours and still green


Battery Life Update - Just hit red between 10-25% at the 5.5 hour mark. 12 hours is not gonna happen on this run, i would guess lucky to hit 7.5

Also MQA and high res PCM activate the proper audio format LED color, but DSD64 shows up as green instead of Cyan (.dsf file)
 
Last edited:
Feb 23, 2020 at 11:16 PM Post #26 of 768
I was going to compare Hip Dac to my Nx4 Dsd if I was fortunate enough to get chosen. It sounds like Nx4 will win the battery show down by a lot.

But something tells me the Bur Brown maybe doing like a class A kind of thing to make the sound better. And class A is bad for battery life.

I realize class A is amp talk and Bur Brown is a signal converter. But you know what I mean.
 
Last edited:
Feb 23, 2020 at 11:32 PM Post #27 of 768
I was going to compare Hip Dac to my Nx4 Dsd if I was fortunate enough to get chosen. It sounds like Nx4 will win the battery show down by a lot.

the battery life is the only thing that doesn't match what is being advertised. plenty of power for most anything you connect to it. my guess is that at 2200 mah the battery is under sized. i dont know if a firmware tweak will help. it has a universal firmware that was developed in March of 2018. i am planning on giving 1 more battery run (that will be #5 since purchase) before it goes back.

i purchased it based on the claimed run time and have spent too much time without even getting much past 50% of the advertised battery life
20200224_042808212_iOS.jpg
 
Feb 24, 2020 at 12:50 PM Post #28 of 768
I just realized in the dac mode on the Nx4, the battery life is not as great as the amp only mode. The Hip dac does not do analog input. So for some reason usb dac mode uses up too much power in everything.

Maybe someone should develop a sleep mode or power conserve mode on the USB. I just bought a battery checker/tester, and it does not use any batteries. So the device turns on immediately when sensing the battery plugged-in off of the test battery's remaining power. The usb should be able to do something similar. Turn off until there is a data signal and then quickly use the data as needed.
 
Last edited:
Feb 26, 2020 at 4:41 PM Post #29 of 768
There is some spilling of the secret sauce of iFi Bur Brown DAC over on the other Hip Dac thread. And my guesstimate is the low frequency is handled by multibit and high frequency is handled by sigma delta beta phi yadi yada delta modulation. AKM and Sabre both use delta method, and they sound very detailed supposedly; and the Audio GD setup uses multi-bit R-R DAC which supposedly sounds smooth. Smooth probably means slow like tubes, just a guess, but sounds good if you're into that kind of thing.

But sonic-ally how does this translate to good sound? I suspect DAC adds delay in the signal chain, even though small. And the multi-bit would be fast enough for low frequency and the sigma-delta is fast enough for high frequency. But I suspect there is adding involved and that could take a long time depending on who is doing it. So I wonder if the XMOS uses the Sigma-Delta to adjust for correction using prediction. Just guessing out loud. I guess who cares if it sounds good.

And I think I know why bluetooth and analog inputs are not supported. It's because the Hip Dac was designed by smart system engineers who otimized the dac-amp for the input PCM and DSD, I think. And the other two inputs would mess up the system design.
 
Last edited:
Feb 26, 2020 at 7:33 PM Post #30 of 768
It is my impression that advanced segment doesn't split it into "bass/hifreq" it shows the upper 6 bits (MSB/positive or negative + 5 bits of a 24 bit signal) determine the segment so the largest chunk of voltage is generated by the ladder, in "big" steps, then delta sigma adds in the other 18 bits (+ MSB for postive or negative) worth of voltage/current for that sample.

It's not split up into high vs low frequency data, it's just data points on the wave form. Low frequency wave form just rises and falls slower, it's not breaking it up by frequency and there's no reason your high frequency signal couldn't be just as much V in any given instant than the low freq hence the R2R part isn't working on a specific section of the wave form it's more like the ladder section is providing a rough estimate of the waveform and the delta sigma adds in the finer resolution between those large steps...
 

Users who are viewing this thread

Back
Top