• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Some measurements of iFi iDSD Pro DAC with comparisons

Veri

Master Contributor
Joined
Feb 6, 2018
Messages
9,596
Likes
12,036
So back when I had an iFi black label which was a really nice, powerful little handheld device, I always wondered about their top-tier iFi iDSD 'Pro', especially considering the iFi black label measured pretty decently (good to very good), with its highlight a very powerful amp.

On Jan 3, stereophile released a review of the iDSD Pro with several very interesting measurements of the device. Suffice to say, if the iFi iDSD was going to be some 'giant killer' upgrade of the iFi black... it received a failing grade especially considering its price. Let's look at and compare some of these graphs to get a better point of view into its performance.

First lets look at the jitter, with comparison to Jotunheim/iDSD micro

119iDSDfig22.jpg

rJOuUbs.png


Considering the jitter artifacts/peaks of the iDSD Pro top out at -130 dB, performance seems more similar to iFi Black than "better". Hardly a definite improvement.

Next up iFi Pro iDSD, spectrum with noise and spuriae of dithered 1kHz tone at –90dBFS
119iDSDfig12.jpg


Compared to Chord Qutest:
1218ChoQutfig09.jpg


Compared to Holo Spring ladder DAC via oversampling:
518HoloSpringfig09.jpg


iFi Pro iDSD, GTO filter, waveform of undithered 1kHz sinewave at –90.31dBFS:
119iDSDfig14.jpg


Compared to Qutest:
1218ChoQutfig11.jpg


Compared to Holo Spring ladder DAC via oversampling:
518HoloSpringfig11.jpg


Finally, iFi Pro iDSD, balanced Solid-State output, spectrum of 50Hz sinewave, DC–1kHz, at 0dBFS and –3dBFS:
119iDSDfig15.jpg

119iDSDfig16.jpg


Compared to Chord Qutest:
1218ChoQutfig12.jpg


Compared to Holo Spring ladder DAC via oversampling:
518HoloSpringfig12.jpg


Needless to say the iFi iDSD Pro here is the least impressive of the bunch. I think one of my first PMs to @amirm was about the possibility of buying and sending him an iDSD Pro. I never followed up with the idea, but I'm kinda glad I didn't. Given the exotic filtering, DSD upsampling and Tube modes of the iDSD Pro, these fancy things do not result in any measured performance benefits. All talk and marketing, little bite. The Qutest which is in fact cheaper, runs circles around it, and the Holo Spring which is a multi-bit ladder DAC seemingly performs better, while also offering DSD upsampling and 'BitPerfect'/NOS modes which are similar to iFi's offering.

If you love iFi audio's "house sound" and sound processing/filters, the iFi micro black label still seems the most attractive option to me, whilst not breaking the bank. If you want to look at more measurements check out Stereophile, many thanks to John Atkinson who makes easy comparisons possible via his (mostly always available) measurements tab in his review articles. Great work :)
 
Last edited:

confucius_zero

Addicted to Fun and Learning
Joined
Nov 22, 2018
Messages
541
Likes
345
Oh my lord?! Does it stand a chance next to JDS EL Dac or rme adi2?
 
OP
Veri

Veri

Master Contributor
Joined
Feb 6, 2018
Messages
9,596
Likes
12,036
Oh my lord?! Does it stand a chance next to JDS EL Dac or rme adi2?
Stereophile's graphs aren't a 1:1 comparison to Amir's due to analyser set-up and settings used so hard to say. The qutest is measured here at ASR though and is double-proven to be great. How deeply 'transparent' does one need a device to be, though ;)
 

despoiler

Member
Joined
Jan 8, 2019
Messages
45
Likes
50
For anyone that didn't see the announcement. iFi replaced their minimum phase filter with the GTO filter on all of their lower end products with the latest MQA compatible firmware. It's one less difference between pro and not.
 
OP
Veri

Veri

Master Contributor
Joined
Feb 6, 2018
Messages
9,596
Likes
12,036
For anyone that didn't see the announcement. iFi replaced their minimum phase filter with the GTO filter on all of their lower end products with the latest MQA compatible firmware. It's one less difference between pro and not.
Did you see the bit about GTO in the Stereophile review? Who in his right mind designed that filter?

The Gibbs Transient Optimized filter (fig.7) offered a slow rolloff with just 20dB attenuation of the image at 25kHz, and a large number of other aliased images are present
119iDSDfig07.jpg


Holy artifacts hell. That looks worse than NOS in some aspects. Is it optimised for the most amount of images possible??

EDIT: explained down below that the tone here is clipped explaining the odd measurement.
 
Last edited:

despoiler

Member
Joined
Jan 8, 2019
Messages
45
Likes
50
Did you see the bit about GTO in the Stereophile review? Who in his right mind designed that filter??

The Gibbs Transient Optimized filter (fig.7) offered a slow rolloff with just 20dB attenuation of the image at 25kHz, and a large number of other aliased images are present
119iDSDfig07.jpg


Holy artifacts hell. That looks worse than NOS in some aspects. Is it optimised for the most amount of images possible??

I had not seen that, but you might be interested in the white paper iFi posted on the filter.

https://ifi-audio.com/wp-content/.../08/iFi-audio-Tech-Note-GTO-filter-FINAL.pdf

For what it's worth I think the GTO filter actually sounds the best out of all of them. It's kind of weird too, because filters are usually extremely subtle and differences between can only be teased out by A/B testing or measurement. The GTO immediately stuck me as sounding really good. It could be general improvements in the overall firmware or placebo. It's odd that I found it compelling given what I objectively know.
 
OP
Veri

Veri

Master Contributor
Joined
Feb 6, 2018
Messages
9,596
Likes
12,036
I had not seen that, but you might be interested in the white paper iFi posted on the filter.

https://ifi-audio.com/wp-content/.../08/iFi-audio-Tech-Note-GTO-filter-FINAL.pdf

For what it's worth I think the GTO filter actually sounds the best out of all of them. It's kind of weird too, because filters are usually extremely subtle and differences between can only be teased out by A/B testing or measurement. The GTO immediately stuck me as sounding really good. It could be general improvements in the overall firmware or placebo. It's odd that I found it compelling given what I objectively know.
The measurement above proves there really would be a difference, theoretically a bad one, but you're not the first to find aliasing artefacts pleasing to the ear. I had personally upgraded my iFi micro to support the GTO filter but could not really hear any real-world difference with it and the minimum phase filter it replaced (I downgraded the firmware to go back to minimum phase). At least not with the garbage music I listen to :D ..
 

bennetng

Major Contributor
Joined
Nov 15, 2017
Messages
1,634
Likes
1,692
Looks like the 19.1k tone is clipped since it is at 0dBFS.
https://www.stereophile.com/content...da-processor–headphone-amplifier-measurements
1016MyBrookfig06.jpg


https://www.audiosciencereview.com/...rements-of-mytek-brooklyn-dac.1828/post-46077
index.php


JA measured another Mytek using 19.1k tone at -3dBFS and the problem is gone.
https://www.stereophile.com/content/mytek-liberty-da-processor-measurements
1018MyLibfig02.jpg


But for unknown reasons the impulse is inverted. Looks like the MQA processing is messing with audio level and polarity.
1018MyLibfig01.jpg
 
OP
Veri

Veri

Master Contributor
Joined
Feb 6, 2018
Messages
9,596
Likes
12,036
Looks like the 19.1k tone is clipped since it is at 0dBFS.
https://www.stereophile.com/content/mytek-hifi-brooklyn-da-processorheadphone-amplifier-measurements
1016MyBrookfig06.jpg


https://www.audiosciencereview.com/...rements-of-mytek-brooklyn-dac.1828/post-46077
index.php


JA measured another Mytek using 19.1k tone at -3dBFS and the problem is gone.
https://www.stereophile.com/content/mytek-liberty-da-processor-measurements
1018MyLibfig02.jpg


But for unknown reasons the impulse is inverted. Looks like the MQA processing is messing with audio level and polarity.
1018MyLibfig01.jpg
Interesting find thanks for chiming in. Could be that the GTO performs like a proper (hybrid?) minimum phase filter in real-world scenario. Would explain why I could discern zero difference before and after updating firmware! I'm not mad after all..! :)

But for unknown reasons the impulse is inverted. Looks like the MQA processing is messing with audio level and polarity.
Ouch. This is why I don't want the MQA filter forced down my throat in future devices.
 

firedog

Senior Member
Joined
Jun 19, 2016
Messages
378
Likes
642
Interesting find thanks for chiming in. Could be that the GTO performs like a proper (hybrid?) minimum phase filter in real-world scenario. Would explain why I could discern zero difference before and after updating firmware! I'm not mad after all..! :)


Ouch. This is why I don't want the MQA filter forced down my throat in future devices.
One of the good points of this DAC is that it uses the MQA filter only when playing back mqa files. Most other MQA dacs don't turn off the MQA filters once it is engaged.
 

ahofer

Major Contributor
Forum Donor
Joined
Jun 3, 2019
Messages
4,952
Likes
8,698
Location
New York City
I have one of these and it is a disaster. Every month or two it refuses to reboot and I have to re-flash the firmware. The wifi is buggy, and my Sennheisers sound bad with it (power? impedance?). Garbage.
 
OP
Veri

Veri

Master Contributor
Joined
Feb 6, 2018
Messages
9,596
Likes
12,036
I have one of these and it is a disaster. Every month or two it refuses to reboot and I have to re-flash the firmware. The wifi is buggy, and my Sennheisers sound bad with it (power? impedance?). Garbage.
:( well that's disappointing.
 

ahofer

Major Contributor
Forum Donor
Joined
Jun 3, 2019
Messages
4,952
Likes
8,698
Location
New York City
:( well that's disappointing.
Yes. It just did it again this morning. And the firmware app can't access the DAC to re-flash it because it...says it needs to reboot. The software aspect of this thing is a hot mess.
 

Haze

Member
Joined
Jun 28, 2018
Messages
93
Likes
34
Inverted or not that lack of pre-ringing is rather impressive.
 
Top Bottom