Which filter setting should I use on Hugo 2, TT2 or Qutest, and what does the HF filter do?

The four settings

On Hugo 2, TT2 and Qutest the filter button controls two things: whether the WTA2 filter (16FS to 256FS) is in circuit, and whether the HF filter is on. His mapping, given in March 2022: "Fil1 = white HF off WTA2 on, Fil2 = green HF on WTA2 on, Fil3 = orange HF off WTA2 off, Fil4 = red HF fil on WTA2 off" Head-Fi, Mar 2022. In 2019 he gave the same numbering by color: "Filter 1 (TT2) is white (H2)Filter 2 is greenFilter 3 is orangeand Filter 4 is red" Head-Fi, Jan 2019. He has said the HF filter for TT2, Hugo 2 and Qutest is done in the same place on all three, in the filters after WTA2 Head-Fi, Jun 2022, and that the incisive (WTA2) filter is on Hugo 2, TT2 and Qutest, while on DAVE "it's always on" Head-Fi, Jul 2018. DAVE has an HF filter Head-Fi, Dec 2015 but "no option to lose WTA2" Head-Fi, Jan 2019.

Which setting he recommends

Hugo 2, from launch (March 2017): "Normally you should be using white or green - I run with green all the time as its useful with 192 recordings" Head-Fi, Mar 2017. He repeated that he runs green in July 2017 ("For 99.9% of the time I just run with green") and August 2017 Head-Fi, Jul 2017Head-Fi, Aug 2017, and in 2018 and 2019 said he leaves it on green because it helps with his 96k or other HD recordings Head-Fi, Apr 2018Head-Fi, Oct 2018; the January 2019 statement "I never use orange/red" came in a post about his TT2 Head-Fi, Jan 2019.

In August 2021, correcting the Hugo 2 manual, he wrote "I recommend filter 2 for both DSD and all PCM, but the benefits at 44.1k is smaller" Head-Fi, Aug 2021. In March 2022 on the TT2 thread he wrote "Technically, the best sound is filter 2 or green" and said green is his preferred setting, while "using the other settings is OK if you need it - but you are using distortions" Head-Fi, Mar 2022. At CanJam SoCal in September 2022, discussing the TT2, DAVE and Hugo 2 filters, he said that if a system is too sharp "you want to tame that down" The Source AV 40:47; when the host asked whether the best technical basis is filter two or green, he repeated the host's words: "It's filter two or the green. And that's what you need to be moving towards." The Source AV 41:02. His most recent statement in this archive, February 2025, concerns TT2 with the Hugo M Scaler: "For TT2 use Fil1 or 2 (I use 2) with the Hugo M scaler but not 3 and 4." Head-Fi, Feb 2025.

For Qutest specifically, in March 2019 he told a listener who preferred orange over white that this was "the equivalent of soft focus for images", suggested the system was too bright, and said the owner "may" get better results staying on white, possibly with DSP EQ at the source, though source EQ costs transparency because the data is no longer bit perfect; in the long term he said the bright component should be replaced (or, with loudspeakers, repositioned) so the system is optimized "around white or green filters" Head-Fi, Mar 2019.

For DSD, in March 2017 he said he leaves Hugo 2 on green "as DSD is a tiny percentage of the music I enjoy listening too", while noting DSD already tends to sound warm and soft Head-Fi, Mar 2017. His 2021 recommendation of filter 2 covers DSD explicitly Head-Fi, Aug 2021.

What the HF filter does

Purpose. He says it exists for high-sample-rate recordings: "its there for HD recordings, particularly 192 or even more with 384 or 768k" Head-Fi, Dec 2017. ADCs put "large amounts of noise shaper noise, starting at about 25 khz" into such recordings, and because that noise is random it "creates noise floor modulation in the analogue electronics", so filtering it out "makes it sound smoother" Head-Fi, Jan 2018. He describes the result as warmer and smoother with better instrument separation and focus, "exactly the SQ changes one expects to hear from reducing noise floor modulation" Head-Fi, Mar 2022. In 2018 he said energy above 30 kHz on HD recordings "is just distortion and noise from the ADC's and needs removing" Head-Fi, Oct 2018.

How big the change is. He has said many times that white to green is small at 44.1k and large at 96k and 192k Head-Fi, Mar 2017Head-Fi, Mar 2017Head-Fi, Jun 2017Head-Fi, Jan 2018Head-Fi, Apr 2018. In June 2017 he said that with a 768 kHz test recording it "becomes essential" Head-Fi, Jun 2017. In 2017 he was surprised that white to green at 44.1k would be audible at a show at all Head-Fi, Mar 2017.

How it works. He has said it is not an extra stage switched into the signal path: "what actually happens is the time constants of the 2048FS filter is changed, not a signal path being switched" Head-Fi, Feb 2019. The same filter is used whether WTA2 is in or out (white/green or orange/red), so "WTA2 always stays the same" Head-Fi, Jun 2022Head-Fi, Aug 2022. For that reason, he said in 2019, it makes no difference what the original input was Head-Fi, Aug 2019. In January 2025, on the Qutest thread, he said white and green "merely adjusts the time constants of the filter that goes from 256FS to 2048FS", while red and orange switch WTA2 out Head-Fi, Jan 2025.

The figures he has given for that filter vary from post to post:

• July 2017: each low-pass filter time constant changes "from 0.7 uS (227 kHz) to 2.8 uS (57 kHz)" Head-Fi, Jul 2017.
• March 2017: "The HF filter is second order, and has a -3dB at 40 kHz" Head-Fi, Mar 2017.
• January 2019: the time constants are adjustable, "about 240 kHz or 60 kHz" Head-Fi, Jan 2019.
• February 2019 (Qutest thread): the -3dB point moves from 150 kHz to 37.5 kHz Head-Fi, Feb 2019.
• June 2019: the time constant "is changed from 160 kHz to 40 kHz" Head-Fi, Jun 2019.
• August 2022 (Qutest thread): "green/orange removes more HF noise and distortion above 50kHz" Head-Fi, Aug 2022. The "green/orange" pairing in that post does not match his mapping elsewhere, where the HF filter is on in green and red Head-Fi, Mar 2022.

Effect on frequency response. He says the HF filter is measurable in the audio band: "the HF filter (green or red) adds -1dB at 20kHz from ruler flat" (December 2017) Head-Fi, Dec 2017Head-Fi, Feb 2018, and in 2019 he added it is "0.24dB down at 10 kHz" Head-Fi, Jun 2019. In 2025 he described the pre-Mojo 2 HF filters as "a 2nd order simple filter" with "a significant drop at 20kHz of around -0.9dB" Head-Fi, Mar 2025. By contrast, switching WTA2 (white to orange) does not show on standard measurements: "You won't measure any change from the 16FS to 256FS WTA filter" Head-Fi, Dec 2017Head-Fi, Jan 2019Passion for Sound 3:38.

Why it is switchable at all. He has said the reason is that 20 kHz roll-off: "the technical reason we have the filter selection is that green is about -1dB down at 20 kHz which is not flat, and would be subject to some criticism if you couldn't defeat it" (March 2022) Head-Fi, Mar 2022. He made the same point at CanJam SoCal 2022 and in 2024 and 2025 The Source AV 38:53Head-Fi, Jun 2024Head-Fi, Mar 2025.

Why it sounds warmer even at 44.1k: how his explanation developed

• 2015 to 2016 (DAVE): he called the benefit on CD "curious" and "very unexpected", since WTA stop-band rejection is already 120 to 140 dB; he said he "very strongly" suspected noise and intended to find out for sure Head-Fi, Nov 2015Head-Fi, Dec 2015Head-Fi, Jan 2016.
• June 2019: he offered three possible explanations (the -1dB at 20k, better stop-band attenuation, and lower noise floor modulation from extra rejection above 12 MHz), said "My guess is issue 3 is the most likely explanation, but I am not too concerned about the issue as green with RBCD is a small difference", and that he would "in due course nail down the exact explanation" Head-Fi, Jun 2019.
• April and May 2020: he reported testing a new HF filter with 20 kHz compensation (a net +0.4dB at 20k), which still sounded warmer, including in a single-blind test with his son, "proving that noise floor modulation is the cause of the HF filters change in sound quality" Head-Fi, Apr 2020Head-Fi, May 2020.
• 2022 to 2025: he says the warmth is "not due to the fact that it's -1dB at 20kHz - nobody can hear that effect, nor is it it due to phase shifts", citing a flat-to-20kHz filter he designed for Mojo 2 that sounded the same Head-Fi, Mar 2022. In June 2024, on the Mojo 2 thread, he described switching the Mojo 2 HF filter in behind an FIR bandwidth-limiting filter and hearing "no change whatsoever", concluding that the benefit of bandwidth limiting comes "entirely from the elimination of HF components" Head-Fi, Jun 2024. He also cited, in February 2025, a test with one of his ADC decimation filters that he says showed the -1dB and phase shifts are inaudible "at least to my old ears" Head-Fi, Feb 2025.

Orange and red (WTA2 switched out)

He describes these as a departure from his usual practice of giving no options Head-Fi, Mar 2017Head-Fi, Jul 2017. He gives two reasons for including them: so people could hear what the WTA2 filter does, and as a way to soften the sound with bright recordings or hard headphones Head-Fi, Mar 2017Head-Fi, Jul 2017Head-Fi, Oct 2018. In March 2017 he wrote "for sure you are using an aberration to hide another problem", and advised that anyone using orange or red all the time "consider getting a warmer set of headphones, or trying out EQ" Head-Fi, Mar 2017. In April 2017 he said the 16FS option "is intended for too hard HP" Head-Fi, Apr 2017, and in July 2017 that it uses "an inaccuracy" to get the balance right Head-Fi, Jul 2017. In September 2023 he summarized: "two wrongs don't make a right", adding that Mojo 2's UHD DSP was the answer for making the sound warmer without degrading transient timing Head-Fi, Sep 2023.

He says the WTA2 change is about perceiving the starting and stopping of notes (piano attacks, wood blocks, rim shots), and that some listeners hear it clearly while others hear no change Head-Fi, Apr 2018Head-Fi, Apr 2018Head-Fi, Oct 2017. In 2017 he said "the real big step" is the change between white or green and orange or red, more than white to green Head-Fi, Mar 2017.

With the Hugo M Scaler

He has said the filter choice is unaffected by the M Scaler, because the M Scaler only replaces WTA1 and passes the recording's ADC noise through, so the HF filter is still needed Head-Fi, Sep 2018Head-Fi, Jan 2019Head-Fi, Jan 2019Head-Fi, Feb 2019Head-Fi, Apr 2020. In September 2018, on the M Scaler thread, he advised "Stick to green filter" Head-Fi, Sep 2018. In January 2019 he said the M Scaler's warmer sound makes WTA2 "even more attractive" Head-Fi, Jan 2019. His February 2025 advice for TT2 with the M Scaler was Fil1 or Fil2, not 3 or 4 Head-Fi, Feb 2025.

DAVE: a change over time

In January 2016 he said of DAVE's HF filter "I just keep it on all the time" Head-Fi, Jan 2016. From 2017 on he said that with the M Scaler and DAVE the HF filter is best off, because the M Scaler's stop-band rejection is better Head-Fi, Jan 2017. In 2022 and 2023 he said he runs DAVE via the M Scaler with the HF filter off, and that without the M Scaler he always had it on Head-Fi, Feb 2022Head-Fi, Jan 2023Head-Fi, Jan 2023. He tied this partly to his own system: "my office system with Dave is on the warm side, so I don't need more warmth from the HF filter" Head-Fi, Jan 2023. In 2022 he also wrote "Go with whichever works best for you" Head-Fi, Feb 2022. These statements concern DAVE with the M Scaler; for TT2 with the M Scaler his 2025 statement is that he uses Fil2 Head-Fi, Feb 2025.

Mojo 2, for comparison

He says Mojo 2's HF filter is permanently on and flat from 20Hz to 20kHz, because the UHD DSP allowed a sharp roll-off without the 20 kHz drop, and that with it on it "simply sounded much better in all circumstances" Head-Fi, Feb 2022Head-Fi, Jun 2024Head-Fi, Jun 2024MFG Audio Reviews 42:27.

What he has not said: this archive has no statement from him on which setting to use for a specific headphone model, and no statement that orange or red is technically better for any source; he has also not reconciled the different cutoff figures he has given for the HF filter in different posts Head-Fi, Jul 2017Head-Fi, Mar 2017Head-Fi, Feb 2019Head-Fi, Jun 2019.

Quotes

“Normally you should be using white or green - I run with green all the time as its useful with 192 recordings”2017-03-15 Head-Fi, Mar 2017

“Technically, the best sound is filter 2 or green”2022-03-21 Head-Fi, Mar 2022

“So the biggest benefit is with HD recordings, but I still hear a small change when using the HF filter with 16/44.1.”2022-03-21 Head-Fi, Mar 2022

“Turning it on gives a warmer and smoother sound, with better instrument separation and focus - exactly the SQ changes one expects to hear from reducing noise floor modulation”2022-03-21 Head-Fi, Mar 2022

Dated positions

Caveats and open points

• The cutoff figures Watts has given for the HF filter differ between posts (for example 57 kHz, 40 kHz -3dB, about 60 kHz, 37.5 kHz -3dB, 40 kHz); they are reported as stated, not reconciled.
• In his August 2022 Qutest post Head-Fi, Aug 2022 the text reads "green/orange removes more HF noise", which does not match his other mappings, where orange is HF off and red is HF on Head-Fi, Mar 2022; the quote is reported as written.
• His DAVE-with-M-Scaler practice (HF filter off) is tied to his own warm office system and to DAVE; it is not a statement about Hugo 2, TT2 or Qutest.
• Statements about audibility are Watts's own listening reports; he notes that some listeners hear little or no difference between settings.

Topics: User filter settings and the HF filter