CHANGSTAR: Audiophile Headphone Reviews and Early 90s Style BBS

Lobby => Amp and DAC Measurements => Topic started by: Hands on May 22, 2014, 10:24:43 AM

Title: Audio-GD NFB-3.2 DAC Measurements
Post by: Hands on May 22, 2014, 10:24:43 AM

(http://i.imgur.com/oDZM0DL.jpg)

Test Hardware and Setup

-Audio-GD NFB-3.2 - SPDIF coax input from JKSPDIF.

-DIY Gaming Desktop - i5 2500K locked at 4.5GHz, Biostar TZ77XE3 motherboard, GTX 780, Antec EarthWatts 650W Green PSU, Auzentech Bravura sound card, 4 140mm fans, goofy UV cathode lights in case, Fractal Design Define R4 case...overall, most likely not optimal for audio, but does well enough as an output in tests and represents a real-life scenario for many. Plugged into APC H10 power conditioner. I did run Fidelizer on it for the tests. Used USB 3.0 ports from the mobo, tested for which ones produce best results and use those. Generally does not cause issues.

-Creative X-FI HD USB (SB1240) - Takes line-out signal from DAC to line-in for recording purposes, powered by USB bus, plugged into Surface Pro. Unit is limited to 48/96KHz input/output.

-1st gen Microsoft Surface Pro - Records from SB1240 via USB. Running Fidelizer tweaks. Battery powered. Screen is touchable and thus makes me a wizard. Produces cleaner and generally better results than running the SB1240 from the desktop as well.

Software

RightMark RMAA 6.3.0, ARTA

Misc. Notes

Creative SB1240 has a slight roll-off in the bass.

I do not guarantee 100% accuracy with these measurements. Good DACs can fairly easily max out the capabilities of my SB1240.

Links

Creative SB1240 - http://us.creative.com/p/sound-blaster/sound-blaster-digital-music-premium-hd (http://us.creative.com/p/sound-blaster/sound-blaster-digital-music-premium-hd)
RightMark Audio Analyzer - http://audio.rightmark.org/index_new.shtml (http://audio.rightmark.org/index_new.shtml)
ARTA: http://www.artalabs.hr/ (http://www.artalabs.hr/)

Can't find an exact page for this DAC, but you can find similar products or tidbits of info randomly on the intertubes.
Title: NFB-3.2 DAC Jitter Measurements
Post by: Hands on May 22, 2014, 10:42:11 AM
24/48 JTest using 4 of the 9 different filters, both "zoomed-in" and full spectrum view. Some filters seem to perform better than others here, but none of them look too bad.
Title: NFB-3.2 1KHz Square Wave Measurements
Post by: Hands on May 22, 2014, 11:05:37 AM
I picked up a cheap, pocket oscilloscope with digital storage, and it generally does a better job with these sort of measurements than the SB1240. The NFB-3.2, and many other AGD Wolfson DACs, have 9 digital filters to pick from. Three 2x, three 4x, and three 8x OS filters, and each of the three groups has one minimum-phase filter. The 2x and 4x MP filters are soft-knee. You might recognize some of these as the same types implemented in the Gamma2.

I had a difficult time getting consistent, accurate results for the 2x filters. They tended to be difficult to lock on with the oscilloscope and, even then, the waves themselves would slowly oscillate between having almost no ringing and a lot of ringing. I have similar troubles getting square wave measurements from my NOS DACs. The 4x and 8x filters gave me no issues. They were consistent and, from what I can tell, accurate.

That said, I might have somewhat accurately captured the square wave from filter 1, the 2x MP soft-knee filter. Maybe not. I will post it either way.

1. F1 - 2x MP soft-knee
2. F4 - 4x MP soft-knee
3. F5 - 4x LP soft-knee
4. F6 - 4x LP brickwall
5. F7 - 8x MP apodising
6. F8 - 8x LP apodising
7. F9 - 8x LP half-band
Title: NFB-3.2 -90.31dB 1KHz Sine Wave Test
Post by: Hands on May 22, 2014, 11:14:10 AM
Here are results of the -90.31dB 1KHz sine wave test. While there are some noticeable differences between this collection of filters at 16/48, the differences were much less prevalent in 24/96. Therefore, I'm only posting one 24/96 test result. Overall, the DAC does a pretty decent job here. Filters tested were 1, 4, 5, and 9, and the 24/96 result is from F1.
Title: NFB-3.2 13/14KHz Test (-3dB)
Post by: Hands on May 22, 2014, 11:20:56 AM
13KHz and 14KHz test (ignore numerical HD readout). I did these just at -3dB (no 0dB tests), both to save myself some time, save space on the thread, because you probably won't listen to recordings with these at 0dB, and because the 0dB tests I tried were looking kinda funky on some filters.

Filters 1, 4, 5, and 9.
Title: NFB-3.2 19KHz Measurements
Post by: Hands on May 22, 2014, 11:22:58 AM
Similar to above, but a single 19KHz sine wave measurement at -3dB. Filters 1, 4, 5, and 9.
Title: NFB-3.2 Filter 1 - 16/48 RightMark Measurements
Post by: Hands on May 22, 2014, 11:31:01 AM
Filter 1 - 16/48 RightMark Measurements. Frequency response measures similar to how most NOS DACs measure. Noise floor is very low for 16-bit, and dynamic range is high. Raised 2nd and 3rd order harmonic distortion, but nothing too bad. Everything beyond that is well controlled. The rest of the 2x filters measure similarly, but I can post them on request. Same for 24/96 tests...until then, I'll stick to one 16/48 measurement each from 2x, 4x, and 8x filters.

Frequency response (from 40 Hz to 15 kHz), dB   +0.12, -1.05Average
Noise level, dB (A)   -97.8Excellent
Dynamic range, dB (A)   97.9Excellent
THD, %   0.0080Very Good
THD + Noise, dB (A)   -79.3Average
IMD + Noise, %   0.0090Very Good
Stereo crosstalk, dB   -97.2Excellent
IMD at 10 kHz, %   0.011Very Good
General performance    Very Good
Title: NFB-3.2 Filter 4 - 16/48 RightMark Measurements
Post by: Hands on May 22, 2014, 11:40:10 AM
16/48 RightMark measurements with Filter 4 (4x, MP, soft-knee).

Frequency response (from 40 Hz to 15 kHz), dB   +0.01, -1.63Average
Noise level, dB (A)   -97.8Excellent
Dynamic range, dB (A)   97.8Excellent
THD, %   0.012Good
THD + Noise, dB (A)   -76.1Average
IMD + Noise, %   0.011Very Good
Stereo crosstalk, dB   -96.9Excellent
IMD at 10 kHz, %   0.012Very Good
General performance    Very Good
Title: NFB-3.2 Filter 9 - 16/48 RightMark Measurements
Post by: Hands on May 22, 2014, 11:47:46 AM
16/48 RightMark measurements with Filter 9, 8x LP half-band.

Frequency response (from 40 Hz to 15 kHz), dB   +0.01, -0.07Excellent
Noise level, dB (A)   -97.5Excellent
Dynamic range, dB (A)   97.6Excellent
THD, %   0.0079Very Good
THD + Noise, dB (A)   -79.3Average
IMD + Noise, %   0.011Very Good
Stereo crosstalk, dB   -97.0Excellent
IMD at 10 kHz, %   0.012Very Good
General performance    Excellent
Title: Re: Audio-GD NFB-3.2 DAC Measurements
Post by: funkmeister on May 22, 2014, 12:25:31 PM
Is it me or are those IMD numbers a bit higher than we picky folk like?!

thx for the measurements.
Title: Re: Audio-GD NFB-3.2 DAC Measurements
Post by: Hands on May 22, 2014, 12:54:53 PM
Yes, THD and IMD measure a touch higher than the "picky folk" would like. :P I'd be curious to see if this would change on better measuring gear or not, but I wasn't entirely surprised given the NFB2 measurements on here. That, and the AGD site has warnings all over it that non-feedback doesn't always measure well. Not sure if that's applicable here or not. If the newer AGD Wolfson DACs are further improved on this, that would be exciting.

IMD is actually decently clean overall. THD "cleanliness" depends on the filter used. At least it's generally just 2nd and 3rd order stuff that is a bit high. Subjectively, though, I think this DAC sounds pretty darn good despite the slightly high numbers in some areas. It also just does really well in some areas. For the $150 I paid, it's probably one of the best audio investments I've made in a long while!

Edit: Touching on how newer AGD Wolfson DACs might measure, the NFB-10.33 is specified to have <0.001% THD with 1KHz sine at 0dB. If that's the case, and IMD follows, I would be very curious to check out one of the newer DACs. Though the NFB-10.33 is a DAC/amp combo unit, if that somehow makes a difference.
Title: Re: Audio-GD NFB-3.2 DAC Measurements
Post by: OJneg on May 22, 2014, 12:59:39 PM
Impressions?
Title: Re: Audio-GD NFB-3.2 DAC Measurements
Post by: Hands on May 22, 2014, 01:04:32 PM
Quote from: OJneg on May 22, 2014, 12:59:39 PM
Impressions?

I have some notes...just need to get around to posting them in my ongoing subjective thread.
Title: Re: Audio-GD NFB-3.2 DAC Measurements
Post by: firev1 on May 22, 2014, 01:17:13 PM
You should have switched on Two Sine mode so we could get DFD2 for the CCIF 13khz 14khz twin tone test. Measurements are completely expected of the DAC that is within the same family as NFB2.0 that I have(same even).

Jitter rejection seems tons better on the newer models though the main difference is that my interface/method of measurement highly influences the jitter readings(JKSPDIF vs SB1240 going full duplex, no guesses for which one is sh1t.) Also better measurements on the 3.2 rather than my old 2.0, not a fair comparion(it should go toe to toe with the 2.32.)

Really interesting to see how the filters work with HF measurements, too bad mine can't be changed :(
Title: Re: Audio-GD NFB-3.2 DAC Measurements
Post by: Hands on May 22, 2014, 01:43:21 PM
I tried switching it to two sine mode, but with how I set it up, it was giving me fits. I probably could have gotten it working if I played around with it more, but sometimes I just don't have the motivation, haha.
Title: Re: Audio-GD NFB-3.2 DAC Measurements
Post by: firev1 on May 23, 2014, 01:18:52 PM
LOL get what you mean, I spent so much time measuring and not listening at times I found myself ridiculous, which was why when I borrowed the Zodiac from a friend, I did not measure it at all. All the time was spent listening.