NAD C 298 Power Amplifier With Purifi Eigentakt Amplification

[Alexanderc]

Edit: I posted this before I saw @Matias had already replied. It appears this is an apples to oranges comparison?

 

[Matias]

It appears this is an apples to oranges comparison?

I think so. If using ASR's 22kHz bandwidth, the high out of band noise does not impact the THD+N, but SoundStage! uses 90kHz which does impact the THD+N....

 

[pozz]

You are right: SoundStage! uses 90 kHz bandwidth, while ASR uses 22 kHz.
@pozz I think we have another problem....

The settings, gear and setup change so much that it's difficult to compare any two sources.

 

[Matias]

The settings, gear and setup change so much that it's difficult to compare any two sources.

I think I need to remove all class D entries (higher out of band noise) in the table coming from SoundStage! measurements since they started measuring in-house, as I think before that Bascom H. King used 22kHz bandwidth too.

Bummer. C 298 removed from the list.

 

[pozz]

I think I need to remove all class D entries (higher out of band noise) in the table coming from SoundStage! measurements since they started measuring in-house, as I think before that Bascom H. King used 22kHz bandwidth too.

Bummer. C 298 removed from the list.

No harm in keeping them in a separate tab. They can still be useful then, just not for comparison.

 

[Joaquinín]

But more than 20 db difference? It seems odd. Notice the "THD+N ratio (unweighted) vs. output power at 4/8 ohms " table at Soundstage doesn´t say a word about the frequency, or does it? If it is measuring just 1 khz, then the result that it shows at 8 ohms/10w (82db!) is in contradiction with FFT spectrum – 1kHz measurement (taken at 8 ohm/10w). Am I wrong?

 

[Alexanderc]

If they're measuring 90khz bandwidth, then would that rise in noise above 20k and the spike at c. 80k all be included in the THD+N graph?

Edit: I'm referring to the FFT spectrum graph that @Joaquinín mentioned.

 

[ebslo]

If they're measuring 90khz bandwidth, then would that rise in noise above 20k and the spike at c. 80k all be included in the THD+N graph?

Edit: I'm referring to the FFT spectrum graph that @Joaquinín mentioned.

Yes, I believe so.

FTA, just above the second table:

Our primary measurements revealed the following using the balanced line-level inputs (unless specified, assume a 1kHz sinewave, 10W output, 8-ohm loading, 10Hz to 90kHz bandwidth):

The "Noise level (unweighted)" from the table agrees with the THD+N plot at 10W, so it appears the 90kHz bandwidth applies to both measurements.

 

[Matias]

It is written there:
"With the exception of frequency response, where the Audio Precision bandwidth is set at its maximum (DC to 1MHz), assume, unless otherwise stated, a measurement input bandwidth of 10Hz to 90kHz, and the worst-case measured result between the left and right channels."

But more than 20 db difference? It seems odd. Notice the "THD+N ratio (unweighted) vs. output power at 4/8 ohms " table at Soundstage doesn´t say a word about the frequency, or does it? If it is measuring just 1 khz, then the result that it shows at 8 ohms/10w (82db!) is in contradiction with FFT spectrum – 1kHz measurement (taken at 8 ohm/10w). Am I wrong?

Yes, noise levels from class D skyrocket after 20 kHz. This is done on purpose, minimize noise in band, and sweep the dirt under the carpet- I mean, shift the noise out of band.

 

[ebslo]

For a good indication of where this amp stands with regards to SINAD by ASR method, compare the "FFT spectrum – 1kHz" chart in this Soundstage! measurement with the corresponding chart in the ASR review of the NAD M28. Highest harmonic peaks looks about 7dB lower than M28, and noise floor from 500Hz to 20kHz, accounting for the 3dB difference in signal power level, also looks several dB lower than M28. Noise below 200Hz seems roughly similar to M28, nowhere in the 20Hz to 20kHz range does the noise floor look higher than M28.

ASR rated the M28 SINAD at 98dB. Based on that and the Sounstage FFT, I would expect this unit would rate a couple dB better than M28 by ASR method.

 

[mbsaph21]

I'm about to take delivery of a C298. Has anyone who's listened to the C298 had a similar experience to that of Evan McCosham at SoundStage Network regarding a slight lack of bass weight?

 

[nothingman]

Here’s a tidbit from the beginning paragraph about A-weighted THD+N, since we don’t have that in the graph: “THD+N (A-weighted) at 10W into 8 ohms measured approximately 0.0004%, while variable mode yielded 0.0006%.”

If my online conversion calculator skills are correct, that is -108db and -104db, respectively.

 

[Matias]

Here’s a tidbit from the beginning paragraph about A-weighted THD+N, since we don’t have that in the graph: “THD+N (A-weighted) at 10W into 8 ohms measured approximately 0.0004%, while variable mode yielded 0.0006%.”

If my online conversion calculator skills are correct, that is -108db and -104db, respectively.

The A-weighting messes up the numbers though....

 

[nothingman]

I'm about to take delivery of a C298. Has anyone who's listened to the C298 had a similar experience to that of Evan McCosham at SoundStage Network regarding a slight lack of bass weight?

I’m not the right guy to answer since I’m running mine into a pair of LS50 Metas (my answer is no, however), but personally I take it with a grain of salt when a reviewer goes well out of their way to qualify with something like “slightly less,” as he does in reference to the difference in bass amount (not even in qualify or speed) between the C298 and Bryston, or “exceedingly small,” as he does in reference to the differences in “textural and tonal” details between the C298 and Purifi demo. Beyond any concerns over one guy doing a bunch of sighted listening, at that point for your average consumer who will never do an A-B comparison, much less ABX, it’s probably splitting hairs.

FWIW I like the review and think Evan McCosham and SS! did a nice job. That new measurement format, whatever its apparent shortcomings in translating to ASR (and Stereophile, I presume?) standards, are as thorough as anything I’ve ever seen.

The A-weighting messes up the numbers though....

Okie dokie. I really shouldn’t wade into these waters since my understanding is extremely layman.

 

[mbsaph21]

I’m not the right guy to answer since I’m running mine into a pair of LS50 Metas (my answer is no, however), but personally I take it with a grain of salt when a reviewer goes well out of their way to qualify with something like “slightly less,” as he does in reference to the difference in bass amount (not even in qualify or speed) between the C298 and Bryston, or “exceedingly small,” as he does in reference to the differences in “textural and tonal” details between the C298 and Purifi demo. Beyond any concerns over one guy doing a bunch of sighted listening, at that point for your average consumer who will never do an A-B comparison, much less ABX, it’s probably splitting hairs.

FWIW I like the review and think Evan McCosham and SS! did a nice job. That new measurement format, whatever its apparent shortcomings in translating to ASR (and Stereophile, I presume?) standards, are as thorough as anything I’ve ever seen.



Okie dokie. I really shouldn’t wade into these waters since my understanding is extremely layman.

Thanks SO MUCH for the insights, @nothingman - your feedback is what I was kind of thinking myself, even before listening. Given the specs/frequency response of the amp, I also find it slightly implausible that it is "lighter" on bass as compared to the rest of the frequency range. I'll be running mine through LS50 (originals), btw!

 

[nothingman]

Thanks! I didn’t expect that to receive an all caps level response but I’ll take it. Hope you enjoy your new amp and LS50s!

 

[Moto]

I see in the beginning of the review, “The Eigentakt amplifiers themselves provide 13dB of gain, so it appears the variable gain circuitry simply acts as an input attenuator.”
Does that seem correct? If so then it seems best to just use fixed gain(max) and attenuate the balanced input from my SMSL SU9 using the 64bit dsp volume control on roon? Correct or???

 

[Kal Rubinson]

I see in the beginning of the review, “The Eigentakt amplifiers themselves provide 13dB of gain, so it appears the variable gain circuitry simply acts as an input attenuator.”
Does that seem correct? If so then it seems best to just use fixed gain(max) and attenuate the balanced input from my SMSL SU9 using the 64bit dsp volume control on roon? Correct or???

While the module itself has 12.8dB of gain, the C298 has fixed/variable gain settings from 8.5dB-28.5dB in stereo mode and from 14.5dB-34.5dB in bridged mode. So, there is some option for attenuation but most users will use the added can of the input buffer.

 

[Moto]

@Kal Rubinson so from 12.8 down to 8.5 is pure attenuation and from 12.8 up to 28.5 is gain from the input buffer Oram I misunderstanding?
Or is it all just pure attenuation from the combined buffer plus purifi total 28.5db gain?

 

[Kal Rubinson]

@Kal Rubinson so from 12.8 down to 8.5 is pure attenuation and from 12.8 up to 28.5 is gain from the input buffer Oram I misunderstanding?
Or is it all just pure attenuation from the combined buffer plus purifi total 28.5db gain?

I think you are conflating reducing gain with attenuation. The former is changing the values in a circuit so that it amplifies less while the latter directly reduces the signal. If manufacturer's are being honest when referring to "variable gain," it is different from attenuation, say with a resistor or potentiometer.

The distinction is important because they affect system noise differently.