• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

3 Generations of Marantz flagship integrated amps measured

GXAlan

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
1,596
Likes
2,425
Three Generations of Marantz flagship integrated amps measured (updated PM-10 test!)
1663797630062.png

One of the biggest challenges with amplifier testing is shipping the gear without damage. We've already seen how the JBL SA600 from 1965 pretty much has an amplifier section that performs similarly to the JBL SDR-35 with processing in 2022. But what about Marantz? Have they improved their technology over time?

On deck for today's review?

Marantz PM-90
1991 - all analog flagship, second only to the PM-95. Last Marantz product BEFORE the introduction of HDAMs. Exact model tested here. ~$3500 in 2022 dollars.

Marantz PM-11s2
2008 - all analog flagship. Integrated amp version of SM-11s1 (tested at Stereophile). Last D+M Flagship during the economic boom right before the Great Recession. HDAM SA3. ~$6500 in 2022 dollars.

Marantz PM-10
2017 - all analog, fully balanced flagship. Marantz's introduction of the HypeX NC500OEM running BTL with a SMPS600 per channel (so 4 NC500's + 2 SMPS600's in the box). The last D+M Flagship before Sound United acquisition. Last Ken Ishiwata flagship. $9000 in 2022 dollars (though BNIB "open box" deals are easily found at ~$5500).

1663797630062.png



Test setup
E1DA Cosmos ADC, Grade A, tested in single channel stereo mode for amplifiers
@pkane's Multitone software
Panasonic UB9000 as source, 1 kHz test tone
Vishay Dale 4.48 ohm 1% NH250 non-inductive resistor

Limitations of the testing:
1. Ideally, I should have a 4 ohm load instead of a 4.48 ohm load, but c'est la vie. The key is comparing between the three units. Resistors are expensive!
2. The UB9000 only outputs 2V from its balanced connections. It's the only balanced DAC I have right now.
3. I have shown that different interconnect cables can result in differences in SINAD even if the differences are inaudible. This can be as much as 3 dB in SINAD. I didn't spend any time focusing on cables.
4. I might get better results if I was testing both amplifier channels simultaneously. Wasn't sure if the 48V minijack benefits from the same summing.

UPDATE: I retested the PM-10 through the XLR inputs allowing for proper grounding.

1663797584061.png



First the UB9000 in E1DA Cosmos ADC mono-summed mode compared to Amir's testing. The E1DA Cosmos ADC does a very nice job of testing line level products.
1663797921240.png


First the PM-90
- My setup is about ~9 dB worse than Amir's APx555 setup and super fancy
- Note the clean noise spectrum above 20 kHz
- My AC power is just as noisy as Amir's. This wasn't been re-built and maybe an upgraded power supply would help
- This *sounds* wonderful. On my test, it's really clear how the majority of the harmonic distortion is the even, not odd distortion.


1663798226636.png


If I turn up the volume ~6 dB higher, the numbers improve slightly
1663798344516.png



Next up, the PM-11s2.

Looks like it'd be a golfing panther! At -94.1 dB SINAD, it's clearly going to be in the blue range if it were tested on the APx555! Would it break 100 dB? Who knows?
1663798397792.png


What if I turn up the volume a bit?

1663798533253.png


Are you not impressed? If there's still another 6 dB on the table with a better resistive load or even a better DAC feeding the amp, you're now looking at one of the very best measuring Class AB amplifiers, period. With all of the Marantz "Sound Master" tuning, convenience of a remote control, and well documented service manuals!

If the PM-11s2 is that good, what about the PM-10?

Well, this is where the story gets complicated. The PM-10 was built after the Great Recession and the rise of headphone hi-fi. Interviews with the design team talk about environmental regulations/certain parts no longer being available, and Marantz Japan making sure everyone knew the switch to Class D was Marantz Europe's request. The measurements clearly show the consequences for Marantz and switching to class D with noise above 20 kHz and more noise from the power supply.

Old Inaccurate Tests
1663799035822.png
1663799046824.png


Is the PM-10 a complete failure? Not quite. If we look a little deeper it becomes pretty clear that PM-10 does have lower THD, but it's the noise that hurts the PM-10 the most.

1663799282009.png


With an improvement in the power supply, the next generation of the PM-10 could be pretty spectacular.

UPDATE: With ASR membership guidance, I re-tested using the XLR inputs and attached the ground to the PM-10's grounding lug.

1663861179066.png


1663861186019.png


What about subjective comparisons?
These tests only measure 1 kHz output and real-world content may be different. The PM-10 essentially can output 600 watts into 8 ohms as compared to the ~120W or so of the PM-11s2. So, for more dynamic music, it's also a big difference to have the extra power on tap.

Subjectively, all three units sound great. I genuinely do think the PM-90 has subjective warmth with vocals and that may be the high 2nd order harmonics and it's a great deal at the used $1K price they go for. The PM-10 and PM-11s2 sound similar but I do believe that I get more subjective bass with the PM-10. That might be the power supply harmonics adding some extra energy in the bass region or improved damping factor. I also do think the PM-10 is better than the PM-11s2 at placing instruments in the space between phantom center and the left and right channel. Not sure if this is purely sighted bias, or if the cleaner signals above 1 khz make it like a cleaner version of the PM-11s2 with a truck idling down the street at -100 dB adding some lower frequency noise.


UPDATE: Whew. The PM-10 sounded better, and thankfully with proper grounding of the test gear, it also measures better.

Conclusion

Overall, I think Marantz has done a pretty good job with their flagship integrated amplifiers. There is clear progression in technology over the time.

The PM-10 improves over the PM-11s2 in the realm of harmonic distortion in the midrange/treble but slips on the power supply performance. It is more power efficient and a little bit of power supply noise for a lot more watts is a reasonable trade off. At MSRP and discounted pricing, you still have to be willing to vote-with-your-wallet and spend the extra money for Made in Japan and for a 5 year warranty with a solid service manual. When the NCx500OEM is used as a drop-in replacement, and a cleaner power supply is used, things could be very interesting.

UPDATE: With the revised testing, and proper grounding, we can see just how good the PM-10 actually is. It's better than the PM-11s2, now with the advantage of up to 600W on tap (even though it's conservatively rated at 200W into 8) and the more power efficient Class D architecture. I am much more comfortable stating that Marantz has shown clear progression/advancement with time.

My comments about pricing still apply. At MSRP or discounted pricing, it's still expensive. That said, you are getting a very good performance out of the unit -- remember that we're testing this as a full integrated amp with features like tone controls, multiple inputs, a phono unit. You still have to be willing to "vote-with-your-wallet" and spend the extra money for Made in Japan, the 5 year warranty you hope to never use, and the solid service manual hoping that it'll be usable when you pass it onto the next generation.

Still, the PM-11s2 is a real testament to what was achieved by Marantz as the peak of their Class AB lineup. It essentially provides identical performance to the PM-10 at lower power. In fact, when I get a chance, it would be interesting to see how the PM-11s2 performs when properly grounded. Cutting down that smaller ground loop is defintely going to improve the numbers a bit.
 
Last edited:

restorer-john

Grand Contributor
Forum Donor
Joined
Mar 1, 2018
Messages
10,392
Likes
29,983
Location
Gold Coast, Queensland, Australia
You should consider re-running the PM-90 test in class AB. Reason? Class A pulls a lot more current and you may be seeing PSU spikes that may not appear in Class AB on the higher voltage windings and lower load.

You will hear the transformer more in Class A mode. My own PM-95 makes more transformer noise in Class A.

Great job though!

Note: What is the 70Hz spike all about? We are 50Hz, so we would be seeing 50/100/150/200 etc. The UB9000 has it, so you may simply be passing that through the system and amplifying it...And it modulates with your mains spikes creating some of the spray.
 
Last edited:

Matias

Major Contributor
Forum Donor
Joined
Jan 1, 2019
Messages
3,694
Likes
7,077
Location
São Paulo, Brazil
Very nice comparison!

I am a little disappointed of the PM-10 power supply noise though. With better isolation from that toroid in the middle it could have zero distortion like most other NCore based amps with SMPS.

Edit: after remeasuring it is fine.

pm10.jpg
 
Last edited:

restorer-john

Grand Contributor
Forum Donor
Joined
Mar 1, 2018
Messages
10,392
Likes
29,983
Location
Gold Coast, Queensland, Australia
The PM-90 is a better unit because it looks better. who cares about sound when amps have with wooden sides. Hook it up to a turntable and it will be the coolest system.

The sides are a magnesium alloy die-cast block, not wood. Around 1kg each.
 

restorer-john

Grand Contributor
Forum Donor
Joined
Mar 1, 2018
Messages
10,392
Likes
29,983
Location
Gold Coast, Queensland, Australia
I am a little disappointed of the PM-10 power supply noise though. With better isolation from that toroid in the middle it could have zero distortion like most other NCore based amps with SMPS.

The PM-10 has twin SMPSs, one for each channel (to the left). The toroid is for the rest of the circuitry and the spikes are not at mains frequency (50Hz multiples for us). It's likely from the source.
 

Matias

Major Contributor
Forum Donor
Joined
Jan 1, 2019
Messages
3,694
Likes
7,077
Location
São Paulo, Brazil
The PM-10 has twin SMPSs, one for each channel (to the left). The toroid is for the rest of the circuitry
Yes, you are right.

and the spikes are not at mains frequency (50Hz multiples for us). It's likely from the source.
Isn't it 60 Hz and multiples, so mains interference? At least the 2 first ones look like 60 and 120 Hz.
 
OP
G

GXAlan

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
1,596
Likes
2,425
You should consider re-running the PM-90 test in class AB. Reason? Class A pulls a lot more current and you may be seeing PSU spikes that mey not appear in Class AB on the higher voltage windings and lower load.
Yeah, I did not think about that and did run it class A. I didn’t spend too much time on the PM-90 since Amir already tested it but wanted to give a comparison to known measurements and what my setup is achieving.

Great job though!

What is the 70Hz spike all about? We are 50Hz, so we would be seeing 50/100/150/200 etc.
I am in the US. The PM-90 is 100V 50/60Hz while the PM-11s2 and PM-10 are normal 120V models, even though gold/champagne is rare in the US and black is more common.

You mention it could be from the source. I will have to try my SACD player once my 1 kHz SACD test disc arrives.
 

Saidera

Senior Member
Joined
Jul 18, 2020
Messages
372
Likes
286
Location
Adelaide, South Australia, Australia
Amazing! In particular, I really like the fact that the E1DA Cosmos ADC is used, and how you have further proven that the impact of historical events did affect audio design, as we'd discussed before!
 

Tks

Major Contributor
Joined
Apr 1, 2019
Messages
3,174
Likes
5,274
Awesome work. Though honestly surprising to see a regression with noise of all things. I thought at least it would be basically the same, but having to hunt down "cleaner" PS and hope that leads to better results is kinda depressing.
 

Grooved

Addicted to Fun and Learning
Joined
Feb 26, 2021
Messages
573
Likes
362
Nice job!
Strange that Marantz say on their website that PM-11S2 and PM-10 have THD at 0.01%, so -80dB while you get -130dB with PM-10
They certainly forgot two "0"
On the other hand, they say between 111 and 113dB SNR, and you get 93dB
 
OP
G

GXAlan

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
1,596
Likes
2,425
Nice job!
Strange that Marantz say on their website that PM-11S2 and PM-10 have THD at 0.01%, so -80dB while you get -130dB with PM-10
They certainly forgot two "0"
On the other hand, they say between 111 and 113dB SNR, and you get 93dB

That SNR is at rated output probably and I am measuring closer to 5 and 10 Watts only.
 

Grooved

Addicted to Fun and Learning
Joined
Feb 26, 2021
Messages
573
Likes
362
That SNR is at rated output probably and I am measuring closer to 5 and 10 Watts only.
I thought about it too, but didn't think it should make such a big difference
 

McFly

Addicted to Fun and Learning
Joined
Mar 12, 2019
Messages
845
Likes
1,666
Location
NZ
yeah those 70hz spikes - something not right there. re measure (perhaps with a topping e30 - super clean and independent mains supply i.e. doesnt use USB power when using as a USB DAC)
 
Last edited:

PeteL

Major Contributor
Joined
Jun 1, 2020
Messages
2,749
Likes
2,953
Isn't it 60 Hz and multiples, so mains interference? At least the 2 first ones look like 60 and 120 Hz.
I think what he means is that it would be 50Hz for him, Restorer John’s Thumb nail indicate he is located in Australia, a part of the world where AC undulate at 50 Hz, as many other countries.
Edit. I do not know the location of the OP. Yes I see what you mean, it looks close to 60, but by »us » John seem to know him , so I guess Your question is relevant. It does look like main induced to me too in the Americas.
Edit 2, hah ok GxAlan answered… OK I know where the exit is thanks…
 
Last edited:

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
1,956
Likes
2,497
Location
Liège, Belgium
My setup is about ~9 dB worse than Amir's APx555
I think it might be that the mains noise you're capturing is due to your cabling.
Everything from amp to load to ADC is unbalanced, isn't it ?
Could you shorten those cables, maybe ?
Did you measure the loopback DAC+ADC with the same cabling (without the load resistor) ?
 
Top Bottom