Audiosciencereview.com/. legit?

AcuDefTechGuy

AcuDefTechGuy

Audioholic Jedi
the only way to know if a system is accurate is to measure at the listening seat and verify & adjust with recordings made in the same room.
Let's say you are intimately familiar with the sound of the drums in your room.

Then you did an Auto Room EQ. Multiple graphs confirm that the FR of the drum sound is ruler-flat.

Setup#1. But when you verify to make sure that it sounds exactly like the drums, it doesn't sound quite like it. Something is off. The graphs all show that the FR is extremely accurate. But the sound is not like the real drums.

Setup#2. Then when you bypass the Auto Room EQ, the graphs are no longer perfect ruler-flat. But now the sound is exactly the same as the real drums.

Which setup would you keep?
 
Verdinut

Verdinut

Audioholic Spartan
A second sample? Amir usually tests products sent to him by consumers who purchased them, not manufacturers. This is one reason why I think his testing is so interesting. So far, Amir is the most serious product tester I've seen in consumer electronics. Even the better channel of the QSC DCA2422 he tested was mediocre in objective performance, though his main issue with the product was fan noise, which he considered unacceptable for home environments. If the QSC out-performed otherwise I'd have to hear the fan noise for myself, but as classical music fan I've already sold one pair of fan-cooled amps in the past (Krell KMA-100 II) because the fan noise annoyed me, so I'd be skeptical. As it happens the QSC looks like a maximum power-optimized implementation that measures no better than most cheap AVRs, if not worse.

In the case of the Benchmark Media DAC3, for example, Amir found that a sample sent to him did not measure as expected. Benchmark's VP of Engineering joined the thread, claimed the sample was defective, and sent Amir a factory sample. Lo and behold the measurements of the factory sample were superior. Supposedly the product owner's unit was sent back to Benchmark, but the outcome never appeared in the thread. I'm curious because I use a DAC3, and I'm wondering about Benchmark's true field performance.
The fan noise on the QSC DCA Series of amps is not that bad. I am using four of them in my HT and I sit about 6 feet away from them. I admit that I can hear the fan noise on the four channel DCA 1824 as it has to cool 4 output stages instead of 2 in the same size chassis. But the fans on the other three 2 channel amps operating simultaneously don't bother me. When I play music, I don't notice either of the four amps.

As for QSC published specs, I think that you should have a look at them. You tend to generalize about their real performance, comparing them to cheap AVR amps. You don't seem to have much knowledge on pro audio gear. QSC has been producing pro audio stuff for over 50 years and if their products did not meet their published specs, professional musicians, studio equipment installers and cinema equipment installers would have stopped using their products, and that also includes loudspeakers, DSPs and other sound processing equipment. That company actually replaced Altec Lansing worldwide for complete audio reinforcement systems in cinemas, stadiums and other large venues. Most Cineplex theaters in North America are exclusively using QSC audio equipment and loudspeakers.

A friend of mine, who is a musician playing in various gigs, and has been using numerous brands of pro amps over the years. has blown up several of them and the only amplifier brand that he couldn't kill is QSC.
 
Irvrobinson

Irvrobinson

Audioholic Spartan
Let's say you are intimately familiar with the sound of the drums in your room.

Then you did an Auto Room EQ. Multiple graphs confirm that the FR of the drum sound is ruler-flat.

Setup#1. But when you verify to make sure that it sounds exactly like the drums, it doesn't sound quite like it. Something is off. The graphs all show that the FR is extremely accurate. But the sound is not like the real drums.

Setup#2. Then when you bypass the Auto Room EQ, the graphs are no longer perfect ruler-flat. But now the sound is exactly the same as the real drums.

Which setup would you keep?
Setup#2, of course, but that situation hasn't happened to me ever. My two biggest challenges have been that the Salon2s like being several feet from any boundary for smooth mids and highs (they especially seem to like high ceilings),and that getting the bass right once the Salon2s are placed for mids and highs is a PITA, and it takes at least one sub with at least four bands of parametric equalization.
 
Last edited:
Irvrobinson

Irvrobinson

Audioholic Spartan
As for QSC published specs, I think that you should have a look at them. You tend to generalize about their real performance, comparing them to cheap AVR amps. You don't seem to have much knowledge on pro audio gear.
Not correct. My wife is a gigging musician, and I'm unfortunately more familiar with pro audio equipment, including amps, mixers, PAs, and mics than I really care to be. Loud and rugged are the success factors, and the sort of performance many of us expect in home equipment or for use with acoustic instruments, especially for speakers and mics, isn't typically there. I'm especially familiar with QSC, JBL, and Crown. I don't mind saying I truly despise all-in-one PA systems, I think they suck, and given the quality I've seen in pro audio passive speakers, which also suck compared to most modern high-quality home passive speakers, I can see that if pro audio amplifiers have marginal performance, like that QSC, no one would notice. I do know of two folks that use Crown amps in their home theaters, but they're action movie fans, and there again loud is what counts, not accuracy.

One thing that really pisses me off, is when a trumpet player or a sax player in a medium venue clips one of those damned horn mics on his bell, and feeds it to the PA system. For the time being at least my wife is only in acoustic groups, so I'm spared the torture of pro audio speakers and PA systems.
 
Irvrobinson

Irvrobinson

Audioholic Spartan
22.5kHz BW means 20Hz-22.5kHz, right?

If he could do 1 watt SNR, that would be more like what we are used to seeing on AH.
On Audio Precision equipment I believe the 22.5KHz limiter is an AES-17 filter, which is low-pass only, so the SNR is 0-22.5KHz.
 
AcuDefTechGuy

AcuDefTechGuy

Audioholic Jedi
Setup#2, of course, but that situation hasn't happened to me ever.
It's good that the measurement matches the sound. It's the same with PENG.

But some people don't have the matching linear-graph-to-sound, including Amir.

I believe Amir was saying this - the REW Auto Room Correction produces a much linear response graph for him, but it actually sounds worse to him than the manual graph that isn't as flat/smooth.

Now I sound like I actually know "Amir" or something. :D
 
Last edited:
lovinthehd

lovinthehd

Audioholic Jedi
Amir has a high-end infatuation but measures $99 DACs? Do you see anyone else measuring low-cost equipment?

I find it fascinating that Amir runs a wide array of measurements, posts the results for anyone to see, takes no advertising, and then gets criticized by people who do nothing but post opinions. I admire the guy because he makes tangible contributions to increasing the objective information available in the field. Who does better? Personally, I take most of the commentary from any site, including this one, as just opinion, and I don't take it all that seriously. The measurements, on the other hand, are invaluable.
Nwavguy designed one as a result of his measuring some. Amir probably tests if anything too many dacs (that's quite a list he has going, none of which are particularly interesting to me as I don't use a stand alone dac). Why is he doing it, just for break-even money? He can afford his hobby/passion it seems. Maybe audioholics should review a couple....
 
Irvrobinson

Irvrobinson

Audioholic Spartan
Nwavguy designed one as a result of his measuring some. Amir probably tests if anything too many dacs (that's quite a list he has going, none of which are particularly interesting to me as I don't use a stand alone dac). Why is he doing it, just for break-even money? He can afford his hobby/passion it seems. Maybe audioholics should review a couple....
Amir seems to be doing it for fun. Being a former CVP for Microsoft (meaning he was a company officer),I would guess Amir is... financially independent. ;-) His web site does accept donations via patrons, but I think it's a fair guess that Amir is just funding most of it himself, like that APx555. I haven't discussed this with Amir, but I think he acquired the AP box and he looks for useful things he can do with it, and the APx555 is perhaps the best unit on the market for testing amps and DACs. Notice that for AVRs & AVPs he doesn't test anything related to video; his testing is audio only. There is a thread on the site now about whether or not ASR should get into testing speakers, which is a heck of a lot more work, but cheaper equipment.
 
Bucknekked

Bucknekked

Audioholic Samurai
AH Brethren:
A most interesting thread. I must confess, most of the conversation was right at the hairy edge of my skill set.
What I find comforting was that everyone's comments were contributing to the positive. Everyone seemed to be trying to clarify and make clear their points. That's refreshing. I enjoyed the conversation.
 
lovinthehd

lovinthehd

Audioholic Jedi
Amir seems to be doing it for fun. Being a former CVP for Microsoft (meaning he was a company officer),I would guess Amir is... financially independent. ;-) His web site does accept donations via patrons, but I think it's a fair guess that Amir is just funding most of it himself, like that APx555. I haven't discussed this with Amir, but I think he acquired the AP box and he looks for useful things he can do with it, and the APx555 is perhaps the best unit on the market for testing amps and DACs. Notice that for AVRs & AVPs he doesn't test anything related to video; his testing is audio only. There is a thread on the site now about whether or not ASR should get into testing speakers, which is a heck of a lot more work, but cheaper equipment.
I'd assume he's quite comfortable. :)
 
P

PENG

Audioholic Slumlord
Let's say you are intimately familiar with the sound of the drums in your room.

Then you did an Auto Room EQ. Multiple graphs confirm that the FR of the drum sound is ruler-flat.

Setup#1. But when you verify to make sure that it sounds exactly like the drums, it doesn't sound quite like it. Something is off. The graphs all show that the FR is extremely accurate. But the sound is not like the real drums.

Setup#2. Then when you bypass the Auto Room EQ, the graphs are no longer perfect ruler-flat. But now the sound is exactly the same as the real drums.

Which setup would you keep?
Obviously setup #2, but If he did the recording in his room and do then do what you described, using the recording, then I would bet money 2:1 setup#1 has a better chance to sound closer, then it would be the one to keep.
 
AcuDefTechGuy

AcuDefTechGuy

Audioholic Jedi
Obviously setup #2, but If he did the recording in his room and do then do what you described, using the recording, then I would bet money 2:1 setup#1 has a better chance to sound closer, then it would be the one to keep.
The point is to use the setup that produces the best sound.

If that happens to be the one with the best graph most of the time, then fine. If not, that’s okay too.
 
B

Brian Smith

Enthusiast
Maybe Gene will test the 8805 and/or the 8500? Any chance Gene? Any comments on the test?
 
P

ProFan

Enthusiast
There is no correlation between Amir’s tests at overloading levels and sound quality of a particular device, and other similar devices.
For example, Amir tested 8805 at 4V RMS output while the manufacturer’s manual clearly stated that the max output voltage is 2.4V RMS. In addition, Amir doesn’t follow the AES17 requirements and instruction manual of his AP555 testing all units at 0 dB FS instead of required -1 dB FS for THD+N.
Another example - my Lynx L22 sound card starts clipping in loop at -0.043 dBFS (SINAD 92 dB at 24 bit 96 kHz),and at -0.044 dBFS there is no clipping (SINAD 106.48 dB at 24 bit 96 kHz). It is just 0.001 dB difference in levels, i.e 0.012%. At 0 dB FS the measured SINAD is 67 dB. It is obvious how SINAD drops down with increase of the clipping levels.
Another Lynx L22 measured by Archimago seems doesn’t clip at 0 dB FS; does it mean my card is defective and his card is good talking about the sound quality? Of course, not: -1 dB FS gap is required by AES17-1998 rev 2009 for THD+N tests using a single sine tone. The specifications with THD+N measurements for Lynx L22 are also made at -1 dB FS, the industry standard.
Tolerance in used by a manufacturer components can create difference in testing at the border level of clipping, the same type of product (just different pieces of it) can clip or not, making a huge difference in measured SINAD. This is why we need some reserve.
Amir’s measurements can not be considered as legit - and correlate with the sound quality of UUTs (units under tests), IMHO.
They can provide some information about overloading capabilities of particularly tested by him units.
 
Last edited:
P

PENG

Audioholic Slumlord
For example, Amir tested 8805 at 4V RMS output while the manufacturer’s manual clearly stated that the max output voltage is 2.4V RMS.
That is not correct, the 2.4 Vrms is the "Rated" XLR balanced output. Marantz is anything but clear about their definition of "Rated output", but it is quite obvious that their "Rated" output is not the same as "maximum" output.

According to the AV8805 manual:

1564772570152.png




For example, according to the Yamaha CX-A5200 manual:

Rated output level RCA unbalanced is 1 V, XLR balanced is 2 V
Maximum output level RCA unbalanced is 4 V, XLR balanced is 8 V. (0.06% THD)

The vol. IC of the AV8805's output spec is 4.2 Vrms unclipped, maximum. So the balanced XLR output will likely be about 8 Vrms as well, like Yamaha's.

Gene measured the Denon AVR-X330W and the Marantz SR8012, below were his comments on the X3300W and the SR8012's pre out voltage.

https://www.audioholics.com/av-receiver-reviews/marantz-sr8012-receiver/marantz-sr8012-bench-tests
https://www.audioholics.com/av-receiver-reviews/denon-avr-x3300w-1/measurements

"One thing I really love about Denon receivers is they NEVER skimp on their preamp out circuits. Like past models, the AVR-X3300W had an ample amount of drive. I measured a whopping 4.5Vrms unclipped output, which is more than double the voltage needed to make virtually any amplifier reach full rated power. Yamaha please pay attention and step your game up particularly with your AV receivers in this price range that clip above 1.6Vrms. A preamp output of less than 2Vrms is unacceptable in my book."

"The SR8012 is capable of outputting 4.5Vrms unclipped from the multi-ch preamp outputs which is more than 2X voltage drive needed to make most external amplifiers reach full unclipped power."

Gene has not measured any Marantz prepro, but it is reasonable to assume the AV8805 will do better than the SR8012 on his bench too.
 

Attachments

Last edited:
P

ProFan

Enthusiast
That is not correct, the 2.4 Vrms is the "Rated" XLR balanced output. Marantz is anything but clear about their definition of "Rated output", but it is quite obvious that their "Rated" output is not the same as "maximum" output.
Thank you, Peng! I believe it is an important matter to discuss.
The definition of "Rated Output Voltage" is not a standard indeed and can be considered by different ways. For example, some electrical engineers believe that:
Rated simply means maximum amount of current and maximum amount of voltage which any particular Electrical Equipment can withstand or carry when it is performing at its peak capacity - by Vatsal Talati, works at Electrical Engineering
While other engineers believe that:
When you say "rated" it means the highest value that the corresponding equipment can withstand - by Prachal Jadeja, M.E. Electrical Engineering, Marwadi Engineering College (2016)
What Marantz means is unclear indeed. However we can not approximate the the Yamaha's numbers, may be they mean completely different things?
Such as Marantz doesn't provide any additional data for the maximum voltage, I believe we shall assume that their numbers correlate with the maximum output voltage due to the following reasons:
1. 1.23Vrms in unbalanced mode means +4dBu Pro Audio Reference Level (ref Rane note 145), anything above it is a "headroom" up to +26 dBu Maximum Output Level in professional Audio. Consumer Reference Level is -10 dBV, and Marantz 8805 is not a professional device. Although it can obviously output higher levels that +4dBu, the measurements have to be performed at this level (or lower)as a rated level with provided level of THD+N.
2. Previously the measurements for Marantz 8802 have been made at 2Vrms output voltage by hometheaterhifi.com with completely different numbers for THD+N comparing to Amir's. They also performed such measurements at 5Vrms to demonstrate the overloading capabilities of the device, and those numbers are pretty similar to Amir's results. It tells us about the actual overloading, and such as the DAC circuit in 8802 and 8805 is similar, we can compare the measurements.
3. "whopping 4.5Vrms unclipped output"
It is unclear for me why do we need it, if majority of amplifiers "will be happy" with significantly lower input levels.
Bryston 4b sensitivity at 1.4V for 250W at 8 ohm (it has switchable gain 23dB for balanced or for unbalanced 29dB).
Onkyo 5501 sensitivity 1V balanced, 2V unbalanced
Proceed AMP 5 sensitivity 2.24V for full rated output (balanced),1.12V for full rated output (single-ended or unbalanced)
 

Attachments

P

PENG

Audioholic Slumlord
Thank you, Peng! I believe it is an important matter to discuss.
The definition of "Rated Output Voltage" is not a standard indeed and can be considered by different ways. For example, some electrical engineers believe that:
Obviously you have now done your research reading so we can discuss (be clear, just discussion) some more. Below are my additional points:

1. "Rated", by itself is just a word, rated speed, temperature, output power, distortions, frequency response would not mean much unless the conditions, e.g. atmospheric pressure (for speed), humidify (for temperature), frequency, load impedance (for amps) etc., under which the equipment, device etc., are measured.

Specifically for pre out voltage, it could be a little more useful if the spec include more details such as:
continuously rated Vrms, Max Vrms (specified max duration), Distortions specs for full bandwidth, input sensitivity, gain, input impedance, output impedance, SNR (A weighted/unweighted)

While other engineers believe that:
That's just what they believe right? I am not going to say the quoted ones were taken out of context as I have not read the whole thing, but I suspect they did not mean what they stated as "definition", because they should know better that one cannot define "rated whatever" without stating all the conditions under which the measurements were taken.

In one of your quote: "Rated simply means maximum amount of current and maximum amount of voltage which any particular Electrical Equipment can withstand or carry when it is performing at its peak capacity - by Vatsal Talati, works at Electrical Engineering.."

That makes little sense, specifically, "Rated" what?, I gathered that he/she meant rated current, or voltage in this case, and I suspect if you had quoted the whole article it would make some, or total sense. Another example, I am sure you would agree that "can withstand............" means little, without the other conditions such as, but not limited to duration, ambient temperature, without/without forced cooling etc..

What Marantz means is unclear indeed. However we can not approximate the the Yamaha's numbers, may be they mean completely different things?
Agreed, and that's the exact point I tried to make.

Such as Marantz doesn't provide any additional data for the maximum voltage, I believe we shall assume that their numbers correlate with the maximum output voltage due to the following reasons:
1. 1.23Vrms in unbalanced mode means +4dBu Pro Audio Reference Level (ref Rane note 145),anything above it is a "headroom" up to +26 dBu Maximum Output Level in professional Audio. Consumer Reference Level is -10 dBV, and Marantz 8805 is not a professional device. Although it can obviously output higher levels that +4dBu, the measurements have to be performed at this level (or lower)as a rated level with provided level of THD+N.
2. Previously the measurements for Marantz 8802 have been made at 2Vrms output voltage by hometheaterhifi.com with completely different numbers for THD+N comparing to Amir's. They also performed such measurements at 5Vrms to demonstrate the overloading capabilities of the device, and those numbers are pretty similar to Amir's results. It tells us about the actual overloading, and such as the DAC circuit in 8802 and 8805 is similar, we can compare the measurements.
Interesting points, but in the service manuals, it does say: "vol IC NJU72343....non clip max signal level 4.2Vrms...", and that's what I based my previously comments on The AV8805's is actually downloadable (try Googling for it) or you can read the data sheet of the volume IC yourself if you are electrically savvy. Note that hometheaterhifi.com's numbers are what they choose for their measurements, just like Amir's who could have used a lower voltage for his measurements too.

https://www.njr.com/semicon/PDF/NJU72343_E.pdf

The Marantz AV8805, SR6012 and up, Denon AVR-X3400H and up all use the same preamp/vol IC, so their "rated" pre out voltage are most likely the same or very similar.

3. "whopping 4.5Vrms unclipped output"
It is unclear for me why do we need it, if majority of amplifiers "will be happy" with significantly lower input levels.
Bryston 4b sensitivity at 1.4V for 250W at 8 ohm (it has switchable gain 23dB for balanced or for unbalanced 29dB).
Onkyo 5501 sensitivity 1V balanced, 2V unbalanced
Proceed AMP 5 sensitivity 2.24V for full rated output (balanced),1.12V for full rated output (single-ended or unbalanced)
You may not need it, but others may as it depends on how much average and peak power you need. THX standard for peaks is 20 dB above the average 85 dB SPL. Someone like @Irvrobinson, @RichB have the Revel Salon 2 that has relatively low sensitivity and impedance specs, so if they listen to media contents that have very say >20 dB dynamic peaks from 4 or even 5 meters in a large room, they may in fact need >4 Vrms or almost 6 V/12 Vp-p peak to avoid any amount of clipping, especially if their power amps have say less than 28 dB of gain. For example, the AT6000 power amp needs 2 Vrms to output 450 W into 4 ohms, or 300 W into 8 ohms. So if they sit 5M from the speakers, even with room gain, they would need more than 4 Vrms to ensure the preamp won't clip during dynamic peaks of 105 dB (yes, that would be rare) or higher from 5 meters assuming their amps such as the AT6000 won't clip during such short duration peak outputs.
 
P

ProFan

Enthusiast
Interesting points, but in the service manuals, it does say: "vol IC NJU72343....non clip max signal level 4.2Vrms
If you will take a look at the datasheet of NJU72343, you will see the following: AC characteristics: Maximum output voltage min 3.6Vrms, typ 4.2Vrms, max -. Those are the limit of tolerance regarding the maximum output voltage of the referred by you component.

It is exactly what I am talking about.

I really doubt that Marantz makes a components screening selecting only chips with typical max output voltage (it is a very expensive process and extremely rarely used in the consumer electronics industry),thus we have a chance that some units will have these components with 3.6Vrms-3.95Vrms maximum output voltage, correct?
Such units, being tested at 4.0Vrms, will show some clipping. Amir's graph shows even higher output voltage at 4.126Vrms/4.130Vrms, meaning that even more components out of the min-typ range can “fail”. Let’s say he got lucky and tested a unit with the chips having the typical maximum output voltage, in this case the results will be nice and he will praise the unit, right?
Alternatively, somebody provided a pretested “known good” unit for Amir’s tests ...
But what if not, and the component had 3.7Vrms maximum output voltage, being still within the allowed limits of the datasheet, but out of the 4.13Vrms testing level? The unit will be blamed, based on just one result of the test (which failed, showing lower SINAD than expected)?
 
Last edited:
P

ProFan

Enthusiast
You may not need it, but others may as it depends on how much average and peak power you need
You are talking about the overloading limits again. The solution is simple - for high power speakers with low sensitivity just use more powerful amps with higher sensitivity. This case is an exception though.
 
Irvrobinson

Irvrobinson

Audioholic Spartan
Alternatively, somebody provided a pretested “known good” unit for Amir’s tests ...
But what if not, and the component had 3.7Vrms maximum output voltage, being still within the allowed limits of the datasheet, but out of the 4.13Vrms testing level? The unit will be blamed, based on just one result of the test (which failed, showing lower SINAD than expected)?
Amir gets the largest fraction of his test samples from site members, not manufacturers. In two cases referenced in this thread, a QSC amplifier and a Marantz pre-pro, and additionally a Benchmark Media DAC3, the samples sent in for testing did not meet specifications or performed oddly.
 
newsletter

  • RBHsound.com
  • BlueJeansCable.com
  • SVS Sound Subwoofers
  • Experience the Martin Logan Montis
Top