Mackie ONYX ... the truth: a big fake!

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

TobWen

Well-known member
Joined
Sep 9, 2004
Messages
195
Location
Germany, Ruhrgebiet, Dortmund
Hi there,

yesterday, I worked with a Mackie Onyx 4880.
I think, you all know it: aprox. $5500 48 mono channels etc.

Website says:
Onyx Mic Preamps
123dB total dynamic range
-129 dBm Equivalent Input Noise
0.0007% Total Harmonic Distortion
and now the funny thing: it says for real

You can read that on: http://www.mackie.com/products/onyxseries/preamps.html

Alright, I was interested on how they did it.
So I opened the device and check the mic-channels.

I began to cry (you can ask a friend of mine).
They're using NJM2068 OPAs and cheap 2SA-Yapan transistors.

It's even theoretically impossible to reach 0,0007% THD because the NJMs have 0,001% at 100% ideal values and this is laboratory only!!
It's a big marketing-fake.

About the "-129.5 dBm Equivalent Input Noise" ... yeah, you could reach this, if you make a short circuit!! But in "real life", you can reach about -110dBm - again only if everything is 100% ideal.

I informed the consumerism / consumer protection, since it is a big fraud on customer. Something like that I've never seen a long time.

What do you think about it?? I am REALLY upset and angry!!!

Best,
Tobias
 
Further proof that specs are worthless. Worse, they lie -- the better sounding product usually has worse specs. I'll bet the specs for the Mackie are accurate, and that a Neve 1073 would spec out worse.
 
[quote author="synthetic"]Further proof that specs are worthless. Worse, they lie -- the better sounding product usually has worse specs. I'll bet the specs for the Mackie are accurate, and that a Neve 1073 would spec out worse.[/quote]

Trust me ... even if they released the correct (worse) specs, it WOULD NOT sound good.
It's a standard circuit. Nothing "magic", nothing "onion" or "onyx" or whatever.

You cannot make good sound with bad silicium, trust me.
Two bad OPAs cannot make one good. Four non-selected BC550 cannot sound like two selected ones.
 
I had an Onyx 800r for a review in a German magazine. I'm not known to be a Mackie lover and in fact don't own any Mackie gear. But I feel I have to correct you there.

The 800R at least uses NJM2114 opamps, which are enhanced NE5532s - double the slew rate and overall improved performance.

EIN of these preamps is in fact very low. The best way to test this in a real life application is to use a low output ribbon mic. I did test that way, and noise here was as low as it gets. Very near the theoretical minimum. -129 dB at 150 ohms seems totaly credible to me.

I don't know how you measured EIN, but I'm sure you didn't do it the correct way. An EIN of -110 dB would mean a very noisy preamp. Here's how you do it: you connect an 150 ohms resistor to XLR pins 2 and 3. Turn the gain all the way up. Measure the noise floor. Audio programs often are not very precise, btw. But let's say your program says -91 dB. You have to take into account the analog to digital level relation. The Onyx 800R converters, I think, were calibrated at +22 dBm -> 0 dBfs. So you have to add 22 dB and get a noise floor of about 69 dB below 0 dBm
Now substract the preamp gain from the noise value. There you go: you have an equivalent input noise of -69dBm -60db = -129 dBm.

The cheapo SA-transistors you refer to are in fact 2SA1316. Those are special ultra low Rbb transistors and about the best transistors you can possibly use for a transformerless mic input stage. Those transistors are responsible for the in fact very low EIN. Get the datasheet from www.alldatasheet.com.

Again, I'm in no way affiliated with Mackie. But you better do your homework and double check your results before you bash a manufacurer or their product, let alone contact consumer protection with wrong information.

Cheers!
 
mackies are extremely consistant and pretty reliable. but, i once saw on a 1604 "192khz digital ready". i mean i guess you could hook a 192 digital thingy to it, but i think they were trying to get google hits.
 
Okay, Rossy, but what do you say about the THD? Watch the datasheet of the NJM2068: http://www.ortodoxism.ro/datasheets/newjapanradio/ae04022.pdf
The value Mackie gave is simply wrong.

And do you really think, it's average -129 dB?
I don't think Mackie selects every transistor or OA ... they just grab, what they get, like every big manufactor.
 
TobWen

You have way too much time on your hands.
#1 Why are you wasting your time.
#2 You could not do better designing a product at that price point.

You need to give it to Greg a nice product a a good price point.
At that price point you cannot have a good workable product
like that and still have it sound great. You had to cut corners
with parts cost which will lead to parts being used that do not sould
great. But try it sometime to design product like that it is not as easy as it looks. And when you succeed total you cost of you bill of materials.

It was bad carma to report them to consumer protection.

No I do not own any Makie gear as I do not like the sonics
of their gear. But I do own a Tapco 6000!
 
You have to cut corners, true. But why lie to customers??
Why not say "people, we have bad specs, but we sound good".

It's not the advert, which lies ... the specs are lying.
If the advert would be wrong: I don't care ... But the measurements are not correct.

And adrianh, who tells you, I can't build a nice gear like that?!
Wait some days after Musikmesse and you'll find some nice projects to DIY.
 
I don't know enough about THD to answer if you can have lower THD than that of the chip you use. But be aware that it always depends on how you measure. The naked THD figure doesn't tell you very much about how clean it sounds. As has been elucidated by more knowledgeable people than myself, the relation between 2nd, 3rd and higher harmonics is more crucial than the total figure.

Yes you can have an EIN of -129 dB without selecting components. Intra-batch consistency is pretty good these days. Plus, these transistors come in different ratings (BL in Mackie's case I think), so they're not all over the place. The opamps behind the transistor stage are not very crucial for the noise performance.

FYI: I have an old B*ringer preamp that has an EIN of -128 at 150 ohms (according to my own measurements). I don't think they selected any components. The key to the good EIN figure is just the 2SB737S transistors - an older super low Rbb type. I have several transformer coupled preamps that sound better but none that produces a lower EIN figure.
 
The sad truth is:

If there's money to be made, companies will make the money.

Every decently large establishment has a room full of people who sit around all day and try to figure out how to milk the consumer for as much as possible. Unfortunately, this attitude seems prevalent over customer service. I guess we only have ourselves to blame for buying the crap, after all we collectively provide the consumer demographic for companies like Behringer and Mackie to F!@#$CK up the A$$.

FYI: I have an old B*ringer preamp that has an EIN of -128 at 150 ohms (according to my own measurements). I don't think they selected any components. The key to the good EIN figure is just the 2SB737S transistors - an older super low Rbb type. I have several transformer coupled preamps that sound better but none that produces a lower EIN figure.

Please post a phase, frequency, and square wave tests too!
 
[quote author="[silent:arts]"][quote author="TobWen"] ... the specs are lying ... [/quote]
did you do any measurements and how ?[/quote]

hell .. just watch the datasheets of the used chips!!!

datasheet of the NJM2068:
http://www.ortodoxism.ro/datasheets/newjapanradio/ae04022.pdf

have a look on the second sheet (page 4-87)
"Equivalent input noise voltage 0,44.....0,56uV"

Get a calculator or Sengpiel-Website and calculate it:
0.0000005 Volt = -123.80 dBu (-126.02 dBV)
Where are the missing 5,2 dB?

If Dunlop tells you, your tires/tyres could go 200 km/h max, they won't work up to 280 km/h.
 
[quote author="TobWen"]If Dunlop tells you, your tires/tyres could go 200 km/h max, they won't work up to 280 km/h.[/quote]

that's just the guaranteed spec, and of course people are driving over 200 km/h with this tire.

I just find it kind of funny complaining the specs / datassheet of one (mackie) in regards to a datasheet of a component they used (NJM2068). may be they found a trick, engineered well, had some measurement tricks, I don't know.

I'm not a fan of Mackie and I don't have any of their equipment, but you are welcome to bring one with you and we will measure it on an Audio Precission System 2.
 
[quote author="Rossi"] 150 ohms resistor to XLR pins 2 and 3[/quote]

exactly this is the "short circuit trick" ...
if you want to simulate a microphone, get a bobbin and some caps
repeat the measures and calculate the avarage ...

that's reality.
 
[quote author="TobWen"][quote author="Rossi"] 150 ohms resistor to XLR pins 2 and 3[/quote]

exactly this is the "short circuit trick" ...
[/quote]

This is not the short circuit trick. Short circuit is short circuit.
This is the standard, accepted way of measuring noise, with an 150 ohm resistor between pin 2 and 3.
It is not a microphone but anyway it is an acceptable solution as long as there are no two identical microphones in specs.

Also you don't know how is the distortion specified in the datasheet. Negative feedback applied can lower distortion significantly.

chrissugar
 
Back
Top