Hitting the lynx aurora best possible

[Nu-tra]

I hit mine about -15

[iangomes]

I don't even really pay attention to that in the sense of sound.
Not too low? Check! Not clipping? Check!
Roll it!

[Gie-Sound]

Originally Posted by iangomes

I don't even really pay attention to that in the sense of sound.
Not to low? Check! Not clipping? Check!
Roll it!

Same here

[nickelironsteel]

ive heard -18 is optimum since converter chips are mainly designed for the cellphone industry these days

could be a rumor though

[Kingtone]

Originally Posted by nickelironsteel

ive heard -18 is optimum since converter chips are mainly designed for the cellphone industry these days

could be a rumor though

[Hyder boy]

Originally Posted by nickelironsteel

ive heard -18 is optimum since converter chips are mainly designed for the cellphone industry these days

could be a rumor though

WOW!
Just, WOW!

[greenhirny]

The aurora converters "sound" best at 96 k in my opinion. Drums hitting at -6 seem to be fine. Even some clipping is fine with this converter.

Generally keep in the -18 to -8 range and you are good

[Doc Mixwell]

Originally Posted by atticmike

Hey there,

The topic might be misleading in the first place but I want to simply know what's the best level to send audio to the aurora while recording?

Just the best level to hit the converter and analog circuitry?

Maybe I should ask this kind of question on the official aurora forums?

Mike

Hi Mike,

Well, it depends on the source material and context really, and how you want to use your analog hardware. But that's just my view of the whole thing. I don't think I typically print everything to the same levels, because everything has different relative volume and loudness. It all depends on the music/context that I am working on. Though one thing is sure! Make sure the Signal Defeat's the Noise, and Keep the Peaks from reaching into Poor Analog Saturation and Harsh Digital Distortion.

As a general guideline, using the reference calibration to 0vu analog, there is a way to see how far over 0vu you are headed, versus how close you are to DEATH [0dbFS] squaring off the waveforms. I seem to recall the Aurora is reference calibrated to -16dbFS = 0VU, +4 [unless your using the "VT" Variable Trim Version] So that means, when your mic preamp is nearing 0vu, you should be seeing a meter reading of around -16DBFS on the Aurora.

Everything in analog has a different threshold for distortion, and as you increase the levels over 0vu, you are increasing saturation. Your reference calibration lets you estimate how far over zero [on the analog gear] you really wanna be traveling, versus how much digital headroom you wanna leave for later on. Headroom on analog gear and in digital, is a very good thing.

I can tell you, that I might travel between -16 and -6dbFS, on the highest peaks on the source, meaning my average signal level is around -18, 16 and my highest and loudest peaks are nearing -6dbfs. But still, I try to scale the headroom for each specific sound, when I am recording. Often takes foresight about how the project will be mixed, arranged, manipulated, processed, et al.

[Nu-tra]

I don't think people realize that about digital recording. They try and get the level as hot as possible like the good old Adat days. My recordings come out much nicer rms -18-14 instead that grey sound you get when clipping the analog path before it hits the AD. I get more clarity and my faders aren't almost all the way down during mixdown.

[greenhirny]

Peaking -18 is way too low. RMS -18 is more like it. That usually means that serious transients (kik/Snare) can peak anywhere between -12 and -6.

[Nu-tra]

Sorry meant rms

[Rev. Eslam]

Originally Posted by greenhirny

Peaking -18 is way too low. RMS -18 is more like it. That usually means that serious transients (kik/Snare) can peak anywhere between -12 and -6.

Agreed!

[atticmike]

Alright, thanks a lot for all your input. Looks like the converter + analog circuitry operates at wide level range properly. You know back in the day with the early digital converters, you had a golden range where the converters sounded the best. Guess not the case with the lynx aurora and contemporary converters anymore?

[dangerousben]

Shouldn't be a problem with converters in the Aurora's class. But watch your levels when going DAD for the use of analogue outboard during mixing.

[greenhirny]

Originally Posted by atticmike

Alright, thanks a lot for all your input. Looks like the converter + analog circuitry operates at wide level range properly. You know back in the day with the early digital converters, you had a golden range where the converters sounded the best. Guess not the case with the lynx aurora and contemporary converters anymore?

this design is from the early 2000's. Every converter DOES have a sweet spot. You have to find it given the material you are working on and the approach you are using. Too many variable to say your mix should be hitting X.

[atticmike]

Originally Posted by greenhirny

this design is from the early 2000's. Every converter DOES have a sweet spot. You have to find it given the material you are working on and the approach you are using. Too many variable to say your mix should be hitting X.

depending on the consistency or rms you're creating but still, it's not going to change, you just have to hit it appropriately then which of course varies by the type of you music you're recording I agree. Yet, you can determine that specific value of rms you want to go for, if not audible.

[EisenAudio]

Lynx Aurora 16, while potentially excellent sounding and fun to use, does not have the most headroom. Here's why:

Assuming a standard +/-15V power supply, the OP-275 op amps comprising the analog section can accept inputs of up to 10.5Vp before exceeding specified acceptable distortion. This equates to 7.425Vrms or +19.63dBu. Ok, so that's lower than the anywhere from +20 to +28dBm max output that quality mic preamps are capable of producing, but Lynx have probably implemented a differential receiver circuit having -6dB loss. This means analog (theoretically) won't actually clip until +25.63dBu (a conservative +24 is probably closer to reality), and even if I'm wrong, we can still watch our levels and make do.

However, there is another limitation. I haven't double checked lately, but I recall the stock (sans variable trim) Aurora A/D is calibrated for -20dBFS = 0dBu, as are many converters. This means a +20dBu input yields 0dBFS, also known as the onset of digital clipping. So, even if the power supply is a few volts higher for a few more dBu in analog domain, the digital output will still clip a few dB sooner, and it all clips before the majority of transformer balanced analog outboard and consoles that people like to track through.

Adding insult to injury is leaving all your software faders at 0 before bouncing to disk. For instance, (and I realize this is not a realistic example) let's say you have 16 tracks of random noise each idling at -12. The summed result is about 0. In other words, loud tracks combine to form an even louder 2-mix which clips your busses.

Now, if you're overdriving transformer-balanced preamps (to get thicker tones), then you can see how it's pretty easy to clip the Aurora A/D. In my experience, this will still sound forgiving on a couple tracks out of several, i.e. your kick and snare hit the red once in a while. However, if you push all of your tracks, then like me you may begin to hear a low level upper midrange haze across the mix. Some of this is the software mixbuss and some of it's the Lynx. This is called distortion, and it makes a lot of sense when considering the numbers explained above.

To answer the original poster's question, I find Aurora 16 sounds best when externally clocked and trying to print tracks which live around -18 in Pro Tools, depending on track count.

[greenhirny]

Originally Posted by atticmike

depending on the consistency or rms you're creating but still, it's not going to change, you just have to hit it appropriately then which of course varies by the type of you music you're recording I agree. Yet, you can determine that specific value of rms you want to go for, if not audible.

no that isn't true. it's completely program dependant. You can have two mixes with the same RMS value and be completely different sounding mixes of course.

RMS doesn't tell you much about relationships in a mix. hitting a converter with a snare, piano and tambourine are vastly different beasts even if they are peaking or averaging the same level.

It's material dependent. No two ways about that. your sweet spot will be different depending on what your are hitting it with.

[psycho_monkey]

Originally Posted by greenhirny

no that isn't true. it's completely program dependant. You can have two mixes with the same RMS value and be completely different sounding mixes of course.

RMS doesn't tell you much about relationships in a mix. hitting a converter with a snare, piano and tambourine are vastly different beasts even if they are peaking or averaging the same level.

It's material dependent. No two ways about that. your sweet spot will be different depending on what your are hitting it with.

Is it just me, or does anyone on a commercial session not have time to worry about converter "sweet spots" and so on?

I consider myself fairly conscientious - read that as "anal" if you like - but in 99% of commercial situations, you're under some form of time pressure. Of course you spend time on mic placement, moving things around, preamp drive, and if tracking to tape then definitely how hot you hit things. After all that, how much time is left to spend working out how to hit a converter's "sweet spot"?

Put me firmly in the "no clipping, sufficient headroom and go" camp. I find a decent yet balance-able level put to DAW, using the preamps in their optimum operating level, would vastly outweigh any advantages of a converter "sweet spot".

I'm not denying the truth of the underlying matter, and obviously the difference would be more pronounced with something like a Burl, the HEDD or a 2192, but the whole "try it at every single level to find the sweet spot for each individual source" kinda smacks of a Sound on Sound article. In the same way that you don't spend ages looking for a mic for the tambourine, you usually just use the vocal mic unless it sounds awful, the amount of time spent on "converter sweet spots" on a professional session is IMO minimal.

[greenhirny]

Yes I agree PM, and I don't spend time looking for converter sweet spots. I am answering the Op's direct question about there being one and his idea that it would be the same always.

I personally track with the " try not to clip, keep it reasonable and even try to get it like you are going to mix it if it is doable" approach

I also find when someone starts a sentence with "is it just me or..." and continues on with things like " in a professional setting" they are trying to cut down the person they just quoted.

Or maybe that's just me.

At any rate. I record professionally. So that argument is out the window. As you stated the 2192 has a much more pronounced sweet spot, and funny enough, the reason I got rid of mine was because I was worrying too much about that. Funny, I also was recording professionally at that time as well. In the end I decided that that was a waste of my time.

See how that works? You can improve your workflow and be a professional all at the same time.

I personally haven't met a professional that is perfect and knows everything, but I've met many that know how to get the job done despite any of their imperfections.

Now is that just me or...

[Kingtone]

Originally Posted by greenhirny

Now is that just me or...

I dunno,
but you definitely get the prize for the most instances of the word 'professional' (or a derivative) in a single gearslutz post!

[psycho_monkey]

Originally Posted by greenhirny

Yes I agree PM, and I don't spend time looking for converter sweet spots. I am answering the Op's direct question about there being one and his idea that it would be the same always.

I personally track with the " try not to clip, keep it reasonable and even try to get it like you are going to mix it if it is doable" approach

I also find when someone starts a sentence with "is it just me or..." and continues on with things like " in a professional setting" they are trying to cut down the person they just quoted.

Or maybe that's just me.

At any rate. I record professionally. So that argument is out the window. As you stated the 2192 has a much more pronounced sweet spot, and funny enough, the reason I got rid of mine was because I was worrying too much about that. Funny, I also was recording professionally at that time as well. In the end I decided that that was a waste of my time.

See how that works? You can improve your workflow and be a professional all at the same time.

I personally haven't met a professional that is perfect and knows everything, but I've met many that know how to get the job done despite any of their imperfections.

Now is that just me or...

I think you're taking all that a bit personally. I was actually asking a genuine question.

I'd say if you have the time to genuinely experiment on a session with this, AND bring things in on time, and keep the band happy...I take my hat off to you!

I just don't think I'd have the time or be able to justify to a band why I was spending time moving levels up and down (whilst of course, counterbalancing with the monitor fader to prevent any "louder is better" syndrome) to see if it made a difference. Converters have analogue electronics; it's perfectly conceivable that a "sweet spot" exists. I think you're totally right to abandon something if that "sweet spot" is small; I would hope however, that the difference between "sweet spot" and everywhere else in a modern converter is minimal, and my time is better spent everywhere else.

As I said - don't take it personally, I'm not trying to cut anyone down to size. I do feel that sometimes GS advice reads like it's parroting magazine articles, and there's a lack of genuine "session pressure" experience, but I'm not pointing fingers in any particular direction.

[Rev. Eslam]

Originally Posted by greenhirny

At any rate. I record professionally. So that argument is out the window. As you stated the 2192 has a much more pronounced sweet spot, and funny enough, the reason I got rid of mine was because I was worrying too much about that. Funny, I also was recording professionally at that time as well. In the end I decided that that was a waste of my time.
.

That must be the most bizarre reason for dispatching an audio interface ever.

Incidentally, welcome to Gearslutz greenhimy. You have certainly been very active in less than one month!
It took me seven years to reach as many posts as you have in a couple of weeks.

[living sounds]

Unlike in the Lynx 2 there are no OP275 on the inputs of the Aurora, they're only in the signal path of the output. Once replaced with better op amps the sound becomes indeed quite a bit more transparent.

The AD of the Aurora uses INA134 line receivers followed by OPA2227 and then MC33078 before the AD.

Originally Posted by EisenAudio

Assuming a standard +/-15V power supply, the OP-275 op amps comprising the analog section can accept inputs of up to 10.5Vp before exceeding specified acceptable distortion. This equates to 7.425Vrms or +19.63dBu. Ok, so that's lower than the anywhere from +20 to +28dBm max output that quality mic preamps are capable of producing, but Lynx have probably implemented a differential receiver circuit having -6dB loss. This means analog (theoretically) won't actually clip until +25.63dBu (a conservative +24 is probably closer to reality), and even if I'm wrong, we can still watch our levels and make do.

However, there is another limitation. I haven't double checked lately, but I recall the stock (sans variable trim) Aurora A/D is calibrated for -20dBFS = 0dBu, as are many converters. This means a +20dBu input yields 0dBFS, also known as the onset of digital clipping. So, even if the power supply is a few volts higher for a few more dBu in analog domain, the digital output will still clip a few dB sooner, and it all clips before the majority of transformer balanced analog outboard and consoles that people like to track through.

Adding insult to injury is leaving all your software faders at 0 before bouncing to disk. For instance, (and I realize this is not a realistic example) let's say you have 16 tracks of random noise each idling at -12. The summed result is about 0. In other words, loud tracks combine to form an even louder 2-mix which clips your busses.

Now, if you're overdriving transformer-balanced preamps (to get thicker tones), then you can see how it's pretty easy to clip the Aurora A/D. In my experience, this will still sound forgiving on a couple tracks out of several, i.e. your kick and snare hit the red once in a while. However, if you push all of your tracks, then like me you may begin to hear a low level upper midrange haze across the mix. Some of this is the software mixbuss and some of it's the Lynx. This is called distortion, and it makes a lot of sense when considering the numbers explained above.

To answer the original poster's question, I find Aurora 16 sounds best when externally clocked and trying to print tracks which live around -18 in Pro Tools, depending on track count.

[EisenAudio]

Originally Posted by living sounds

Unlike in the Lynx 2 there are no OP275 on the inputs of the Aurora, they're only in the signal path of the output. Once replaced with better op amps the sound becomes indeed quite a bit more transparent.

The AD of the Aurora uses INA134 line receivers followed by OPA2227 and then MC33078 before the AD.

Thank you very much for the clarification. INA134 do not have the -6dB negative gain I mentioned, but are still difficult to overload. Perhaps +27dBu @ +/-18V supply if I'm reading the datasheet correctly.

[Kingtone]

Originally Posted by living sounds

Unlike in the Lynx 2 there are no OP275 on the inputs of the Aurora, they're only in the signal path of the output. Once replaced with better op amps the sound becomes indeed quite a bit more transparent.

Have you replaced the op amps on an aurora?
Do you have some more specific details on this?

[living sounds]

Originally Posted by Kingtone

Have you replaced the op amps on an aurora?
Do you have some more specific details on this?

I tried quite a few in several positions and ended up with the AD8512 in place of the OP275. This decision was made by ear, there are other op amps that measured better, but the AD8512 in this place provided the most natural sound to my ears and the signal works better when mixed in the console.

[Kingtone]

Originally Posted by living sounds

I tried quite a few in several positions and ended up with the AD8512 in place of the OP275. This decision was made by ear, there are other op amps that measured better, but the AD8512 in this place provided the most natural sound to my ears and the signal works better when mixed in the console.

Interesting to hear.
How hard/simple was the swap out?

[greenhirny]

Originally Posted by Kingtone

I dunno,
but you definitely get the prize for the most instances of the word 'professional' (or a derivative) in a single gearslutz post!

I was doing that because PM was suggesting that folks aren't professional if they think there is a sweet spot in a converter. so I was letting him know that some of us do, but also agree with him that It might not be that important and that good gain staging is probably more important