Quantcast
SSL Alpha Link or Lynx Aurora D/A for Summing Box - Page 2 - Gearspace.com
The No.1 Website for Pro Audio
SSL Alpha Link or Lynx Aurora D/A for Summing Box
View Poll Results: Which D/As for summing box?
Lynx Aurora 16
39 Votes - 34.51%
SSL Alpha Link
74 Votes - 65.49%
Voters: 113. You may not vote on this poll

Old 12th November 2009 | Show parent
  #31
Gear Nut
 
🎧 15 years
Quote:
Originally Posted by feck ➡️
I just love it when strangers tell me how my ears work.
Haha...yeah, You don't know **** about music, do ya? ;-)
Good sounds on your page.
Old 12th November 2009 | Show parent
  #32
Jai guru deva om
 
warhead's Avatar
 
🎧 15 years
Quote:
Originally Posted by jamwerks ➡️
What plateform & DAW are you on?
Rain PC / Win XP / Nuendo 4

Quote:
Originally Posted by jamwerks ➡️
Wouldn't I want the SX to be able to connect AES/EBU (Bricasti, Eclipse, PCM 96)?
Yes, the SX is the AES/EBU version (AX which I own and use = ADAT for its digital connectivity).

Quote:
Originally Posted by jamwerks ➡️
Are the only two choices for mac/Logic the SSL Xtreme and the RME Madi?
Yes, the new MX4 card currently will route audio etc on a MAC but you won't have functions like the DSP mixer and effects. Best to stick with MADI Extreme 64 in my opinion, I just use Nuendo for routing etc.

Quote:
Originally Posted by jamwerks ➡️
Can you get comparable (or almost) latencies as with a Symphony system?
With my MADI Extreme 64 PCIe at 44.1k in my system I experience 2.04ms of latency round trip (sending a signal DA to AD and back into Nuendo). The MX4 claims .9ms of latency and also 128 channels, plus all the extensive routing / effects one could likely ever use in a complex setup.

I have to say I do not use Apogee or have a relationship with them, and have not directly compared myself or tested.

Hope this helps.

War

Solid State Logic | Music

Solid State Logic | Music
Old 12th November 2009 | Show parent
  #33
Lives for gear
 
jamwerks's Avatar
 
🎧 15 years
Quote:
Originally Posted by warhead ➡️
Hope this helps.
thumbsup
Old 12th November 2009 | Show parent
  #34
Gear Nut
 
🎧 15 years
Quote:
Originally Posted by warhead ➡️

With my MADI Extreme 64 PCIe at 44.1k in my system I experience 2.04ms of latency round trip (sending a signal DA to AD and back into Nuendo). The MX4 claims .9ms of latency and also 128 channels, plus all the extensive routing / effects one could likely ever use in a complex setup.
Hi.
Today I performed an extended latency test with the system I am currently trying out (Madi extreme-alpha link) and found some info which , to me at least , is WAY more interresting than the simple da to ad-test.

This is kinda messy but try to stay with me...;-)

I made an audio file with 4 short sine wave clicks and put that on track 1 in logic and assigned it to output 1. I connected utput 1 with input 9 and on track 2 in logic I recorded the signal coming back in from input 9.
Then I zoomed in on the 2 audiofiles to high resolution and saw that the "new" file on track 2 was delayed. I measured it to somewhere between 1.5 and 2 ms .

Now I extended it even further and found something strange which may explain why I am experiencing latency when I try to put down bass through this system...

I kept sending from output 1 to input nine on track 2 which was recording the incoming signal . THEN I assigned track 2 to output 3 and connected output 3 with input 11 and on track 3 I recorded the signal coming in from input 11.
In other words track 3 recorded the signal that track 2 put out WHILE recording from the original source(track 1 with the aif-file)(Tracks 2 and 3 were both recording at the same time and track 3 trecorded what was coming from "the playback head" from the ALSO recording track 2....!)

Now,still the delay between the original file and track 2 was still between 1.5ms and 2.0ms BUT the delay between track 2 and track 3 was about 5.2ms.....THIS made me think that doing the d/a-a/d test don't really tell the true story because the source is buffered and the computer don't have to compute it because it's already been done before pressing play BUT when it comes around the second time it's another story because then it has been re-routed inside logic, right?
SO...when the signal from my bass enters the a/d it may be quick as hell (like 0.9ms as it seems)BUT then logic has to process it and rout it to whatever output I have chosen and my conclusion( Please feel free to correct me. I am looking for a sollution to an unacceptable problem!) is that THERE's 3.x miliseconds added by logic while routing and then there's another 0.9 ms added by the d/a....sum it up and u get 5.2 ms....which is pretty much more than the madi extremes specs....
When I made the same test in Logic with my protools mix-rig I got the same result ,2.0 ms, no matter how many in-out-in-out's I routed and recorded..

Am I on to something here or am I just lost in latency hell? ;-)

I would SO like to hear if the same happens in every native system..
Old 12th November 2009 | Show parent
  #35
Lives for gear
 
feck's Avatar
 
🎧 10 years
Quote:
Originally Posted by figgebass ➡️
Good sounds on your page.
Thanks, man.
Old 13th November 2009 | Show parent
  #36
Lives for gear
 
🎧 15 years
Quote:
Originally Posted by figgebass ➡️

Am I on to something here or am I just lost in latency hell? ;-)

I would SO like to hear if the same happens in every native system..
I had a similar experience once while helping a friend record some demos into Ableton Live with some firewire interface (can't remember which). Everything was placed late in the timeline compared to input by a factor of the buffer (tested it with a click) which made overdubbing impossible. Curious, I tried the same experiment with an MBox and Protools LE I had laying around, and the clicks lined up EXACTLY when re-recording.

You'd think in this day and age of DAW development that placing performances accurately in the timeline would be a given - be it audio or midi. What a slippery mess we're still sloshing around in...
Old 13th November 2009 | Show parent
  #37
Gear Nut
 
🎧 15 years
Quote:
Originally Posted by juniorhifikit ➡️
I had a similar experience once while helping a friend record some demos into Ableton Live with some firewire interface (can't remember which). Everything was placed late in the timeline compared to input by a factor of the buffer (tested it with a click) which made overdubbing impossible. Curious, I tried the same experiment with an MBox and Protools LE I had laying around, and the clicks lined up EXACTLY when re-recording.

You'd think in this day and age of DAW development that placing performances accurately in the timeline would be a given - be it audio or midi. What a slippery mess we're still sloshing around in...
yeah, and I am more and more leaning towards going protools HD to avoid the latency that's applied by logic....=/ Expensive sollution just to be able to record without problems...
PISSES ME OFF BIG TIME cos the SSL sound GREAT.....and the SSL madi card and the d/a is quick, no prob, but what logic does there in the middle....
Old 13th November 2009 | Show parent
  #38
Lives for gear
 
jamwerks's Avatar
 
🎧 15 years
Quote:
Originally Posted by figgebass ➡️
yeah, and I am more and more leaning towards going protools HD to avoid the latency that's applied by logic....=/ Expensive sollution just to be able to record without problems...
I'm quite sure that your problem is just one of the parameter settings. You might want to start a new thread here at GS or maybe on a User Group and I'm sure you could get the help that you need.

Native systems now have latencies just as low (or lower) that HD systems. There are tons of thread here where lots of issues are discussed.
Old 13th November 2009 | Show parent
  #39
Gear Nut
 
🎧 15 years
Quote:
Originally Posted by jamwerks ➡️
I'm quite sure that your problem is just one of the parameter settings. You might want to start a new thread here at GS or maybe on a User Group and I'm sure you could get the help that you need.

Native systems now have latencies just as low (or lower) that HD systems. There are tons of thread here where lots of issues are discussed.
I wish you were right but I seriously doubt it cos now I have run the test on 3 different setups and they all told the same story: There is a serious latency happening inside logic while routing incoming signal back out again....The Ssl beat the **** out of the RME and the Motu setups although they were all performed on different specs computers so it doesn't necessarily mean that the last 2 are slower on the a/d-d/a.

When I play a audiofile out through one output and into another while recording they all (all 3 systems)give a latency of slightly less than 2 ms BUT when I record something that the computer haven't buffered I get 5.2 ms latency on the quickest of the systems. What I did was play a file uot and back in on track 2 which was recording AND routing out again and back in on track 3 and the difference between track 2 and 3 was 5.2 ms. (On the slowest system it was 10 ms plus....)

My conclusion is that when you press play to play an already recorded file ,before there is any sound coming out ,the computer already has calculated what it's going to process and hence there's no latency applied by the outgoing routing .
When the system records and then plays back it hasnt had any chance to calculate in advance what it's going to put out and where but has to calculate in realtime and that seems to take about 3.4 ms in the fastest computer of this test (an 8 core 2.26 nehalem).

When you are doing the same thing but with my protools-logic-rig there is 2 ms latency between the original and track 2 and 2 ms latency between track 2 and 3 etc etc forever. This must be down to the signal never actually being calculated by the computer because it never leaves the dsp's on the hd-card.

Please , anyone who has 3 minutes over, try this test. Make an aif-file with a click , put it on track 1, send it out and back in to track 2 , send that out again and into track 3 and press rec on tracks 2 and 3. I am pretty sure you'll get a small latency between track 1 and 2 and then a worse latency between track 2 and 3.

Or if someone knows which box to check , pls let me in on the secret..;-)
Old 13th November 2009 | Show parent
  #40
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by feck ➡️
Nope. The way the transients and stereo imaging add up to a combined improvement is noticeable enough that even my clients noticed a difference between the new mixes (using the Alpha Link instead of the RME with the same exact settings) and the previous ones without me telling them about the new converters. I just love it when strangers tell me how my ears work.
Do you mean the same mix done with both RME and SSL converters and same outboard settings or just different mixes at all ? If you're talking about different material then the comparison can't be fair, too many variables and most likely the converters are the last thing going to make a difference, your skills and approach probably the first.
Old 14th November 2009 | Show parent
  #41
Lives for gear
 
feck's Avatar
 
🎧 10 years
Quote:
Originally Posted by Ciozzi ➡️
Do you mean the same mix done with both RME and SSL converters and same outboard settings or just different mixes at all ? If you're talking about different material then the comparison can't be fair, too many variables and most likely the converters are the last thing going to make a difference, your skills and approach probably the first.
The same mixes, subbed out through 8 stems of SSL instead of 8 stems of RME. The only difference being the converters. Believe me, I was skeptical of the level of improvement it would/might add. I am no longer skeptical.
Old 14th November 2009 | Show parent
  #42
Lives for gear
 
jamwerks's Avatar
 
🎧 15 years
Quote:
Originally Posted by figgebass ➡️
Or if someone knows which box to check , pls let me in on the secret..;-)
Post your question on the Logic user group and you'll get all the help that need.
Old 14th November 2009 | Show parent
  #43
pan
Lives for gear
 
pan's Avatar
 
🎧 15 years
Quote:
Originally Posted by figgebass ➡️
Or if someone knows which box to check , pls let me in on the secret..;-)
To compensate for input latency you have to adjust the Recording Delay in Audio Preferences. This will take care of the delay on Track 2 in your test-setting. Now you can accurately measure the ad-da roundtrip latency.
Attached Thumbnails
SSL Alpha Link or Lynx Aurora D/A for Summing Box-screenshot-1.jpg  
Old 14th November 2009 | Show parent
  #44
Gear Nut
 
🎧 15 years
Quote:
Originally Posted by pan ➡️
To compensate for input latency you have to adjust the Recording Delay in Audio Preferences. This will take care of the delay on Track 2 in your test-setting. Now you can accurately measure the ad-da roundtrip latency.
Ok, but all that will do is put the file in the place it should be, right , and not in any way take care of the sound that I am hearing when playing bass through the computer,which is what the problem is about, or am I wrong?
Old 14th November 2009 | Show parent
  #45
pan
Lives for gear
 
pan's Avatar
 
🎧 15 years
Quote:
Originally Posted by figgebass ➡️
Ok, but all that will do is put the file in the place it should be, right , and not in any way take care of the sound that I am hearing when playing bass through the computer,which is what the problem is about, or am I wrong?
Yes, it can not reduce the LATENCY.
Old 14th November 2009 | Show parent
  #46
Gear Nut
 
🎧 15 years
Quote:
Originally Posted by pan ➡️
Yes, it can not reduce the LATENCY.
Exactly...and the latency is the problem. Sorry if I didn't make myself clear. (Dang english language!! ;-))
I can not play my best when my brain says "HEY?!? What's THIS?! What I hear is behind the beat! Better speed up! OOps...now I am ahead! Back off! Oops...behind again"....;-)
When what my fingers do don't match the signals my ears send to my brain then I have a problem....
"Digidesign, here I come...."...I'm afraid...
Old 16th November 2009 | Show parent
  #47
pan
Lives for gear
 
pan's Avatar
 
🎧 15 years
Quote:
I am working in logic 8 in leopard. Is there nothing native that comes down to the , in my opinion, non noticable latency of my OLD protools rig?
unfortunately - NO


Old 16th November 2009 | Show parent
  #48
Gear Addict
 
bassman's Avatar
 
🎧 15 years
Quote:
Originally Posted by feck ➡️
The same mixes, subbed out through 8 stems of SSL instead of 8 stems of RME. The only difference being the converters. Believe me, I was skeptical of the level of improvement it would/might add. I am no longer skeptical.
Hold on a sec..... You cannot do a direct swap. The converters have differing output levels which will throw the test off. Here are the hard specs:

• SSL output level: +22dBu max

• RME Hi Gain: +19dBu (this is confusing as they list the +4dBu standard as peaking at +13 dBu suggesting they are using -12dBFS [or even less??] as a ref. go figger....)

• Lynx Aurora: +20dBu max

With this in mind and no calibration between tests, the SSL will win hands downs because it will be louder. Run tones. Check with meters, including SPL meter so you don't get hosed in the test.....

That being said, I tested with calibration and still liked the SSL over the Lynx. But the diff is subtle. Low end is definitely clearer and defined better on the SSL. Upper mids are more forward on the SSL, with the Lynx being "sweeter." However, I hear more detail on the SSL in that range as well. The lynx was smoothing over some stuff.

This is just DA, playback of sources only. YMMV

Also, watch out for ground loops between the Alphalink and the MADI card via wordclock as this will foul the wordclock connection. Happened to me and took a bit of suss'ing to figure it out.

-bassman
Old 16th November 2009 | Show parent
  #49
Lives for gear
 
DaVogi's Avatar
 
🎧 10 years
calibration is one thing but you just posted the different headroom-specs of the converters.... they have nothing to do with the actual output levels.
Old 16th November 2009 | Show parent
  #50
Lives for gear
 
🎧 15 years
Quote:
Originally Posted by bassman ➡️
Hold on a sec..... You cannot do a direct swap. The converters have differing output levels which will throw the test off. Here are the hard specs:

• SSL output level: +22dBu max

• RME Hi Gain: +19dBu (this is confusing as they list the +4dBu standard as peaking at +13 dBu suggesting they are using -12dBFS [or even less??] as a ref. go figger....)

• Lynx Aurora: +20dBu max

With this in mind and no calibration between tests, the SSL will win hands downs because it will be louder. Run tones. Check with meters, including SPL meter so you don't get hosed in the test.....

That being said, I tested with calibration and still liked the SSL over the Lynx. But the diff is subtle. Low end is definitely clearer and defined better on the SSL. Upper mids are more forward on the SSL, with the Lynx being "sweeter." However, I hear more detail on the SSL in that range as well. The lynx was smoothing over some stuff.

This is just DA, playback of sources only. YMMV

Also, watch out for ground loops between the Alphalink and the MADI card via wordclock as this will foul the wordclock connection. Happened to me and took a bit of suss'ing to figure it out.

-bassman
Did you compare the SSL to the Aurora or your Lynx Two?
Old 17th November 2009 | Show parent
  #51
btw, does someone know if we can calibrate the output of the Aurora to +22dB?

cheers
G

Quote:
Originally Posted by bassman ➡️
Hold on a sec..... You cannot do a direct swap. The converters have differing output levels which will throw the test off. Here are the hard specs:

• SSL output level: +22dBu max

• RME Hi Gain: +19dBu (this is confusing as they list the +4dBu standard as peaking at +13 dBu suggesting they are using -12dBFS [or even less??] as a ref. go figger....)

• Lynx Aurora: +20dBu max

With this in mind and no calibration between tests, the SSL will win hands downs because it will be louder. Run tones. Check with meters, including SPL meter so you don't get hosed in the test.....

That being said, I tested with calibration and still liked the SSL over the Lynx. But the diff is subtle. Low end is definitely clearer and defined better on the SSL. Upper mids are more forward on the SSL, with the Lynx being "sweeter." However, I hear more detail on the SSL in that range as well. The lynx was smoothing over some stuff.

This is just DA, playback of sources only. YMMV

Also, watch out for ground loops between the Alphalink and the MADI card via wordclock as this will foul the wordclock connection. Happened to me and took a bit of suss'ing to figure it out.

-bassman
Old 17th November 2009 | Show parent
  #52
Lives for gear
 
feck's Avatar
 
🎧 10 years
Quote:
Originally Posted by bassman ➡️
Hold on a sec..... You cannot do a direct swap. The converters have differing output levels which will throw the test off. Here are the hard specs:

• SSL output level: +22dBu max

• RME Hi Gain: +19dBu (this is confusing as they list the +4dBu standard as peaking at +13 dBu suggesting they are using -12dBFS [or even less??] as a ref. go figger....)

• Lynx Aurora: +20dBu max

With this in mind and no calibration between tests, the SSL will win hands downs because it will be louder. Run tones. Check with meters, including SPL meter so you don't get hosed in the test.....

That being said, I tested with calibration and still liked the SSL over the Lynx. But the diff is subtle. Low end is definitely clearer and defined better on the SSL. Upper mids are more forward on the SSL, with the Lynx being "sweeter." However, I hear more detail on the SSL in that range as well. The lynx was smoothing over some stuff.

This is just DA, playback of sources only. YMMV

Also, watch out for ground loops between the Alphalink and the MADI card via wordclock as this will foul the wordclock connection. Happened to me and took a bit of suss'ing to figure it out.

-bassman
Even so, I am talking about a fully stemmed mix with all ITB processing bounced to a stereo mix, and then run through an L2 with 3 db max gain reduction. There is no difference I can foresee even if there is a difference in the output level, as all of the audio tracks are still the same and processing remained the same as well.
Old 17th November 2009 | Show parent
  #53
Gear Addict
 
bassman's Avatar
 
🎧 15 years
I don't think there's a clear understanding of what the max output level means. It has nothing whatsoever to do with headroom.

Headroom is the amount of level before distortion that exists above a reference level. It depends on the ref level. For professional use, digital uses 1kHz @ -20dBFS as a ref level, leaving 20dB of "headroom" above it before digital clipping. If you use -12dBFS as a ref, then you only have 12dB of headroom above that before digital clipping. Remember, decibels formats are relative!

So........

DAC's have their own maximum output level measured in dBu, which is a voltage level, not dBFS, which is a digital audio level (bits). When it says +22dBu max, that means a 0dBFS 1kHz signal coming out of the SSL will measure +22dBu (voltage). Coming out of the Lynx Aurora, it will only be +20dBu, less voltage = lower ANALOG level going into the summer that feck is using.

feck, what setup are you using? Did you run levels through your summer and calibrate the final gain stage for equal levels? Just a 1k tone will do out of one of your stem outputs.

And, yes, I was comparing the SSL to my LynxTWO in the test, however, I do have an Aurora here that I have heard as well. The LynxTWO also has a max output of +20dBu. Just for comparison, the Dangerous mastering console has a max output of +27dBu, that's professional output level, 23dB of headroom above nominal +4dBu levels, nice....

-bassman

Wiki on dB, read way down on electrical measurements.
Old 17th November 2009 | Show parent
  #54
Lives for gear
 
feck's Avatar
 
🎧 10 years
Quote:
Originally Posted by bassman ➡️
I don't think there's a clear understanding of what the max output level means. It has nothing whatsoever to do with headroom.

Headroom is the amount of level before distortion that exists above a reference level. It depends on the ref level. For professional use, digital uses 1kHz @ -20dBFS as a ref level, leaving 20dB of "headroom" above it before digital clipping. If you use -12dBFS as a ref, then you only have 12dB of headroom above that before digital clipping. Remember, decibels formats are relative!

So........

DAC's have their own maximum output level measured in dBu, which is a voltage level, not dBFS, which is a digital audio level (bits). When it says +22dBu max, that means a 0dBFS 1kHz signal coming out of the SSL will measure +22dBu (voltage). Coming out of the Lynx Aurora, it will only be +20dBu, less voltage = lower ANALOG level going into the summer that feck is using.

feck, what setup are you using? Did you run levels through your summer and calibrate the final gain stage for equal levels? Just a 1k tone will do out of one of your stem outputs.

And, yes, I was comparing the SSL to my LynxTWO in the test, however, I do have an Aurora here that I have heard as well. The LynxTWO also has a max output of +20dBu. Just for comparison, the Dangerous mastering console has a max output of +27dBu, that's professional output level, 23dB of headroom above nominal +4dBu levels, nice....

-bassman

Wiki on dB, read way down on electrical measurements.
Hey Bassman. Thanks for the info. The setup is (was) an RME HDSPe and RME 8 I/O running into a Neve 8816. Then I subbed out the HDSPe for a RAYDAT and the SSL Alpha Link and am using the ADAT connections now. I pulled up a few of the mixes I had previously done on the RME converters and just ran the same mixes through the Alpha Link. The transients and stereo imaging overall are much more articulate than they were with the RME (and the RME 8 I/O was a Nuendo 8 I/O which apparently has an even older converter build, inferior to the newer RME stuff).
Old 17th November 2009 | Show parent
  #55
Gear Addict
 
bassman's Avatar
 
🎧 15 years
Quote:
Originally Posted by feck ➡️
The transients and stereo imaging overall are much more articulate than they were with the RME
My experience as well. They're good converters plus a ton of IO with the MADI card...

Setting up a mix with 22ch stems as we type....

-bassman
Old 17th November 2009 | Show parent
  #56
Gear Nut
 
🎧 15 years
Today I took away the ssl-system I had been working on for little less than a week and installed a hd3 with an 192....It was NOT an improvement in the treble-area that's for sure.....!!!
Old 18th November 2009 | Show parent
  #57
Lives for gear
 
feck's Avatar
 
🎧 10 years
Quote:
Originally Posted by bassman ➡️
My experience as well. They're good converters plus a ton of IO with the MADI card...

Setting up a mix with 22ch stems as we type....

-bassman
Nice. By the way, you have a very sweet studio going there!
Old 18th November 2009 | Show parent
  #58
Gear Addict
 
bassman's Avatar
 
🎧 15 years
Quote:
Originally Posted by feck ➡️
Nice. By the way, you have a very sweet studio going there!
Why thankee much! Nice work you have on your site as well.....

-bassman
Old 18th November 2009 | Show parent
  #59
Lives for gear
 
DONNX's Avatar
 
🎧 10 years
Out of the two..I voted for the SSL. And SSL is usually humble about its stats...meaning I believe there +22DBU is somewhat more accurate that the others...

Two to consider are the Mytek 8x192. The headroom on that is huge!
And the apogee DA16x. +26dbu. with a soft limit feature (nice pro level)
Old 18th November 2009 | Show parent
  #60
Quote:
Originally Posted by George Necola ➡️
btw, does someone know if we can calibrate the output of the Aurora to +22dB?

cheers
G
check..
📝 Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 55 views: 32334
Avatar for IM WHO YOU THINK
IM WHO YOU THINK 13th October 2020
replies: 15929 views: 1533275
Avatar for Ragan
Ragan 11th January 2019
replies: 1296 views: 181792
Avatar for heraldo_jones
heraldo_jones 1st February 2016
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump