The No.1 Website for Pro Audio
Evaluating AD/DA loops by means of Audio Diffmaker
Old 8th October 2020 | Show parent
  #2101
Gear Maniac
 
Dan O's Avatar
 

Quote:
Originally Posted by didier.brest View Post
It seems OK.

-The uploaded recorded file is a little bit longer than the original. Can you fix that?

It does not matter.

- In logic, when bouncing, I chose split and not interleaved, correct?
Yes. Both should be OK. Your file seems to be interleaved. Split wav stereo is composed of two files: one for the left channel and one for the right channel.
Ah Ok I Think it generated both split and interleaved then. I got two additional files L and R.


Not so good results though. I'm gonna do some listening comparisons later as well. It's funny because we did som blindtests earlier with your files and compared some MOTU, antelope and lynx aurora files. The lynx which have the best result sounded good but was also the easiest to separate from the others and actually sounded the least similiar to the original file.
Show replies
Old 8th October 2020 | Show parent
  #2102
Quote:
Originally Posted by Dan O View Post
Ah Ok I Think it generated both split and interleaved then. I got two additional files L and R.


Not so good results though. I'm gonna do some listening comparisons later as well. It's funny because we did som blindtests earlier with your files and compared some MOTU, antelope and lynx aurora files. The lynx which have the best result sounded good but was also the easiest to separate from the others and actually sounded the least similiar to the original file.
Out of interest which one sounded the closest then?
Show replies
Old 9th October 2020 | Show parent
  #2103
Quote:
Originally Posted by Dan O View Post
It's funny because we did som blindtests earlier with your files and compared some MOTU, antelope and lynx aurora files.
Which files did you listen to ?

Did you apply the corrective gains ?

If one listens to the raw files dowloaded from the list of the results

Lynx Aurora(n), 1.1 dB (L), 1.1 dB (R), -58.9 dBFS (L), -59.0 dBFS (R) (test recorded by Mikathechou)

Antelope Pure2, -0.0 dB (L), -0.2 dB (R), -56.8 dBFS (L), -58.2 dBFS (R) (test recorded by Touf)

MOTU 2408mk3, 0.1 dB (L), 0.1 dB (R), -56,7 dBFS (L), -58,1 dBFS (R) (test recorded by ram 75)

without applying the indicated corrective gains, one may find that

Quote:
Originally Posted by Dan O View Post
The lynx which have the best result sounded good but was also the easiest to separate from the others and actually sounded the least similiar to the original file.
Show replies

Last edited by didier.brest; 9th October 2020 at 04:57 AM..
Old 9th October 2020 | Show parent
  #2104
Gear Maniac
 
Dan O's Avatar
 

Quote:
Originally Posted by didier.brest View Post
Which files did you listen to ?

Did you apply the corrective gains ?

If one listens to the raw files dowloaded from the list of the results

Lynx Aurora(n), 1.1 dB (L), 1.1 dB (R), -58.9 dBFS (L), -59.0 dBFS (R) (test recorded by Mikathechou)

Antelope Pure2, -0.0 dB (L), -0.2 dB (R), -56.8 dBFS (L), -58.2 dBFS (R) (test recorded by Touf)

MOTU 2408mk3, 0.1 dB (L), 0.1 dB (R), -56,7 dBFS (L), -58,1 dBFS (R) (test recorded by ram 75)

without applying the indicated corrective gains, one may find that
Yes, we downloaded the files from the list and level matched. I didn't realize though that the right dB correction for level matching was included in the list but I've checked with the levelmatching we did on our own and it seems to be spot on. Maybe I'll give it another listen when I find the time. The interfaces we were interested in was lynx aurora n, MOTU 16a, MOTU 24 i/o, orion studio and just for interest antelope pure 2. What we were missing was orion32 gen3 which we later ended up buying.
Show replies
Old 10th October 2020
  #2105
Gear Maniac
 
Dan O's Avatar
 

Ok, so here it goes. New loopbacks of the Orion 32 gen3. The results will probably be more or less the same but just for the sake of it this is the best it's gonna get. ( I'm not going to solder a d-sub to trs connectors for the monitor loopback. )

The first version is a straight connection (mogami cables if anyone cares) from line out 1-2 to line in 1-2:

http://www.mediafire.com/file/sjkhbl...ction.wav/file

and secondly. From Monitor out 1-2 via console and EDAC patchbay to Line in 1-2:

http://www.mediafire.com/file/kgbzu6...chbay.wav/file



( Edit. I just realized that the monitor out trim was set to 20dbu and the line in and out to 24dbu but I guess that shouldn't really matter. )

( 2nd Edit. I just realized that on the monitor out loopback the R signal, my adapter cable had been phase flipped. Didier, can you work around that or do I need to redo it? )
Show replies

Last edited by Dan O; 10th October 2020 at 07:08 PM..
Old 10th October 2020 | Show parent
  #2106
Gear Nut
 

Quote:
Originally Posted by Dan O View Post
Ok, so here it goes. New loopbacks of the Orion 32 gen3. The results will probably be more or less the same but just for the sake of it this is the best it's gonna get. ( I'm not going to solder a d-sub to trs connectors for the monitor loopback. )

The first version is a straight connection (mogami cables if anyone cares) from line out 1-2 to line in 1-2:

...
according to my experience with this kind of loopback test, using a short dedicated cable (0.5m) or a patch bay with a cheap (the cheapest on Thomann) patch cable connecting two d-sub looms 1.5m each makes barely no difference.
Show replies
Old 10th October 2020 | Show parent
  #2107
Quote:
Originally Posted by Dan O View Post
The first version is a straight connection (mogami cables if anyone cares) from line out 1-2 to line in 1-2
1.030648 ms, 0.3835 dB (L), 0.3284 dB (R), -44.1181 dBFS (L), -45.1804 dBFS (R)

Quote:
Originally Posted by Dan O View Post
I just realized that on the monitor out loopback the R signal, my adapter cable had been phase flipped.
No matter. I'll measure your second file tomorrow.
Show replies
Old 10th October 2020 | Show parent
  #2108
Gear Maniac
 
Dan O's Avatar
 

Quote:
Originally Posted by didier.brest View Post


No matter. I'll measure your second file tomorrow.
Ok ๐Ÿ‘Œ Thanks!
Show replies
Old 11th October 2020 | Show parent
  #2109
Quote:
Originally Posted by Dan O View Post
From Monitor out 1-2 via console and EDAC patchbay to Line in 1-2
399 ns, 0.2046 dB (L), 0.2382 dB (R), -48.8590 dBFS (L), -49.8813 dBFS (R)

To be added to the next issue of the list of the results.
Show replies
Old 11th October 2020 | Show parent
  #2110
Gear Maniac
 
Dan O's Avatar
 

Quote:
Originally Posted by didier.brest View Post
399 ns, 0.2046 dB (L), 0.2382 dB (R), -48.8590 dBFS (L), -49.8813 dBFS (R)

To be added to the next issue of the list of the results.

Ok, interesting. A bit better results. I think I've percieved the monitor outs to sound a little better than the line outs but I'm gonna take a second listen.

Still not so impressive results though. If one were to go only by these results, two Motu 16A would have been more for the money. I think I need to redo som listening tests now that we also got the files from the antelope.

Btw. I wonder what the real difference is between the line outs and the monitor outs. If i assign 31-32 to the monitor outs I can still get the same signal from the line out 31-32. I'm guessing then that it's the same converterchip but a different analog output stage.
Show replies

Last edited by Dan O; 11th October 2020 at 02:51 PM..
Old 11th October 2020 | Show parent
  #2111
Quote:
Originally Posted by Dan O View Post
I wonder what the real difference is between the line outs and the monitor outs.
The test result of the Orion Rev. 2017 is also better on the monitor outs than on the line outs (see the list of the results). The reason might be that the monitor outs of the Orion Rev. 2017 are unbalanced (see p.11 of this user manual). But the monitors outs of the Orion32+ are balanced (see ยง5 point 18 of this user manual). Did you use TRS or TS cables ?
Show replies
Old 11th October 2020 | Show parent
  #2112
Gear Maniac
 
Dan O's Avatar
 

Quote:
Originally Posted by didier.brest View Post
The test result of the Orion Rev. 2017 is also better on the monitor outs than on the line outs (see the list of the results). The reason might be that the monitor outs of the Orion Rev. 2017 are unbalanced (see p.11 of this user manual). But the monitors outs of the Orion32+ are balanced (see ยง5 point 18 of this user manual). Did you use TRS or TS cables ?
Ok, interesting. I used TRS connections.
Show replies
Old 4 weeks ago | Show parent
  #2113
Lives for gear
 

I've been doing a lot of A/B tests with the samples in this thread as I am looking for a new interface. Great stuff and very helpful!

One question I have, though (perhaps it has been answered already), is do we know what AD converter was used to create the original sample? It sounds quite good.
Show replies
Old 3 weeks ago | Show parent
  #2114
Quote:
Originally Posted by shoepedals View Post
do we know what AD converter was used to create the original sample?
I don't.

Test of the Tascam US-122L by KindYak (send by PM):

-1.569786603 s, 12.5437 dB (L), 12.8052 dB (R), -41.3201 dBFS (L), -42.7394 dBFS (R)

To be added to the next issue of the list of the results.
Show replies
Old 2 weeks ago | Show parent
  #2115
Gear Maniac
 

Quote:
Originally Posted by Dan O View Post
Ah Ok I Think it generated both split and interleaved then. I got two additional files L and R.


Not so good results though. I'm gonna do some listening comparisons later as well. It's funny because we did som blindtests earlier with your files and compared some MOTU, antelope and lynx aurora files. The lynx which have the best result sounded good but was also the easiest to separate from the others and actually sounded the least similiar to the original file.
I mean, I don't agree, I've used Apogee, RME and Lynx in the past and at some point got the Cranesong Solaris which sounded amazing at first listen, like it sounded open, 3d and like very exciting but then I was like wait, is it actually a clear representation or is it sweetening the sound and sure enough, looping back the Solaris through several devices (RME, Lynx and Apogee) gave me that hyped sound, which I did Love from an experience point but it wasn't an honest representation of what the signal actually sounds like and this upset me by a lot because it's sole purpose is monitoring.

Actually to me it had this smiley face EQ curve sound with a feel of transient designer response, like things were popping out unrealistically and when you actually printed it back through another device and phased both signals out, you could hear that fake transient emphasis that was left over of kicks, snares and very transient heavy things.

The smiley curve I figured out actually came from the device phasing out in the mid range very well but becoming increasingly less precise to the sides of the frequency spectrum so actually having owned the Cranesong Solaris made me google converter shootout and find this thread.

I also tried the Merging Anubis because some people said it sounded very much like the Solaris and sure enough, it had a very similar sonic effect going through probably this internal up and downsampling for better transient response which to my ears sounded more fake than genuine.
Show replies
Old 2 weeks ago | Show parent
  #2116
Gear Maniac
 

Quote:
Originally Posted by didier.brest View Post
-1.337595 ms, -0.0071 dB (L), -0.0554 dB (R), -77.9426 dBFS (L), -77.9483 dBFS (R)

To be added to the next issue of the list of the results.


Fine.
That's very odd, is it possibly up and down sampling internally since it is a dsp based fx unit? Because all the other converters at higher sample rates will create a more precise translation as well which will possibly be more accurate when it comes back through the same device downsampled with the same filtering again?
Show replies
Old 2 weeks ago | Show parent
  #2117
Quote:
Originally Posted by spankjam View Post
That's very odd, is it possibly up and down sampling internally since it is a dsp based fx unit?
1) Not so odd: there are better results involving a Focusrite or a MSB AD converter in the list of the results.

2) The test rule is that the DA converter must be fed with the 44.1 kHz playback of the original wav file and the AD converter must deliver the 44.1 kHz digital audio signal recorded as the loopback wav file. Internal sampling rate conversion is OK. I don't know whether it is used by the Eventide H9000R.
Show replies
Old 2 weeks ago | Show parent
  #2118
Gear Maniac
 

Quote:
Originally Posted by didier.brest View Post
1) Not so odd: there are better results involving a Focusrite or a MSB AD converter in the list of the results.

2) The test rule is that the DA converter must be fed with the 44.1 kHz playback of the original wav file and the AD converter must deliver the 44.1 kHz digital audio signal recorded as the loopback wav file. Internal sampling rate conversion is OK. I don't know whether it is used by the Eventide H9000R.
Yeah I just mean when you go with higher sample rates, the phase out will be better and some devices could do it internally.

Would you be able to do a comparison with 192k and calculate the difference?
Show replies
Old 2 weeks ago | Show parent
  #2119
Quote:
Originally Posted by spankjam View Post
Would you be able to do a comparison with 192k and calculate the difference?
Yes after downsampling from 192 to 44.1.
Show replies
๐Ÿ“ Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 183 views: 25711
Avatar for baikonour
baikonour 10th October 2006
replies: 360 views: 93292
Avatar for symphonicmind
symphonicmind 4 weeks ago
replies: 4582 views: 1202955
Avatar for amberience
amberience 1 minute ago
replies: 61 views: 9836
Avatar for Transistor
Transistor 21st February 2015
Post Reply

Welcome to the Gearslutz 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…
๐Ÿ–จ๏ธ Show Printable Version
โœ‰๏ธ Email this Page
๐Ÿ” Search thread
๐ŸŽ™๏ธ View mentioned gear
Forum Jump
Forum Jump