|
|
Post by scumbum on Nov 14, 2014 12:35:10 GMT -6
Hey guys I was wondering if I could get some help . I did a test to compare Pro Tools LE 7.4 vs a Dedicated Recorder ( Black Face ADAT ) and my findings is that the dedicated recorder (BF ADAT) sounds better than Pro Tools ! Ok stop rolling your eyes ....... I actually don't want this to be true , but my tests reveal that it is . This is where I need your help , PROVE ME WRONG , Prove that my test is flawed . In all honesty I want to keep using Pro Tools LE because its VERY easy to use to record/edit , but if my findings are true I'll be moving to a dedicated recorder like an Alesis HD24 . So find the flaw in my test so I can keep using PT LE ! Over the years I've read on audio forums how people have had dedicated recorders then switched to Pro Tools and their music suffered . The responses to them were always "Your Crazy !! , Its all 1's & 0's !! , Digital is digital !!" .......So in the back of my mind I was 99% sure they were right and there is no difference between Pro Tools or a dedicated recorder . But that 1% uncertainty was always in the back of my mind so I finally did a test to see who was right . Heres my test : Step : 1. Record 6 tracks of audio to my Blackface ADAT tape ( Drums , Bass , Guitar ) 2. Play back ADAT Tape - mix audio through a Mackie mixer - stereo out of Mackie mixer into an Mbox connected to a "Laptop" running PT LE 7.4 - record stereo track of mix into a 24bit 48k session . Ok I now have a mix down of the ADAT tape recorder . 3. Connect Black face ADAT to a 002r for a digital transfer ( optical cables , In & Out ) the 002r is running on a "desktop" computer with PT LE 7.4 . 4. Set the Black Face ADAT's clock to internal , set the 002r master clock to Optical (ADAT) . 5. Create a Session ( I tried both a 24bit 48K and a 16bit 48K session with the same results ) create 6 audio tracks and record the ADAT tape into pro tools through the optical connection . 6. With the BF ADAT still set as Master clock , playback recorded audio in Pro Tools through the same BF D/A converters , through the same cables and same Mackie Mixer ( Did not change a thing , same cables , same mixer settings) into the Mbox and Laptop session. The end results is the two mixes don't null , theres a good deal of audio loss between the two . The settings are all the same , same D/A converters , same Master clock , same cables , same mixer settings......only difference is playback from BF ADAT vs PT LE . The only thing I can think of is the audio is degraded some how in the optical transfer ?? BUT!!!!!! The other day before I did this test I recorded a mix from PT to the ADAT tape and then back to PT through the optical connection to see if there was any audio loss and that nulled in PT............ PROVE MY TEST FLAWED!! I want to keep using PT LE . Heres the two versions , the adat playback the other is the Pro Tools play back . Listen for yourself . www.dropbox.com/sh/apyln9e680xvb88/AAA8968fdaKdSQuRs9rL7liCa?dl=0 |
|
|
|
Post by popmann on Nov 14, 2014 13:20:48 GMT -6
My only suggestion would be there are two things unaccounted for....one you can actually check--which is to print six test tone on the end of all the track--one after another--just make sure the output of the DAC is actually teh same coming from PT, because hitting an analog board a few DB differently will certainly affect the difference.
Then....the other IS what you're getting at....which is that while an ADAT (or most hardware multitracks WITH NO MIXING FUNCTION) the process is ADC in....file creation....DAC output--all at 16 (in this case) or 24bit depth. Where ProTools (I'm assuming software) will take the ADC (or in this case digital in) in, create a file, inject it into a 32bit float mixer....often to an output buss (?)....and provide that stream to a hardware mixer (in the interface which may be floating or fixed point) which then sends that to the 16bit DAC. There's no doubt that in that there's potential for loss. How significant it is....I suppose becomes a matter of subjective opinion....but, the idea that it passes all that with default settings untouched/nullable, is IMO delusional thinking.
|
|
|
|
Post by scumbum on Nov 14, 2014 13:49:42 GMT -6
My only suggestion would be there are two things unaccounted for....one you can actually check--which is to print six test tone on the end of all the track--one after another--just make sure the output of the DAC is actually teh same coming from PT, because hitting an analog board a few DB differently will certainly affect the difference. Then....the other IS what you're getting at....which is that while an ADAT (or most hardware multitracks WITH NO MIXING FUNCTION) the process is ADC in....file creation....DAC output--all at 16 (in this case) or 24bit depth. Where ProTools (I'm assuming software) will take the ADC (or in this case digital in) in, create a file, inject it into a 32bit float mixer....often to an output buss (?)....and provide that stream to a hardware mixer (in the interface which may be floating or fixed point) which then sends that to the 16bit DAC. There's no doubt that in that there's potential for loss. How significant it is....I suppose becomes a matter of subjective opinion....but, the idea that it passes all that with default settings untouched/nullable, is IMO delusional thinking. So how do I do the test tone , use the signal generator plugin in pro tools , then record through optical onto the ADAT tape tracks ? What test tone do I use , frequency , wave form ?? I'm not a technical guy by any means , but I do know what I hear and its the sound of Pro Tools . I've heard this sound since my friend got a TDM rig back in the mid 90's , kinda flat , undefined low end . Maybe HD is different . I do have reaper and I'm gonna try the same test to see how reaper does . I'd like to find a DAW if possible that doesn't degrade the audio just by playing it back . Otherwise I'm going dedicated recorder . All this might be subjective opinion like you said but I can hear that big difference in the low end . |
|
|
|
Post by popmann on Nov 14, 2014 14:19:55 GMT -6
I guess a typical 500hz sine wouldn't tell you if it was a frequency dependent difference....but, I was thinking that--just a sine to make sure the volume is the same. I don't think i matters how you generate it--as long as it's recorded onto all the ADAT tape tracks the same way--then transferred along with the audio. The sine would tell you if there's a volume difference. If there's a no volume difference, but IS a frequency difference....then there's "a sound"....but, if it's sound is dropping the 1.5DB off the signal, it shouldn't be important.
Curious to see how Reaper does....but, I honestly think whether it's just a level drop or a "sound", it's going to be somewhere in the handoff between app and hardware digital mixer on the interface...although....did the 002r HAVE a hardware mixer?
I've saved the WAVs off...I'm about to get started for the day--I'll pull them in and see what I hear....
|
|
|
|
Post by drbill on Nov 14, 2014 14:32:41 GMT -6
Oy.....maybe you just LIKE that 90's sound. heh heh My eyes weren't rolling - they actually popped out. Well....I guess they were rolling when they hit the floor...
There's a huge difference IMO. In favor of PT, but I'm not using a native system..... I don't think that would make a difference though. My thoughts are that you probably like that nostalgic early 90's sound and the ADAT's give it to you,,,
|
|
|
|
Post by scumbum on Nov 14, 2014 14:39:01 GMT -6
Oy.....maybe you just LIKE that 90's sound. heh heh My eyes weren't rolling - they actually popped out. Well....I guess they were rolling when they hit the floor... There's a huge difference IMO. In favor of PT, but I'm not using a native system..... I don't think that would make a difference though. My thoughts are that you probably like that nostalgic early 90's sound and the ADAT's give it to you,,, My test wasn't "Are ADATs better sounding than Pro Tools"? The test was DEDICATED recorder vs Pro Tools . The converters (blackface ADAT) don't mean anything in the test . The test was about is there a difference in the quality of the audio playback between the two mediums . The Black Face ADAT was used as converter only because thats all I have for a dedicated recorder . My bet is the HD24 and Radar dedicated recorder system will also sound better than Pro Tools in just the simple task of playing back audio . |
|
|
|
Post by scumbum on Nov 14, 2014 14:44:17 GMT -6
I guess a typical 500hz sine wouldn't tell you if it was a frequency dependent difference....but, I was thinking that--just a sine to make sure the volume is the same. I don't think i matters how you generate it--as long as it's recorded onto all the ADAT tape tracks the same way--then transferred along with the audio. The sine would tell you if there's a volume difference. If there's a no volume difference, but IS a frequency difference....then there's "a sound"....but, if it's sound is dropping the 1.5DB off the signal, it shouldn't be important. Curious to see how Reaper does....but, I honestly think whether it's just a level drop or a "sound", it's going to be somewhere in the handoff between app and hardware digital mixer on the interface...although....did the 002r HAVE a hardware mixer? I've saved the WAVs off...I'm about to get started for the day--I'll pull them in and see what I hear.... Thanks for the tip ! I'll do the sine wave test . I'm gonna try doing the test in reaper.....but my reaper skills are very much lacking !! I get lost just trying to do simple things . |
|
|
|
Post by popmann on Nov 14, 2014 14:52:36 GMT -6
I listened and did quick analysis. Couple things....the timing isn't stable--they cancel pretty well (not all the way)....and then they don't....then they do...it comes and goes. Now maybe the second point explains it to some degree....but, something's not consistent in the timing enough to actually do cancellation tests.
Second point--there's a couple DB more sub information on the ADAT file. meaning <=60hz. While I would describe it as any more "consistent" it's certainly "more"--which I can see why you'd call it that.
I'll listen again with fresh ears later....
|
|
|
|
Post by popmann on Nov 14, 2014 15:04:35 GMT -6
BTW---doing the tone, since I see there's more sub--run two....a 40hz and a 400hz. You should be able to record it once onto an ADAT track and copy it to the others...redo the transfers.
|
|
|
|
Post by scumbum on Nov 14, 2014 17:33:05 GMT -6
BTW---doing the tone, since I see there's more sub--run two....a 40hz and a 400hz. You should be able to record it once onto an ADAT track and copy it to the others...redo the transfers. Thanks for the help popmann ! I recorded a sine wave at 500hz , 400hz and 40hz to the ADAT tape then re-recorded an ADAT tape , Pro Tools and Reaper bounce to the Mbox and laptop . Looking at the wave forms they all seem slightly different , so trying to line them up to null test was difficult . Maybe I should have recorded one click at the beginning from a metronome to make it easier . Thanks again for analyzing these , heres the link , www.dropbox.com/sh/u89g0x43v7isth8/AABu61CIROr4hNuquPzwMgJJa?dl=0 |
|
ericn
Temp
Balance Engineer
Posts: 16,099 
|
Post by ericn on Nov 14, 2014 17:59:19 GMT -6
Your test shows that you like the playback of The ADAt better.
For a real recorder test take mic pre to a line level splitter match both recorders and compare the playback.
|
|
|
|
Post by scumbum on Nov 14, 2014 18:46:43 GMT -6
Your test shows that you like the playback of The ADAt better. For a real recorder test take mic pre to a line level splitter match both recorders and compare the playback. But then you'd be comparing A/D & D/A converters , and cables..........that alone is a WHOLE different test . My test is digital audio played back from two different digital storage devices and thats it . Everything is the same in all the tests except audio coming from a DAW (computer) and audio coming from a dedicated recorder (ADAT) . Same cables , converters , same performance.....EVERYTHING is the same , The test shows there is a difference , Pro Tools LE in some way degrades the audio , other wise you wouldn't hear a difference or even be able to tell them apart . popmann will help with the technical stuff about whats going on . All I know is the dedicated recorder is sounding better . |
|
|
|
Post by drbill on Nov 14, 2014 18:47:09 GMT -6
Your test shows that you like the playback of The ADAt better. Or that the clocking is not the greatest, or that PT runs better on INTERNAL clocking vs. EXTERNAL clocking like most DAW's (the most likely culprit - per Dan Lavry), or the the levels coming out of PT might be different than the ADAT or the optical cables might be cheap or the pan laws are different (is everything up the middle?), or that the 002 is poorly designed and can't track ext. clocking as well as a 192, or...... We're talking ones and zero's here. Once converted, it's no different than a seagate HD playing back vs a WD hard drive. The difference is in how the clocking it arrived at. Next test, clock the ADAT to the 002. Keeping 002 as master and ADAT as slave. |
|
|
|
Post by scumbum on Nov 14, 2014 19:34:22 GMT -6
Your test shows that you like the playback of The ADAt better. Or that the clocking is not the greatest, or that PT runs better on INTERNAL clocking vs. EXTERNAL clocking like most DAW's (the most likely culprit - per Dan Lavry), or the the levels coming out of PT might be different than the ADAT or the optical cables might be cheap or the pan laws are different (is everything up the middle?), or that the 002 is poorly designed and can't track ext. clocking as well as a 192, or...... We're talking ones and zero's here. Once converted, it's no different than a seagate HD playing back vs a WD hard drive. The difference is in how the clocking it arrived at. Next test, clock the ADAT to the 002. Keeping 002 as master and ADAT as slave. From popmann's advice I added Sine waves on each track , 6 tracks , on the adat tape and then redid the test so levels could be checked on each bounce . Optical cables - I did another test earlier , which I mentioned in the above post , that I transferred audio with the same cables from PT to ADAT tape and then back through the optical connection and that nulled in PT . So a transfer back and forth didn't cause a problem earlier . Theres no Master fader and nothing is panned in PT . Each track comes out its own converter to the mackie where the settings do not change between the bounces . The ADAT gets nonstop pops and clicks when it slaves to the 002r . So the ADAT converter was the Master clock the whole time . But even if I could slave the ADAT from the 002r clock , and I used the ADAT converter for the test , there would be jitter ?? I thought if I used the converters own master clock I eliminate jitter , so that would make the test more concrete . Kinda like the 002r is just a PT dongle , not even being used in the chain . |
|
|
|
Post by popmann on Nov 14, 2014 20:18:54 GMT -6
I'll grab those....what's the signal flow? Recorded on the ADAT tape (from who cares where)-transferred via optical into software....then these are the ADAT playing back into the Mbox....002 playing back into the Mbox using ProTools...and the same files opened and played back through Reaper? There's not really a lot to analyze....are they the same level? If not, how much difference? The purpose of the tones isn't to cancel...it's to see that the DAC is playing back the same level from the tape and software.
Before you go too far down the rabbit hole....let's assume that you get a purer playback from a hardware recorder--that's still not apples to apples....because now you're saying you'd rather get a purer feed of a 90s Alesis ADC at 48 than a "compromised" playback of a 96khz modern ADC? That would of course be absurd, IME. But, that's the actual thing you're proposing. I suppose you could get a newer RADAR with Nyquist conversion....but, you're still comparing the fidelity of one recorder to another--as a total package. You know? And that's even setting aside that much of sonics comes from function. A LARGE amount depending on the style....
Regarding the 002r as "dongle"--no, it's still the audio interface. It's still the source of software's clock. Your slaving IT to the ADAT simply means protools is two clocks down the line. If you're using a Mac running CoreAudio for Reaper, it's three clocks down the line. If it does internally have a hardware routing mixer (like any non Avid interface) it still passes that....
|
|
|
|
Post by popmann on Nov 14, 2014 20:49:29 GMT -6
They're all uniformly close enough to the same level....minus .1db. ADAT=reference level....PT= -.1db.....Reaper= -.2db...for whatever reason, Reaper drops an extra .1 in only the left channel.
So, it's solid as theoretical tests go....problem is (see above)....it's not a real world test. You like the playback of a 16bit tracks on the 16bit machine best. Ok. Fair enough....but, if your 002 at 24/96 interface clock sounds better than the ADAT at 16/48....what does it mean to you?
|
|
|
|
Post by tonycamphd on Nov 14, 2014 20:49:45 GMT -6
scum, i feel your pain man, but do yourself a favor, call BLA and get that 002 sig modded with a FM or XB clock and SMOKE any converter out there! Yeah, i said it, ANY CONVERTER! you will be blown away bromee, i promise! 8)
Remember Mike Shipley loved the BLA FM192 over every converter out there, and Alison Krause and her band found it necessary to start over with the tracking of Paper Airplane when they heard the diffs that converter made, this caliber of artists don't make shit up pertaining to what they hear, believe it.
and you're welcome ahead of time broman.
|
|
|
|
Post by scumbum on Nov 14, 2014 21:38:34 GMT -6
They're all uniformly close enough to the same level....minus .1db. ADAT=reference level....PT= -.1db.....Reaper= -.2db...for whatever reason, Reaper drops an extra .1 in only the left channel. So, it's solid as theoretical tests go....problem is (see above)....it's not a real world test. You like the playback of a 16bit tracks on the 16bit machine best. Ok. Fair enough....but, if your 002 at 24/96 interface clock sounds better than the ADAT at 16/48....what does it mean to you? Thanks for checking things out ! So the 002 is basically the problem no matter what way you use it . Even if you try to bypass it through the ADAT input , your still screwed . I can say for sure that in every situation even at 24/96 the 002r is gonna fail compared to the ADAT running its own clock and tape . I've done too many tests over the years and the 002r converters always fail . This is exactly what I needed though to figure out where to go next . Even if I got a really high end 8 channel converter and ran it through the 002r's ADAT optical I'm shooting myself in the foot because there will always be a clocking issue . |
|
|
|
Post by scumbum on Nov 14, 2014 21:51:26 GMT -6
scum, i feel your pain man, but do yourself a favor, call BLA and get that 002 sig modded with a FM or XB clock and SMOKE any converter out there! Yeah, i said it, ANY CONVERTER! you will be blown away bromee, i promise! 8) Remember Mike Shipley loved the BLA FM192 over every converter out there, and Alison Krause and her band found it necessary to start over with the tracking of Paper Airplane when they heard the diffs that converter made, this caliber of artists don't make shit up pertaining to what they hear, believe it. and you're welcome ahead of time broman. I can't afford it right now . I'll probably either switch over to reaper and use a different interface , go HD24 , or upgrade to PT 9 to use another interface . |
|
|
|
Post by scumbum on Nov 14, 2014 22:04:27 GMT -6
They're all uniformly close enough to the same level....minus .1db. ADAT=reference level....PT= -.1db.....Reaper= -.2db...for whatever reason, Reaper drops an extra .1 in only the left channel. So, it's solid as theoretical tests go....problem is (see above)....it's not a real world test. You like the playback of a 16bit tracks on the 16bit machine best. Ok. Fair enough....but, if your 002 at 24/96 interface clock sounds better than the ADAT at 16/48....what does it mean to you? I have a clock question for you popmann . How about when you have "two" interfaces that are the "same brand" like two steinburg ur824's , does that help with the ADAT clock issue , help to eliminate jitter ?? Since both interfaces have the same type of clock inside them they will communicate better than say a steinburg hooked up to a MOTU through ADAT . Or does that not make a difference ? No matter what slaving to another clock even if its two of the same brand interfaces you'll have jitter problems ? |
|
|
|
Post by drbill on Nov 14, 2014 22:08:01 GMT -6
Thanks for checking things out ! So the 002 is basically the problem no matter what way you use it . Even if you try to bypass it through the ADAT input , your still screwed . I can say for sure that in every situation even at 24/96 the 002r is gonna fail compared to the ADAT running its own clock and tape . I've done too many tests over the years and the 002r converters always fail . This is exactly what I needed though to figure out where to go next . Even if I got a really high end 8 channel converter and ran it through the 002r's ADAT optical I'm shooting myself in the foot because there will always be a clocking issue . You're making a lot of assumptions that your test isn't backing up. First, it's not PT that's the problem as initial thread title and assumptions pointed towards. It's the 10 year old digidesign bottom of the line interface - the 002r. There's no doubt that it's low end. Tony's suggestion is a good one. The 002 is a good candidate for an upgrade. The fact that the ADAT is popping and clicking when clocked to the 002 shows that there's something wanky going on. Could be cabling, could be internal clocking issues, could be something set wrong in the software, could be a component failure. But one thing is certain, there have been hundreds of records made with ADAT's clocking back and forth with PT. You cannot assume that the 002 cannot clock better to a third party alternative than it does to the ADAT. Your test has not proved that. Perhaps borrow a 3rd party interface with ADAT lightpipe and give it a try. But know this, AES clock is better than ADAT clock. Perhaps a 3rd party interface would allow you to clock with AES, and transfer with lightpipe. Too many options, too many variables to make definitave statements at this point. Good luck finding the problem. Tony's option is a no brainer. I'd talk to the people at BLA and ask how it would react to being clocked via a 3rd party interface with their mod. Or perhaps, the 3rd party interface will work fine with the BLAclockified 002. Either way, I see no evidence that -- with converters out of the equation -- a dedicated playback unit is any better than PT or any other DAW. |
|
|
|
Post by popmann on Nov 14, 2014 22:38:49 GMT -6
I don't think your conclusion is coming from the right place....an upgraded ADC will run on it's internal clock resulting in better sounding raw capture. I also don't think you can stretch a digital tape machine from 1990 to an Alesis from 97 or any random hardware recorder....and no way you can compare that at 48 to any ADC I've used at 24/96.
Do you know what level the 002 is calibrated for as zero? I'm curious, because AI maintain that ALL ADCs are non linear....the better ones just have a much wider dynamic window of sounding good. But, fed the level they're designed for....they're mostly interchangeable. But, then I guess fwiw, when I listen to those clips I don't hear anything more different than a hundred other things. Ie--less difference than any two converters I'd call interchangeable...less difference than preamps I really don't care a lot about-meaning they sound different but not inherently better/worse in a broad sense....and nothing compared to changing MICS or moving them an inch.
I would caution that going with an old deck and affordable desk, you're literally giving up all the sonic related function for what would be a nuance of purer playback. By that, I mean time aligning drum MICS....gates and expanders as needed....minor automations...being able to track without concern for track count....more flexible EQ....being able to buss tracks together into the same compressor or analog channel...as many compressors or EQs as a mix needs....etc--all functions that directly affect the sonics of the end result. And since I promise any 24/96 ADC I've used will lay waste to an old ADAT....you don't really even get better raw playback fidelity unless your budget is for a newer RADAR with Nyquist cards.
It's also possible you just love the sound of the ADAT converters....you know as...your own subjective preference. As we've seen time and again, fidelity isn't always what people crave. If this studio is for your art....just use BF ADATS. I had someone once tell me how they never could beat their ADATs get this--mixed on a digital Mackie db8 with some low end Lexi and Alesis FX boxes. I think to some degree, all of this is getting the most out of whatever tools you have. And we all have our sensitivities...
|
|
|
|
Post by popmann on Nov 14, 2014 22:44:58 GMT -6
To answer you clock question, you will never get better performance from and ADC than running on it's internal clock. DACs make less difference. You don't want "two interfaces"....the same make will mean the converters have the same latency...but, no--you need a single System with the IO you need for ideal performance. Which is why there really aren't many capable of feeding a desk until you get into professional systems--HD IO and Symphony (and elderly MOTU PCIe systems)....which is what makes the 32x32 Orion noteworthy--but not something alive heard. Just functionally, it turns any laptop into a "deck" to feed a board.
|
|
|
|
Post by Bob Olhsson on Nov 14, 2014 23:20:42 GMT -6
Blackface ADATs used some kind of weird pre-emphasis that sounds like ass using the digital output into an ordinary converter. I always had to come out analog for it to sound right.
|
|
|
|
Post by jimwilliams on Nov 15, 2014 10:19:20 GMT -6
I used to rebuild those things. They used a pretty good delta sigma Crystal ADC chip for those days. The DAC was a decent Burrbrown sign magnatude converter. The design was not set for oversampling so they used a 10 pole passive LPF to roll-off above Nyquist. That filter created about 2500 degrees of phase shift, or group delay.
The ADC converters did a fine job as long as the analog was done right. I have many ADAT releases here that sound wonderful, some acoustic. Those same tracks continue to show their quality by improving their playback with better DAC's.
|
|
