[LFF]

Quote:

Thought I would wake this thread from the dead...
 
I have been using my M-Audio 2496 to digitise my Vinyl - lots of it is 50's and in middling condition which means the rip is quick and the cleanup takes forever!!! (CoolEdit)
 
I recently started testing the M-Audio using Headphones (just getting back into HP's after a 15+ year hiatus)... and I found the M-Audio analogue out to be a little less transparent than the digital out into my AVR and out into a Headamp.
 
This caused me to start looking askance at the 2496.
 
So my question is has anyone compared different ADC's for vinyl conversion (or any other recording purposes) - and without getting into stratospheric pricing categories, is there something better than the 2496, or should I look at mods for the 2496?
 
The MSB PAD sounds like an interesting option as does the MSB Audio Director - has anyone compared these to direct PCI based ADC's?
 
Has anyone compared some of the Xonars to M-Audio and E-MU ADC's?

I have used all types and frankly, the best bang for the buck is the EMU. It's fairly transparent and does a great job for vinyl rips. Your average joe will NOT be able to tell the difference between the ADC of an EMU or that of a Lavry Gold or a Benchmark. Stick with the EMU as the law of diminishing returns really applies to the DAC and ADC world. Most importantly, I'd would say that the most critical thing about doing a proper vinyl rip is the the following:
 
1. Quality of the vinyl.
2. Quality of the mastering on the LP.
3. Being able to clean the vinyl properly with the proper solution
4. Capture resolution.
5. Processing order.
6. Post-processing polishing.
 
A well done vinyl rip usually takes me around 3 hours if the record is clean and minty. Badly treated records take more time. When all things are the same, it's very hard telling the difference between the 16/44.1 rip and a 32/196 rip.

 

[dlaloum]

My current process for ripping is as follows:
 
1) Initial dusting
2) Preliminary cleaning with Knosti Disco-Antistat and my own custom cleaning solution (alcohol + Teepol + purified water)
3) First Rinse - followed by Drying on the Sota Vacuum cleaner
4) AIVS Record Cleaning process (enzymatic cleaner soak, purified water rinse on the Sota)
5) Last Record cleaner followed by Preservative
 
Possibly a little extreme - But I want to do this ONLY ONCE - and then have a digital copy from there on!
 
Due to the noise of the Sota cleaner, I put on my Sennheiser PXC-450 noise cancellers while doing the cleaning cycles, and I tend to do the records in batches of 5-10 at a time.... the soak stage breaks things up a bit but it takes a while regardless.
 
Once that is done, I move to the turntable (Revox B795 with BenzMicro MC1HO into Creek OBH18 connected directly to Analogue in of the M-Audio 2496) cue things up and check the recording levels.
 
I do the recordings at 24-96 and I then do the post processing in CoolEdit - the post processing can take from 30min to a day or more. 
 
Some of the scratchy oldies that are not replaceable and rare have taken 8-10 hours to clean up.
 
After processing I may drop res to 16/44 depending on the record.... 
 
BUT: given the amount of time and effort involved, it would just kill me to find that I had not achieved the best possible transparency...
The end result should be as much as possible indistinguishable from the original, and I am wondering whether I can achieve that with the 2496.
 
M-Audio is one of E-Mu's main competitors, and the 2496 is a competitor to the 0404.... but I have not seen head to head comparisons of the two cards, or of competitors to them with a focus on recording quality (ADC) as opposed to playback (DAC).
 
I was also hoping to avoid getting into the higher price categories.... 

 

[LFF]

Quote:

My current process for ripping is as follows:
 
1) Initial dusting
2) Preliminary cleaning with Knosti Disco-Antistat and my own custom cleaning solution (alcohol + Teepol + purified water)
3) First Rinse - followed by Drying on the Sota Vacuum cleaner
4) AIVS Record Cleaning process (enzymatic cleaner soak, purified water rinse on the Sota)
5) Last Record cleaner followed by Preservative
 
Possibly a little extreme - But I want to do this ONLY ONCE - and then have a digital copy from there on!
 
Due to the noise of the Sota cleaner, I put on my Sennheiser PXC-450 noise cancellers while doing the cleaning cycles, and I tend to do the records in batches of 5-10 at a time.... the soak stage breaks things up a bit but it takes a while regardless.
 
Once that is done, I move to the turntable (Revox B795 with BenzMicro MC1HO into Creek OBH18 connected directly to Analogue in of the M-Audio 2496) cue things up and check the recording levels.
 
I do the recordings at 24-96 and I then do the post processing in CoolEdit - the post processing can take from 30min to a day or more. 
 
Some of the scratchy oldies that are not replaceable and rare have taken 8-10 hours to clean up.
 
After processing I may drop res to 16/44 depending on the record.... 
 
BUT: given the amount of time and effort involved, it would just kill me to find that I had not achieved the best possible transparency...
The end result should be as much as possible indistinguishable from the original, and I am wondering whether I can achieve that with the 2496.
 
M-Audio is one of E-Mu's main competitors, and the 2496 is a competitor to the 0404.... but I have not seen head to head comparisons of the two cards, or of competitors to them with a focus on recording quality (ADC) as opposed to playback (DAC).
 
I was also hoping to avoid getting into the higher price categories.... 

 
You should be ok. I know a few other professional engineers who use M-Audio and most use them for stuff like vinyl transfers or cassette tape transfers. The results depend mainly on your skill for listening and processing rather than the equipment. 24/96 should be more than enough IMHO. Aiming to make it indistinguishable from the original should be your main goal UNLESS you want to improve it. Post a clip of your work....I'd love to have a listen.
 

 

[dlaloum]

Quote:

Aiming to make it indistinguishable from the original should be your main goal UNLESS you want to improve it. Post a clip of your work....I'd love to have a listen.
 

What I need is for the ADC to record indistinguishably from the original.... and then my objective with the cleanup work in the digital domain is remove flaws (scratches, clicks, pops) with as little impact on the original recording as possible.
That means going though the recording at times second by second to pick up flaws.... brute force automated plug-ins tend to impact the recording too much (although when set to lighter settings they seem to work well)... So I start with the automated plug-ins to pick up the obvious stuff - then go to manual which takes forever on badly treated/handled LP's.
 
I have been noticing low level hash/noise, which seems to be caused by digital signals - not immediately obvious - and difficult to compare as listening to the recording requires the PC to be on as the primary source.... so my hash generator is in residence!
 
Which leads me to consider an external ADC - which would allow me to bypass the PC altogether for analogue, and deliver a clean signal - then the PC sends the clean bit perfect signal on to the DAC.... 
 

 

[dlaloum]

Just ran a test comparing M-Audio 2496 recording to the original LP - it is very very close...
 
Without substantial further time and effort hard to tell whether it is possible for me to differentiate between the two.
I do have some left/right balance issues - which are being exposed by the test... it seems as if my recording ADC's may be marginally misaligned - which requires a calibration LP to set up etc... (which I have but cannot be bothered with right now)
 
I made two recordings 1st via my AVR (Onkyo TX-SR876) phono input - then out from the Pre-out (which allows gain adjustment) into the M-Audio, the second from my Creek OBH18/OBH2 into the M-Stage (acting as pre) into the M-Audio.
 
The difference between these two recordings (ie the two phono stages) is far far greater than any difference between the original LP played directly into the M-Stage for HP listening, and the recording decoded by the AVR DAC's and out into the M-Stage for HP listening.
 
Within the constraints of my monitoring & playback system the recording is effectively identical to the original.
 
I might for the time being stop worrying about ADC's.... 
 

 

[jp11801]

Here is a sample of a vinyl rip I just completed, if vinyl is low-fi then I love low-fi!



Description: BW Sample



You can use the following link to retrieve your file:



http://www.sendspace.com/file/nb3hww

 

[silverxxx]

i wonder, can you post a higher sampling rate sample if you have one, it would be interesting to compare.

 

[googleborg]

i've downloaded a few vinyl rips done by people who obviously care about it, and was quite surprised at the quality...24/196, but my DAC only does 24/96 or 16/196 so no wasapi, heh.
 
 
edit : listening to Tin pan alley by SRV now from the vinyl rip, via Headphones, really nice, but i still prefer the CD rip

 

[jp11801]

Here is the same 55 sec clip originally done in Sound Designer ll 24/192 and each sample was taken from that file to produce the three AIFF files. Resampling and Dither done in wave editor using the izoptope Mbit codes.
 
IDLM 16 44 24 96 24 192

You can use the following link to view the folder:

http://www.sendspace.com/folder/f50j8l

 

[googleborg]

thanks! i'll give them a listen tonight when i can concentrate best

 

[googleborg]

well i did a test between the 16/44 and 24/96, and it was actually easy!  i couldn't do the 192 as my dac (Aune) wont take it and just 'unlocks' 

 
if you ask me the one sounded more, uhm 'airy'? than the other, clearer too.
 
i should have hidden results also, but it really was quite obvious.
 

 

 

[silverxxx]

Yes, interesting samples, thanks for posting them. The 192 sample is not so different from the 96 one to my ears (i cant tell them apart). On the 44khz sample the sibilance is also worse.
 
I wonder if the differences are due to the sampling rate or bit depth.  I'll do some testings now...