Hi. This is a really good site. I certainly wish I'd found it before i started spending money.

I am commited to an amateur project which requires me to splice together approx 5 hours of music audio in the digital domain. At least 3 hours of which I have to obtain from vinyl as it simply does not exist in any other format.

I do not feel I am obtaining the quality of result from my attempted rips that I would expect. I outline my equipment and method below in the hope that some knowledgeable and kind soul can identify the weak points in my technique. Failing that if spending more money, commensurate with my existing spend - say £200-£300 max. - would lead to significant improvement.

Equipment.

Signal Path In

Denon DL-103 MC cartridge - I also have access to an AT 120E & Stanton Trackmaster Mk II + standard Stanton 500AL but the Denon sounds best to me atm.
Technics SL-1210 Mk (standard arm)
CA Azur 640P-B MM/MC Phono pre-amp.
( Soundcraft Spirit M4 analogue/digital mixer - that's 44kHz x 24 bit but VERY low output from digital - optional)
M-Audio Fast Track Pro USB soundcard (up to 96kHz x 24 bit but 44khz x 16 bit is normal)
Intel i7 quad core PC running Vista Home Premium- 64 bit
Audacity Vn 1.2.6. (also possible access to Ableton LE 6.0.2)
Ashampoo Burning Studio 6 Vn 6.75
Optiarc DVD RW SATA 7200
Verbatim super Az0 CD-R disks

Signal Path Out - Monitoring & Playback

via the PC - FTP - Soundcraft to M-Audio AV40 near field monitor OR Kenwood KR7040 & TDL RTL 3 Hi Fi playback.

Technique.

Not much to say really. i have read the guides on this site. My vinyl, half electronic and half what might be loosely termed rock, is of excellent rather than mint condition. Cleaned dry. All equipment is set up as per manufacturers recomended settings. Mixer and/or FTP just below red line. I've used manual settings through Audacity trying to get either -3 dB or 0 db. Or automatic which comes in around -10db then normalising either/or via the Audacity plug in or on Ashampoo neither of which goes above -3db so manual workls best but see below).

Results.

Well I'm quite aware that I'm going to get some quality loss due to format change but what i'm getting still seems below what i might expect.

The first issue is one of volume equalisation of course. Mixing between audio ripped or played direct from CD and or WAV/MP3 320 bps downloaded from professional sites such as Beatport the volume is about 20% down.

However even after adjusting levels using the mixer the quality of my rips is down significantly. I can play some of the same samples direct from vinyl and compare to WAV/MP3 320bps. Different but within tolerable limits for the type of listener the project to designed for but comparing my rip to the equivalent WAV/MP3 320 bps the difference is immiediately detectable (disinterested subject = girlfriend). In some ways the result might be acceptable if the total project could be ripped but it cannot and the flow is broken and jerky as the different source samples come in and out. The main issue is one of loss of presense and dynamics. My rips sound 'flat' and dead. They are warm enough but lack presense and 'bite'.

I know we are expected not to be subjective on this site but in lack of experience I can only estimate atm. I have found, to my surprise, that it's difficult to tell the difference between a downloaded or ripped .WAV and a very high quality MP3 (? sry) but my rips are below that I get via Spotify direct (196bps AAC I think) and Last FM 9196 bps MP3 I think) but perhaps rather better than standard 96 bps MP3.

I have access to a compressor/limiter/gate (RNC 17373) but i don't think thats the issue. It seems to my, admitedly amateur ear, that what I need is something to revitalise the sound.

Any ideas?

edited for spll/grmr.

I am not a mixing or mastering engineer but these are the thoughts I have.

All of this certainly sounds like you're dealing with a mastering issue. You're trying to compete with a professionally mixed/mastered product; what you mix is always going to sound quieter. And the problems you seem to be describing are not related to quality per se - the dynamics of each mix element are not doing what you want them to do, so that they advance and recede in the mix when you do not want them to.

Apply loudness equalization on ALL the samples before mixing them, and apply loudness eq on all professional mixes you're comparing against, and your own mix, when you compare them. Differences in volume should not be of concern at this point. If you're concerned about it in the final product, hire a mastering engineer.

Try running each track through a dynamic range compressor before mixing them, to avoid the issues you mention with samples fading in and out.

You might try using the AT120E instead of the DL103, as the latter has a bit of a high frequency rolloff and added distortion; the colder sound of the 120E might not sound preferential in isolation but it might work better in your mix.

I doubt any of this has to do with the bitrates you are encountering. The other quality issues you seem to be dealing with (presence/bite/etc) really sound like mastering differences - the flatness you are hearing, depending on your source material, may be completely expected. You might seriously want to consider hiring a mastering engineer to deal with that.

You could try record using 24/96 'resolution' and then SRC/dither down to 16/44.1. Voxengo r8brain is quite good software for SRC/dithering. IIRC, you need to install device drivers from M-Audio to get 24/96 available.

Also, try using native ASIO drivers for M-Audio and recording using ASIO compatible software. Sometimes you get better input levels through ASIO path.

Check articles from Bob Katz
http://www.digido.com/honor-roll.html
http://www.digido.com/articles-demos.html


Juha

Thank you to Axon and Juha for their informative and useful replies and also to anyone who read the thread and didn't make a rude reply to a 1st post.

I'd like to restate my problem in a simpler way.

If I mix sections of my selection 'live' alternating between turntable and PC/soundcrd sources I can detect the difference between the vinl and .WAV/MP3 but my audience cannot. Well at least not to the extent that it interferes with their enjoyment of the narrative flow.

However if I previously record my vinyl (see above post #1) then mix the selection in it's entirety the difference between the original sources is immediately apparent and it does indeed distract the listener even after I have compensated for volume differences via the mixer. Of course to some extent this is to be expected but the extent has surprised me.

Why I ask is because so many of you have already digitised vinyl and report good to excellent results. Perhaps if I played a whole side of a album, took a break, then played the same abum side as recorded, maybe adjusted a few settings I wouldn't notice but mixed directly I certainly do. Why don't you?

Axon's suggestion seems best. As professional mastering is out of the question at this stage the mest way to proceed is to process the pre mastered segments to match what I can achieve by recording and then reprocess the final result. Does seem a shame though to effectively drop quality to the lowest common denominator rather than raise up my samples to match the best.

maybe i should try and obtain access to professional DJs and find out how they deal with this issue although at, I suspect they must have access to proffesional mastering equipment as axon suggests.

Could it be possible for you to upload one or couple 30 second sample(s) (Uploads sub forum) from your vinyl rip(s)?

Have you also tried if

Har-Bal - http://www.har-bal.com/
or
AAMS - http://www.curioza.com/

would help you on this?



Juha

Hi,
Well, I'm not familiar with any of the recording chain you're using so I don't have any concrete suggestions and I've never had vinyl rips sound flat or dead. However, it sounds like you're saying that your rips are missing some upper mid-range and treble and maybe don't sound as "loud" as CD rips. If that's the case you could apply some eq and some light compression with an audio editor (try the freeware "Audacity" if you don't already have one). Another thing you might try if "loudness" is the main problem would be to run everything through Wavgain. That should level out the volume differences. I've never actually used Wavgain so can't advise but if you search the forum I'm sure you will find plenty of help

That's a good idea.

It's holidays atm so it'll take me a day or so to find the time to do it properly.

The suggestion to use the AT120E rather than DL-103 for recording was a good one too even if the DL103 still sounds better 'live'.



but have you ever tried mixing recordings back together with the live or CD quality originals?

Yes

I agree with botface, that it sounds mostly like an equalization problem (frequency balance). A lot of records are a bit "dull" sounding (lacking in high frequencies). The cartridge can make a difference, but EQ can make a bigger difference. (Use the best sounding cartridge). Most records were made with older technology, and I think most producers didn't really care about frequency response... "Good enough for AM radio!"

You may also need some dynamic compression. Most modern CDs are over-compressed, and this is one reason that some people prefer vinyl. (See Loudness Wars.)

I'd start by boosting the high frequencies. Choose a digital source as your reference, and try to make the vinyl transfers "sound like" the reference. (Of course, different recordings, different musicians, different equipment, different producers, different times... You're going to get different sound, and there's only so much you can do.)

Then, match the levels. This can be tricky, because some quiet sounding tracks may have high-level peaks, and if you increase the level you can get clipping (distortion)...

The general procedure is to normalize everything for peaks at (or near) 0dB. Then, find the quietest sounding track, and bring the levels of everything else down to match the quietest track. (by ear).

If you're still "not there", you can try some compression. Or, you may find that your overall level seems too low after lowering the volume of almost everything. Again, compression is the answer. Compression/limiting can bring up the average level without increasing the peaks (or clipping). You can use Audacity or your hardware compressor. (There are advantages to software compression, but if you're familiar with your hardware compressor, it might be easier to get the "right sound".)

Note that compression reduces dynamics, so don't abuse it! You can generally "get away with" more dynamic range in a live venue than in a home/studio. And, since it's normally used to boost levels, compression will boost any vinyl noise on the records.

Weird! I'd expect very similar results. You are probably constantly monitoring/adjusting volume levels during the performance. Are you using the compressor when working live? Or, any EQ adjustments on the vinyl channels?