Members Kaux Posted February 8, 2010 Members Share Posted February 8, 2010 Hi... Reading the thread about high resolution audio I started thinking... that, well... Sample rate and bit depth conversion is a critical step, isnt it? So anyone using higher sampler rates should be even more careful on how they get down to 44.1. I am wrong? Any way, the thing I would like everyone to share is what do they use to make the conversions... I have heard of people exporting full res from the DAW and then using another program just for the conversion, Have heard also of people going out analog through hi q DAC and then recording the music again with another HI Q ADC... So how and when do you make your sample rate and bit depth conversions? Why/how did you choose that method? Link to comment Share on other sites More sharing options...
Members blue2blue Posted February 8, 2010 Members Share Posted February 8, 2010 Here's a handy comparator tool for comparing test results from a range of SRC softwares: http://src.infinitewave.ca/ Of course, you've still got to know what they're testing and the significance. A naive user looking at some comparisons might see 'obvious' differences visible in the the graphs that are nonetheless inaudible. And, possibly, vice versa. Link to comment Share on other sites More sharing options...
Members Kaux Posted February 8, 2010 Author Members Share Posted February 8, 2010 A naive user looking at some comparisons might see 'obvious' differences visible in the the graphs that are nonetheless inaudible. And, possibly, vice versa. Yeah, I found lots of that graphs and studies... just wanted to know how you guys are doing it, and your reasons to do it that way. We I got my first interface 96 Khz capable, and that was when I didnt know much about recording, I started using this capability. After 4 tracks my computer starter to slow down! I wondered why! Now I know that for ME and with MY equipment, 96 Khz doesnt have any practical advantage. So, how do you guys convert sample rates and bit depths? Link to comment Share on other sites More sharing options...
Members Anderton Posted February 8, 2010 Members Share Posted February 8, 2010 When doing 96kHz projects, I use the sample rate conversion in BIAS Peak if I'm on the Mac, or Adobe Audition if I'm on Windows. When I visited the VSL factory several years ago, I noticed they had Cool Edit Pro (now Adobe Audition) on their computers. Given that they also had big-bucks digital audio editing programs, I asked why they used Cool Edit Pro. They said it did the cleanest sample rate conversions of anything they tested. However, all programs have improved since that time. Back then the sample rate conversion in most DAWs was pretty bad, but that's change, as the reference cited by bluetoblue indicates. Link to comment Share on other sites More sharing options...
Members rasputin1963 Posted February 8, 2010 Members Share Posted February 8, 2010 What about me? I'm going from 22k (recording) down to 11k. WINDOWS '95 seems to handle it pretty well. Doing what I can to ensure the best sound.... The final medium is a FLAC... it's lossless, you know, so it should sound really good. There are some rich highs and thumping lows and delicate mono imaging that I'm seeking to preserve in the bump-down. Advice? Should I dither? OK, I'll come clean: I'm marketing a new line of microwave "done" beeps. There is a untapped market for custom beeps in people's kitchens; my prospective customers are discerning types who prefer a more cultured, elegant beep instead of the brusque, bog standard one that comes with so many units. Really smelling "success" with what I'm producing here. Link to comment Share on other sites More sharing options...
Members blue2blue Posted February 8, 2010 Members Share Posted February 8, 2010 FWIW, the SR in Apple's Leopard AU looks very good (and much improved over that in Tiger) in the SRC comparator above. Link to comment Share on other sites More sharing options...
Members blue2blue Posted February 8, 2010 Members Share Posted February 8, 2010 What about me? I'm going from 22k down to 11k. Doing what I can to ensure the best sound.... I believe this GS thread is for you: The Idea of Grit Link to comment Share on other sites More sharing options...
CMS Author MikeRivers Posted February 8, 2010 CMS Author Share Posted February 8, 2010 Reading the thread about high resolution audio I started thinking... that, well... Sample rate and bit depth conversion is a critical step, isnt it? So anyone using higher sampler rates should be even more careful on how they get down to 44.1. I am wrong? You're not wrong, but you may be somewhat out of date. These days most sample rate conversion algorithms work very well. There are some that are marginally better than others, but generally it's not something we agonize over these days. I rarely record at anything other than at 44.1 kHz unless the client asks for higher sample rate files, and then I let him worry about what to do with them. I frequently record at 24-bit resolution, and when I need to convert to 16-bit, I usually use Apogee UV22HR in WaveLab just because I have it. I don't hear any real difference no matter what I use including just feeding the 24-bit file to CD Architect and let it do whatever it does, which I suspect is truncation. Link to comment Share on other sites More sharing options...
Members Surrealistic Posted February 8, 2010 Members Share Posted February 8, 2010 I use Cool Edit Pro - and quite an old version I think. Seems to work pretty well - I can't hear any difference between the original and the downsampled versions. (Maybe because my originals are crud to begin with - who knows? ) Link to comment Share on other sites More sharing options...
Members UstadKhanAli Posted February 8, 2010 Members Share Posted February 8, 2010 I bypass this whole thing by recording in 44.1!!! Wheeeeee!!!!!!!! Link to comment Share on other sites More sharing options...
Members Kaux Posted February 9, 2010 Author Members Share Posted February 9, 2010 I bypass this whole thing by recording in 44.1!!! Wheeeeee!!!!!!!! Not the whole thing unless you record to 16 bit !!! tel me ... tell me... Link to comment Share on other sites More sharing options...
Members spokenward Posted February 9, 2010 Members Share Posted February 9, 2010 ok, I'll play because I recently changed my practice. In Sound Forge 10 they licensed the Izotope SRC. I had been using r8brain for the journey from 96/24 to 44/16. I had sworn off Forge after 8. I still used the Batch converter from 5. The batch converter and the SRC brought me back for this version 10. Link to comment Share on other sites More sharing options...
Members Anderton Posted February 9, 2010 Members Share Posted February 9, 2010 Sound Forge 10 has AMAZING pitch transposition, algorithms licensed from zPlane. Check it out. Link to comment Share on other sites More sharing options...
Members UstadKhanAli Posted February 9, 2010 Members Share Posted February 9, 2010 Not the whole thing unless you record to 16 bit !!! tel me ... tell me... Naw, I record 24-bit. I was just being a smart-ass. Again. Link to comment Share on other sites More sharing options...
Members Kaux Posted February 9, 2010 Author Members Share Posted February 9, 2010 Naw, I record 24-bit. I was just being a smart-ass. Again. So how do you go from 24 to 16? I guess I'll use Audition... since I have it, and it is what Craig uses... If Craig and VSL use it... then it is good for me too. Link to comment Share on other sites More sharing options...
Members Rudolf von Hagenwil Posted February 9, 2010 Members Share Posted February 9, 2010 "Any way, the thing I would like everyone to share is what do they use to make the conversions." sample rate and bit depth conversion and conversion from any format to any format is done here with with: - ProCoder3, converts any format to any format- Grass Valley imports all formats and export to any format- Adobe Media Encoder- WaveLab Crystal Resampler- Waveart Final Plug Link to comment Share on other sites More sharing options...
Members Rudolf von Hagenwil Posted February 9, 2010 Members Share Posted February 9, 2010 Why? At export, the proper trancode setting are dialed according to the specs of the final product, e.g. CDA red book standard, Blu-ray: Dolby Digital (AC-3), DTS, linear PCM. Dolby Digital Plus and DTS-HD High Resolution, Dolby TrueHD and DTS-HD Master Audio. Video editing requires a sample rate of 48 kHz. For DVD-VIDEO, the audio files are converted to the proper audio file type accepted as asset in the authoring: - DTS mono, stereo or up to 7.1 datareduced (compressed)- MPEG-1 Audio Layer 2, stereo or mono datareduced.- MPEG-2 Multichannel bis 7.1 datareduced.- Dolby Digital AC-3 (up to 5.1, datareduced). - PCM 48 kHz, mono, stereo or up to 7.1 uncompressed. So how and when do you make your sample rate and bit depth conversions? When it is necessary for the next work step. Link to comment Share on other sites More sharing options...
Members Rudolf von Hagenwil Posted February 9, 2010 Members Share Posted February 9, 2010 So anyone using higher sampler rates should be even more careful on how they get down to 44.1. I am wrong? SRC and bit depth reduction is done by software. There isn't much you can do wrong. Link to comment Share on other sites More sharing options...
Members Beck Posted February 9, 2010 Members Share Posted February 9, 2010 I resample rather than convert. I Link to comment Share on other sites More sharing options...
Members UstadKhanAli Posted February 9, 2010 Members Share Posted February 9, 2010 So how do you go from 24 to 16? I guess I'll use Audition... since I have it, and it is what Craig uses... If Craig and VSL use it... then it is good for me too. Honestly, no special way. I use Pro Tools and sometimes I use POW-R dither and sometimes I don't. I honestly don't care. I can't hear the freakin' difference, and I have ADAM monitors and good ears. If I'm in a hurry, I'll take a 24-bit mix and let iTunes or Roxio Toast Titanium and let those programs drop 8 bits. This is NOT for a final CD or anything important, of course. Link to comment Share on other sites More sharing options...
CMS Author MikeRivers Posted February 10, 2010 CMS Author Share Posted February 10, 2010 SRC and bit depth reduction is done by software. There isn't much you can do wrong. Well, you could choose the wrong dither algorithm, though most people can't hear the difference on most material. But there are lots of things that the software writers could do wrong. But of course nobody makes those mistakes any more, except for those who have never done it before. Best to use a standard algorithm from a well established vendor. Link to comment Share on other sites More sharing options...
Members bp Posted February 10, 2010 Members Share Posted February 10, 2010 If I'm going to downconvert to 44.1k I usually record at 88.2k. The end result sounds much better to me than 96k to 44.1k. If I'm doing a DVD project that will end up at 48k I may record at 96k. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.