|
Vinyl Touch Pitch Lock
|
View this Thread in Original format
| DJ Visionary |
| I been readin about this pitch lock from Vinyl Touch that keeps the pitch of a track true, even when you speed it up. sounds like a good investment to me, but i need to know a place or website where i can pick up the module....anyone got info on it? |
|
|
| St. Michael |
I looked into it a while ago. I assume it's the same one as Vinyl Touch. It requires a simple installation style modification. The reports I was getting from DJs on several forums was that it incorporates a low quality digital pitch correction with the normal pitch lock flaws found on many devices and software that uses the same technology. So far, I have not found any high quality pitch correction devices or software that really does the trick without unbearble distortion and choppiness.
Maybe what the technology needs is 48 kHz sample rate and simultaneous 44.1 kHz conversion with 32-bit processing and real-time 16 Bit audio conversion for built in pitch correction in CD decks and simply 32 Bit processing of analog audio for TTs (thats an intense amount of processing power). I still don't know if that would correct the problems of processed pitch integrity for tempo changes. Naturally, with the existing processing engines and techniques, as you drift farther from the native tempo the distortion and choppiness gets worse.
I really don't know what the solution is, but so far, nothing really lives up to the desire for pitch integrity for tewmpo changes. |
|
|
| Nic |
| quote: | Originally posted by St. Michael
I looked into it a while ago. I assume it's the same one as Vinyl Touch. It requires a simple installation style modification. The reports I was getting from DJs on several forums was that it incorporates a low quality digital pitch correction with the normal pitch lock flaws found on many devices and software that uses the same technology. So far, I have not found any high quality pitch correction devices or software that really does the trick without unbearble distortion and choppiness.
Maybe what the technology needs is 48 kHz sample rate and simultaneous 44.1 kHz conversion with 32-bit processing and real-time 16 Bit audio conversion for built in pitch correction in CD decks and simply 32 Bit processing of analog audio for TTs (thats an intense amount of processing power). I still don't know if that would correct the problems of processed pitch integrity for tempo changes. Naturally, with the existing processing engines and techniques, as you drift farther from the native tempo the distortion and choppiness gets worse.
I really don't know what the solution is, but so far, nothing really lives up to the desire for pitch integrity for tewmpo changes. |
ableton live seams to work pretty well, of course its not going to do the job for turntables |
|
|
| Dervish |
| quote: | Originally posted by St. Michael
Maybe what the technology needs is 48 kHz sample rate and simultaneous 44.1 kHz conversion with 32-bit processing and real-time 16 Bit audio conversion for built in pitch correction in CD decks and simply 32 Bit processing of analog audio for TTs (thats an intense amount of processing power). |
How do you mean (I'm confused)? What are all these conversions for (internal to the CD player?)? I don't get bit where it goes to 16 bits and stuff.
Also why would it need to be 32 bit exactly? Surely 24 bits would be enough?
It sounds like an interesting challenge anyway. |
|
|
| St. Michael |
I didn't put an incredible amt of thought into the methodology but let me see. You have a CD at 16 Bit 44.1 Sample rate (reading the CD image). This gets converted to an analog signal to output to the master or whatever. This is where the pitch lock comes into play. The analog signal gets sampled at 48 KHZ for better quality representation of the analog frequencies and is processed at 32 bit for the pitch lock effect. Next, the audio must again get reconverted back to analog so yeah I suppose a manufacturer could use a D/A converter from 48 KHZ to analog. the point here is there are alot of real-time steps which pose a likely latency issue due to all of that processing. All of this technology would reside internally inside the CD deck.
As far as Ableton goes I dunno I've heard it sound pretty lousy using the pitch correction when warping. However, I am not a regular user of the program. Maybe I don't really use it correctly. I do have it, and I have tinkered with Ableton, but in my own experience it sounds quite "digitized" or unnatural. |
|
|
| Nic |
| quote: | Originally posted by St. Michael
As far as Ableton goes I dunno I've heard it sound pretty lousy using the pitch correction when warping. However, I am not a regular user of the program. Maybe I don't really use it correctly. I do have it, and I have tinkered with Ableton, but in my own experience it sounds quite "digitized" or unnatural. |
ive found it sounds lousy unless you click the high quality button, it uses more cpu but definatly worth it |
|
|
| Nic |
| quote: | Originally posted by St. Michael
I didn't put an incredible amt of thought into the methodology but let me see. You have a CD at 16 Bit 44.1 Sample rate (reading the CD image). This gets converted to an analog signal to output to the master or whatever. This is where the pitch lock comes into play. The analog signal gets sampled at 48 KHZ for better quality representation of the analog frequencies and is processed at 32 bit for the pitch lock effect. Next, the audio must again get reconverted back to analog so yeah I suppose a manufacturer could use a D/A converter from 48 KHZ to analog. the point here is there are alot of real-time steps which pose a likely latency issue due to all of that processing. All of this technology would reside internally inside the CD deck. |
what would the point be of resampling the cd at a higher bitrate/frequency, if anything this is going to degrade the quality. and why on earth would you want to convert from digital to analog, then back to digital again, surely the pitch correction is done on the raw digital cd data, and then its converted to analog for the final output. |
|
|
| Dervish |
Yeah thats what was confusing me, what I'd do it keep it digital straight off the disc. Then convert it later on.
Your right enough about he possible latency issue, but I recon it would more be because the digital signal processor would be taking samples of a given length. So if for example you wanted to jump to a new bit you'd need to reload the sample buffer.
Though thinking about it you could load a given amount of cue point samples into a buffer too. So that you already have samples for thouse cue points if you jump to them.
But it wouldn't work for fast-forwarding and so on. It would for instant start though as you'd be sitting at the point you need instant start for a bit to load the sample buffer (probebly how they do instant start actually).
I wrote a bit about this kinda thing in another thread yesterday or something. |
|
|
| Stu Cox |
The reason the sound goes grainy IS NOT BECAUSE OF INSUFFICIENT PROCESSING POWER! Well, in hardware it may be an extra limitation but it isn't the main reason.
It's simply because of the processes required to time stretch (which is what has to happen when you change the speed of something but not the pitch). The sound is broken down into its constituent waveforms and these are the extended/contracted as required. There are various different algorithms (Sound Forge has about 20 you can choose from) that can be used to do this which will have varying results (depending on both the algorithm and the nature of the sound). Things like CDJs use the same algorithm for everything, so sometimes can be really effective even at great changes in speed and at other times it starts to sound grainy very quickly.
The obvious next step is to create a system which can work out which the best algorithm to use would be and then apply that, which may be what Ableton uses for it's "Complex" time stretching method, but if so it's only a relatively new technology. It may be something totally different, I'm just speculating here.
On the subject of the Vinyl Touch, it probably is a combination of low digital sampling quality and very simple algorithm being used that results in it sounding a bit rubbish, but my main point is really that you can't expect to improve the quality of timestretching drastically just by sampling at a higher quality - the algorithms you use are going to have a much greater effect!
And Nic's right, the pitch lock processing will be done well before it's converted to analogue, they wouldn't bother resampling. |
|
|
| St. Michael |
| I see all of your points and well taken. Nonetheless, I haven't really heard anyone who is a fan of the pitch lock function on various devices. I have tried to use it on CDJs and it just has that choppiness and distorted quality. |
|
|
|
|
