Recording an Internal Mix in Traktor over 2048MBs :(

Recording an Internal Mix in Traktor over 2048MBs :disappointed:

Hello, fellow DJ professionals, music junkies, gear heads and music maestros! I’m here because my mix files continue to split at around 32-45 minutes in and I don’t know when it will happen. I can’t record an outro to the mix, and I feel it’s a race against time. I would much rather be able to record a set at whatever file size I like and whatever length.

The file cuts after exceeding 2048MBs, is there any way around this? I’m using a Traktor Kontrol S4 MK2 but I haven’t found a way to record “External”.

mod edit: You cant post mixes until you reach 25 posts

thats 3 1/2 hours mate.

Very strange that it’s splitting where it is… as Jester says, that’s a lot more than 45 minutes.

The technical reason for the 2048MB limit is due to Traktor still being 32 bit software - 32 bit operating systems could not address files larger than 2GB.

I’m using a 64-bit operating system, if that makes any difference. Though I know what you mean, Traktor is a 32-bit software. Maybe it’s that I’m recording at such a high sample rate (192000 Hz)? I’ll try to record yet another mix again and see where I get cut off at, hopefully I’ll be able to record beyond an hour’s length of time; we’ll have to see. All of my settings are exactly as they should be despite my sound card not being ASIO compliant. It’s a Sound Blaster Recon3D PCIe.

Can you take the two halves and join them in a DAW? Audacity is free.

Are you really using files that take advantage of the higher sample rate? otherwise you are just creating larger files for the sake of it. Anyway, once Traktor crosses the threshold it will start recording a new file and you can easily merge them with no audible spot using Audacity or any other audio editor.

I think we have a winner

That’s the sample rate I record mixes at … For my dog.

Yeah, that, right there.

Even if somehow you had tracks that were recorded and encoded at that high a rate (a not easy feat in itself), you have exceeded human hearing by 335% (that is 4.35 times higher frequency than what a human can hear). There is no need to do this at all. Even 48KHz is overkill.

Heck, this is 1.6 times higher frequency than what a dog can hear.

haha

Okay, so…

Then, what sample rate should I be selecting instead? Here are the options …

192000
176400
96000
88200
48000
44100
24000
22050
11025

I want to spare everything for the sake of audio quality and lag has yet to be an issue, I have a multi-core processor. All of your help and hilariousness has been a great reassurance and aid of great quality! :thumbsup:

44100 … zero reason for recording at higher mate.

Quick audio lesson:

44100 Hz is the optimal sampling frequency. Human hearing is generally from 20 Hz to 20 kHz (although it fluctuates in people, with older people not able to hear as high pitched sounds). The Nyquist-Shannon sampling theory states that anything you’re trying to reproduce must have a sampling rate at least twice as high the original. Think about when you’re in a very fast car and staring at another car’s rims. Even though you’re going forward, your eyes don’t have a high enough sample rate, so it looks like the rims are moving in reverse. The additional 4.1 kHz is to allow for anti-aliasing which I don’t remember enough of the lecture to regurgitate but you really don’t need to choose anything other than 44100. Nothing higher, nothing lower.

Just so to set the record straight.. Sample rate has to do with how many samples or points of data are taken, not tone.

So say 48000 vs 96000 means that there are more samples being taken giving you a more accurate/detailed sound, not that it will be high pitched.

However, you are totally right in that it’s not necessary to record at such a high rate. Most tracks are mastered at 44100/16 bit depth anyways.

It’s just a clarity thing :slight_smile:

That is not quite right, but I think you are misunderstanding the conversation.

The number of samples you take directly correlates to the maximum frequency that is captured… because frequency is “number of oscillations per second,” and sample rate is “number of samples per second.”

It sounds like you think, though, that what we’re saying is higher sample rates will make the recording higher frequency - that is not what is being said here at all. What is being said is that the number of samples taken affects the RANGE of frequency that is captured. It is equal to half the sample rate, so a sample rate of 44100 Hz is equal to a maximum frequency range of 22050Hz - which is beyond the maximum range of human hearing. As stated earlier, the extra bit is there is nice for dithering, which is another topic that is beyond the scope of this thread, I think.

Not quite. You get most of your accuracy and detail from your bit rate, not your sample rate. An 8 bit sound is much less accurate than a 16 bit sound, and sounds noisier because there are larger jumps between the values of each sample in your curve. This is especially pronounced in 8 bit vs. 16 bit (as opposed to 16 bit vs. higher rates). Signed 8 bit has a minimum value of -128 and a maximum value of 127… when your wave is converted (since it is typically represented as -1 to 1), that means that you have a whole lot of graininess in your sound because the jumps are relatively huge. Once you get to 16 bit, the gap of 1 bit in the total range is pretty small (because the minimum value is -32768 and maximum is 32767), and while noticeable to a discerning ear, isn’t a huge deal.

Now, at a low sample rate, yeah, it will lose quality, but that is because you lose frequency range. For example, if your sample rate is half of CD quality, 22050Hz, then that means the highest frequency you can capture is 11025Hz - which is actually pretty low, you lose almost all of your sizzle and most of your hats that way.

However, any sampling above 44100Hz is outside the range of human hearing, except for particular (rare) mutations. I suspect if you were playing an old folks home, you could probably sample at about 33100Hz and do just fine. :wink:

Why is Traktor only allowed to create a 2gb file being a 32bit piece of software, but I can record 7-8 hours (4.5gb) worth of .wav audio at 44.1k in Cool Edit Pro every weekend without any issue. Doesn’t sound like a 32bit limitation to me…

I have the same beef too. I go on binges and record em every time. I re-listen while Im driving to learn my tunes and see what mixes worked and what simply doesn’t.

riggghhhht … all the dude needs to know is to use 44100 khz .. stop confusing the lad :slight_smile:

riggghhhht … all the dude needs to know is to use 44100 khz .. stop confusing the lad (or use audition/audacity/soundforge to record instead)

I LOVED Cool Edit Pro. :+1:

Rather than clutter this post up further, here’s an article about it. It is a holdover, related to FAT.