Review of DJ equipment?

[kemmler3D]

Measurements of the Pioneer DJ / CDJ-2000 NXS2 and CDJ-3000 with TEMPO slider set to 0%, +0.15% and +15%. MASTER TEMPO is not used.
TEMPO RESET: ON gave the same results as TEMPO RESET: OFF + TEMPO slider: 0%, so it is not included.

Thanks for posting these!

Do you know what Master Tempo is doing such that it creates a 30-40dB SINAD penalty? The output looks pretty gnarly with it on.. some kind of time-stretching on by default?

 

[nagster]

Measurements of Pioneer DJ / CDJ-2000 NXS2 and CDJ-3000 playing an inter-sample test signal.
The test signal was generated by Multitone Analyzer @pkane. An RME / ADI-2 DAC was added for comparison.

 

[nagster]

Do you know what Master Tempo is doing such that it creates a 30-40dB SINAD penalty? The output looks pretty gnarly with it on.. some kind of time-stretching on by default?

I think the additional calculations make it worse, but I don't know the details. For some reason it seems to have gotten even worse with the CDJ-3000.
The image was cut from the manual.

 

[nagster]

These are FFT Spectrum of Pioneer DJ / CDJ-2000 NXS2 and CDJ-3000 playing the J-test signal. MASTER TEMPO is on and off.

 

[naagratrax]

Measurements of the Pioneer DJ / CDJ-2000 NXS2 and CDJ-3000 with TEMPO slider set to 0%, +0.15% and +15%. MASTER TEMPO is not used.
TEMPO RESET: ON gave the same results as TEMPO RESET: OFF + TEMPO slider: 0%, so it is not included.

This is soooo cool, thanks a lots !

Finally some AP measurements for The Pioneer CDJ

It's possible to test the digital output ?

Do you have access to some digital Pioneer Mixer or some Allen & Heath Xone 96/92 ?

I'm semi professional club DJ myself in paris, the THD Noise is correct/good without the master tempo but with it it' really bad

 

[oozlum]

Hi, thanks for publishing these measurements! One thing I don’t understand is that jitter is ‘more spurious’ when master tempo is off. Shouldn’t it be the opposite since master tempo on introduced unwanted artefacts in your previous measurements? Thanks for your reply.

 

[Severian]

Very interesting, thanks for posting. Would love to see some mixer measurements.

No self-respecting DJ should be using master tempo anyway...

 

[nagster]

It's possible to test the digital output ?

Below are just a few of the measurement results for the digital output of the CDJ-2000NXS2 and CDJ-3000.
There is also an AP loopback measurement for comparison.

The images are numbered and arranged in order to make them easier to understand, but for some reason the order may change after a few days. There is nothing I can do about it.
In that case, please view them in numerical order.

edit: The order of the images is already messed up... I'm sorry it's hard to see.

 

[nagster]

One thing I don’t understand is that jitter is ‘more spurious’ when master tempo is off. Shouldn’t it be the opposite since master tempo on introduced unwanted artefacts in your previous measurements?

You may be wondering. Anyway, that's what I got with the J-test signal. In particular, I measured the CDJ-3000's MASTER TEMPO: on / 0% plot multiple times because it also looked clean at first glance.
The appearance is more dramatic with MASTER TEMPO: off. With MASTER TEMPO: on the spurious becomes less noticeable but the skirt becomes wider. Also, a rise around 11kHz can be seen in all plots.
Attached is a re-measurement of J-test signal playback on the CDJ-3000. This time the X-axis display range has been expanded.

 

[nagster]

Do you have access to some digital Pioneer Mixer or some Allen & Heath Xone 96/92 ?

I went to a company I know, but both seemed to be in use. I'll put it on my to-do list.
I brought the remaining DJ mixer. I took some measurements, but I don't have time to edit them so I'll just attach the photos for now.

 

[naagratrax]

Don't hesitate if you can post the measurements later

I've been waiting a long time for CDJ measurements

I'd be curious to see the difference on Pioneer's competitors with the Denon SC6000

I heard the frequency response with a -6db at 20khz

I went to a company I know, but both seemed to be in use. I'll put it on my to-do list.
I brought the remaining DJ mixer. I took some measurements, but I don't have time to edit them so I'll just attach the photos for now.

 

[nagster]

I asked around to see if anyone had DENON DJ, but I couldn't find anyone.

Below is the frequency response of the CDJ-3000.
When MASTER TEMPO was ON, there was constant slight fluctuation around 20Hz and 2kHz.
When MASTER TEMPO was OFF, the FR was stable regardless of the slider position.
Repeated measurements yielded similar results.
Just to be sure, I added another unit, but there was no noticeable change.

edit: There seems to be an emergency with the ASR server. Depending on the situation, we may delete the post temporarily.

 

[nagster]

When I post measurements, I usually try to stay in the role of tester, but this time I'll add a quick comment.
I heard that Pioneer is the mainstream digital player.
It would be difficult for the PA side to remedy these results, and I felt it was disappointing considering the possibility that tens of thousands of people would be listening to the CDJ sound.
One measure that DJs can take is to prepare a sound source with a level that is lowered by about 4dB to 5dB. This will avoid problems with inter-sample over, and may also reduce degradation when using MASTER TEMPO.
However, manually processing hundreds or thousands of songs is a nightmare. I think some kind of automation is needed.
Ideally, this is something I'd like to see dealt with on the CDJ side.
MASTER TEMPO may be necessary for DJs. I don't know. The trade-off is increased distortion, noise, and FR fluctuation.
It was more transparent to leave the TEMPO slider unchanged, but this difference was small compared to the impact of MASTER TEMPO.

 

[Querkle]

thanks for these measurements, pioneer should be launching a new player soon so im curious to see how that does, id also love to see measurements on their A9, V10, and Euphonia Mixers and players like XDJ700

 

[Severian]

thanks for these measurements, pioneer should be launching a new player soon so im curious to see how that does, id also love to see measurements on their A9, V10, and Euphonia Mixers and players like XDJ700

Has there been any actual info on a new standalone player from Pioneer?

I use a DDJ-1000 controller and would really like to move to a standalone setup, but CDJs are simply ludicrously expensive and the XDJ 700/1000 are laughably outdated at this point.

I've been seriously considering a pair of used Denon SC6000 players but I don't particularly want to add new library management software to the mix. I like Rekordbox.

 

[Querkle]

i wonder if SD slot performs the same as USB stick? also new CDJ3000X has been teased/semi announced by AlphaTheta/Pioneer, curious how that will measure up.

 

[NTTY]

Below are just a few of the measurement results for the digital output of the CDJ-2000NXS2 and CDJ-3000.
There is also an AP loopback measurement for comparison.

The images are numbered and arranged in order to make them easier to understand, but for some reason the order may change after a few days. There is nothing I can do about it.
In that case, please view them in numerical order.

edit: The order of the images is already messed up... I'm sorry it's hard to see.

Thanks for all the details measurement which probably means an ASRC is at play to output 48kHz whatever it's fed with, and its internal interpolator has no digital headroom and is not protected from overflow.

 

[nagster]

Thanks for all the details measurement which probably means an ASRC is at play to output 48kHz whatever it's fed with, and its internal interpolator has no digital headroom and is not protected from overflow.

I don't have access to a CDJ right now, but from what I remember, it was as follows:
44.1 kHz source -> 44.1 kHz coaxial output
48/88.2/96 kHz source -> 96 kHz coaxial output

I can only guess at the sampling frequency of the DAC section, but I will check the coaxial output again next time.

 

[NTTY]

Even for a 1:1 conversion, it’s likely to process the digital signal via its ASRC, before digital output, and so will suffer same overload in case of digital sample overs. To check indeed

 

[Reticuli]

Wow, the CDJ-2000NXS2 upsamples 48khz to 96khz. I was not expecting that. You should make sure that you're testing 44.1khz files over SPDIF to a 44.1khz digital receiver, and if you're testing 96khz files, make sure the receiver is running at 96khz, too, rather than SRCing any of the digital outs to the analyzers.

The CDJ-3000, by the way, resamples everything to 96khz similar to the way the new Denons do, rather than having any change in the SPDIF rate, but the CDJ-3000 appears to use a better quality interpolation than the new Denons that seem to use a very computationally efficient method. The new Denons use superior key correction, though, but distortion is of course cummulative. If you're not using key correction, the CDJ-3000 should give superior SPDIF output, while if you're using keylock, it starts to get into a matter of taste in some respects. Personally, I still give the overall sound nod to the CDJ-3000 even with key correction, but others disagree with me.

InMusic actually raised the corner frequency of their anti-aliasing filters for the digital audio processing that's happening all the time on the players after we complained about the substantial high frequency roll off in the DSP phase. While they did improve the extension a little, especially on higher rate tracks, they now appear to let in more aliasing and therefore intermodulation distortion plus noise that particularly shows up on frequency sweeps and on a 20-tone IMD test, indicating an IMD 'hump' on complex mid-amplitude material. In contrast, if you run a 2-tone SMPTE IMD test that simulates only more minimalist music, it comes up pretty clean on the new Denons. There's also some ultrasonic grunge still getting through. Personally, I'd rather have more treble roll off than more aliasing, and I wish when you turned the B layer (and effects) off you got better quality SRC interpolation since you'd be freeing up available resources. IMO, the current overall sound with lossless on the Denon Prime units is closer to lossy audio files on older models... assuming you're comparing keylock off. Again, if you're doing keylock, that's dominating the distortion and frequency bandwidth changes. Here are some videos of test waveforms:

The last versions of Virtual DJ 7 were actually bit perfect at zero pitch regardless of master tempo being on or off, and while for a while there later versions had some issues, I have been told that their latest versions once again can do that... of course assuming the rest of your system is not SRCing the stream and you don't have VDJ's mixer, effects, and stems separation in the path.

The oldest Pioneer DJ CDJ-2000 were usually bit perfect from their SPDIF at zero pitch regardless of whether keylock was on or off. Their Wolfson DACs weren't too bad, if a bit shimmery. The older pre-InMusic Denon DJ decks tended to have better-sounding DACs, though. I don't think I ever bothered testing the old Denons because they always sounded amazing. I did also test the Gemini MDJ and Hanpin players, though: