Hi folks,
I am back for another review. Today's one is about the
Rotel DT-6000 CD player and DAC. The latter is self-qualified by the company as a
"DAC Transport", which sounds a bit unusual. It has been launched in 2022, as part of the Diamond Series line, on the occasion of the Rotel's 60th anniversary (which was really in 2021). The Diamond Series is supposed to fill the gap between the regular 15X2 midrange line and the higher-end Michi electronics. There are only two products: one integrated amplifier, the
RA-6000, and the player measured here, the DT-6000. It comes at a fairly high price tag, costing
2299€ as the time of this writing. This sample has been kindly loaned to me for the purpose of this review by Masimo France (formerly Sound United), which is the distributor of Rotel products, at least for EU market.
Rotel definitely remains a traditional, old-school manufacturer. It should be one of the very last companies to still make FM tuners by 2023, as they have very few, if none, raw DACs or standalone streamers listed in their product line-up. As such, they also have a long tradition with classical optical players. While keep releasing CD players in the era of streaming is certainly kind of hazardous, I personally do have a loving relationship with the Red Book format, and would still be in pursuit of buying such a product. Besides, while I am familiar with their amplifiers, I look forward to seeing how a D/A converter from this iconic brand would perform. This unit in particular features an ESS ES9028PRO chip, and would be, as such, the first Sabre implementation in a Rotel product, which previously used Wolfson, AKM and now, TI chips.
The DT-6000 design is quite a departure from the rounded corners we used to see with their previous CD Players and Amps, such as the 1572 or 1592. You got here a much more rectangular drawing. This Player simply looks magnificent. These sides, made of polished metal, are just amazing and there is a very nice grained painting (that you cannot see on pictures) on the top steel. In brief, the build quality is top-notch. The player is rather big (41cm wide, 32cm deep) and heavy (8.1kg) and feel very solid in hands.
You get quite a few buttons on the front, and basically, every command you would need without using the remote. The disc tray seems to be quite sturdy, and works smoothly and silently. The CD playback is also pretty fast, with no disk that I could not read, even the most scratched ones. Definitely a strong quality device overall, for the few weeks I used it.
The rear first seems very conventional, with both XLR balanced and RCA single-ended outputs. What is not that common, however, is to not see digital outputs at the back of a CD player. Yes, these S/PDIF ports are inputs only, for both optical and coaxial. There is also a USB-B input, that works through and XMOS chip and a dedicated driver. The USB-A is only here for software update.
Since Masimo allowed me to open the unit, you get as bonus a (uncommented) teardrown :
View attachment 302492View attachment 302493View attachment 302494View attachment 302495
Measurements
Disclaimer: Measurements you are about to see are not intended to be as precise or extensive than what you get from a 30k€ AP. There is obviously both hardware and software limitations here, so not quite apples to apples comparison with Amir's testing. Still, this data is enough to have a pretty good idea if the gear is bad or not, stellar, broken, or sub-par...
- Instruments : RME ADI-2/4 PRO SE. E1DA Cosmos ADC (Grade B), Minimum phase filter. E1DA Cosmos APU 60dB preamp is used for DR measurements. Output voltage is measured separately, using a DMM with a 0dBFS 1kHz tone.
- Software : REW V5.20.14, Multitone Loopback Analyzer 1.0.80 and RMAA 6.4.5 PRO,
- Method : 8 runs for each test, then I choose the closest to the average. Bandwidth and sampling rate to be specified for each test.
Note: Prior to this publication, I showed my measurements to Masimo, who put me in contact with Rotel to discuss the results.
I started my tests with the USB-B input and XLR balanced outputs.
Tests | Manufacturer's specs | Results (L & R) | Rating |
Noise Level (REW) | -115dB (stated as SNR) | -116.0dBA | Excellent |
Dynamic Range (REW) | 99dB | 110.1dB | Very Good |
THD (REW) | / | 0.00017% | Very Good |
THD+Noise / SINAD (REW) | -103.1dB | -105.3dB | Very Good |
IMD SMPTE (REW) | -98.4dB | -99.5dB | Good |
Stereo crosstalk (RMAA) | -115dB | -115.3dB | Very Good |
IMD+Noise @10 kHz (RMAA) | / | 0.00149% | Good |
Multitone 32 TD+N (Multitone) | / | -88.9dB | Average |
Note: I now use a new column for published manufacturer's specs (when some are available), that gives an interesting comparison between what is advertised and what you buy at the end.
Alright. My first thoughts here were about the surprising correlation between advertised specifications, and my measured results. For almost every part, the DT-6000 performed according to specs, or a touch above. You may assume that this should be the norm, but it is not. Apart from some manufacturers we often see measured here on ASR, a lot of products on the market are sub-performing in regard to exaggerated numbers. In this case, I am pleased to observe such cohesive (while good but not great) benchmarks.
Now, if I take a step back, I could admit that the performance is not quite there for the high price tag. The SINAD is effectively good, but a lot of cheaper DACs will give you better numbers. Should it be audibly transparent? Most probably: yes. Yet, I have got a couple of issues (say, three, really). First is the USB-B input that shows both strange sidebands in my THD+N FFT, for which I would not mind, since it doesn't affect the THD, not to mention in an inaudible way. The second concern is... a terrible jitter performance. While the noise floor seems acceptably low, the main 12kHz sine wave seems to be kind of desynchronized in every J-tests I have run. I shared this results with Rotel, and they told that they will investigate on it. Sadly, the poor Jitter performance is not related to my specific sample, as you may see in SoundStage's (Audio Precision) measurements here. Since they did not measure the USB-B input, tho, I will come back to the matter when commenting S/PDIF inputs. At last, I am not satisfied to see, for a product released in 2022, such an "ESS Hump" with IMD VS Level, that most Chinese manufacturers already fixed a long time ago. These aside, the Rotel DT-6000 is doing alright, for a product from a traditional manufacturer, that stayed out of the "SINAD Race" we witness here on ASR.
What about unbalanced RCA outputs?
Tests | XLR | RCA |
Output Voltage @1Khz | 4.540Vrms | 2.127Vrms |
Noise level (REW) | -116.0dBA | -111.8dBA |
Dynamic range, dB (REW) | 110.1dB | 109.6dB |
THD (REW) | 0.00017% | 0.00020% |
THD+N (REW) | -105.3dB | -104.3dB |
IMD SMPTE (REW) | -99.5dB | -99.1dB |
Stereo crosstalk (RMAA) | -115.3dB | -113.5dB |
IMD+Noise @10kHz (RMAA) | 0.00149% | 0.00160% |
Multitone 32 TD+N (Multitone) | -88.9dB | -88.6dB |
Overall, very consistent results. It is rare to observe such a small drop in performance from XLR. While single-ended obviously catches more hums, it keeps the noise at reasonable levels, and both THD and IMD remain almost on par with XLR. Good.
Let's see if coaxial and optical have the same "issues" as USB...
View attachment 302682
Jitter - 24b/48kHz
View attachment 302683
Overall, both coax and toslink are performing equal, meaning, a hair better than USB-B. Not regarding raw numbers, which are basically the same, but by looking at the shape of the 1kHz sine wave, that definitely shows a cleaner picture.
Back to Jitter, we do have here a much more stable presentation, with somehow a "readable" 12kHz tone... and a ton of sidebands around it. It still is not a good performance, even if considering the absolute level ultimately too low to be audible. As said earlier, these tests are unfortunately backed-up by
AP measurements from SoundStage, that show mostly the same picture. However, when I showed these specific results to Rotel, they reacted very professionally and took the problem quite seriously. They said that their team would investigate and could work on a fix. Could this be done by firmware update only (hypothesis: a DPLL setting too aggressive) ? I hope so. In any case, there was a proposal for testing an eventual updated sample when the investigation would be done.
Now, wait. I just realized that the initial purpose of a CD Player would eventually be to play... CDs.
For these last tests, I spent a short time to burn several disks to then decide which tones I would use. Since I would not be able to synchronize sweeps, there were not an option. The continuous tones, repeated for a long period (allowing me repeatability of testing) worked great. As CD is obviously limited to 16bits, I generated my files from both REW and APx Tone Generator, which are basically the same test tones I use for my 24bits testing. The tones are generated 16bits dithered, 44.1kHz.
Total Harmonic Distortion + Noise (SINAD) - 16b/44.1kHz
View attachment 302685
This is basically the maximum I could measure from my 16bits dithered file, and as, such, as transparent as it can get for a CD Playback. I remember
@amirm finding
almost the exact same picture in his review of the Marantz SA-10.
Just for experiment, I tried with a still 16bits PCM file, but this time generated with a 24bits dither:
Total Harmonic Distortion + Noise (SINAD) - 16b/44.1kHz
View attachment 302686
While it gives better SINAD numbers, it makes the harmonics unreadable with a very busy and blurry noise floor. I would stick to the 16bits dithered for next CD player reviews.
For some reason, my Noise Level test is as good as what I get from 24b/44.1kHz through digital inputs.
Dynamic Range is as expected.
While not looking that good, this is indeed a texbook perfect 16bits J-test picture, the same I may get when measuring the tone directly in the digital domain. Its ENOB even surpasses 16bits. So, the DT-6000 may do a great job with Jitter after all. It is just a pity that other digital inputs are somewhat flawed on that matter.
Conclusions
It is not so easy to come with a wrap-up about the Rotel DT-6000. As a pure audio product, I really liked it. It is extremely well-built, it looks gorgeous, and it sounds good. When focusing on specifications, I would say that I praise Rotel for selling a product that performs right on specs, even when it should the very minimum from any manufacturer out there. Now, what about the performance? Essentially good, absolutely transparent for CD playback on one side. On the other hand, mostly sufficient for 24bits contents in audible thresholds. Now, is it technically perfect? No. There are some issues that should not be observable for a product at the price point. Nor that there would be any excuse from such an established company to make some compromises in designing a D/A architecture. As such, I could not give my "competent" rating to the DT-6000. The Jitter disparities come to mind at first, and their results are not acceptable at this point. Is this specific flaw audible? Not from what I heard, honestly. I also assume that most potential buyers for this Player would intend to pair it with the RA-6000 Amplifier, and I highly doubt that they would be any kind of disappointed, or disturbed by any irregularities in the analog signal. That is for sure, but what we are evaluating here are raw objective results. And in that regard, it falls short of its potential.
On a positive note, I am glad to know Rotel being aware of these measurements, and recognizing that the Jitter behavior would need for a rework. I really hope that they could come up with a fix, as I said, preferably by firmware update. If it is a DPLL adjustment as I suggested, I am sure that they could solve it that way. I sure would be glad to re-test an "updated" sample if Rotel makes the proposal. At last, I would like to thank Masimo France for sending me this sample, always knowing the risk of a bad measuring product that they actually sell. Their efforts for transparency are, again, much appreciated.
Flanker rating: Good Enough