Okay, call me a dummy, but I’ve had a lifetime Roon membership for over a year now, and I’ve not bothered with the DSP functionality out of some “purist” idea that the output is not what was actually recorded. Well, today I decided to try out upsampling via DSD 64 and… Well, it’s utterly transformed my system. It’s like a top to bottom upgrade!
At first I noticed the difference in the upper frequencies - everything is just so much clearer. The additional detail in the sound staging, woah! The layers of detail and creativity in the recording now revealed… Amazing!!! But then I paid attention to the bass, and the differences are just as positive. Every deep note is distinct from one another. I thought I had an awesome system before, but switching on upsampling is just night and day! I’m utterly content with the sound of my system now.
Mac Mini into an Oppo 205. You think 128 is that much better??? I’ll give it a whirl, but it’s hard to believe that the sound could improve any further.
Yep, give it a whirl. DSD128 should be better (in theory). Of course let your ears do the deciding because it can be so dependant on so many parts of the chain, but some of the theory is in the link below, if you’re bored
"Double DSD seemed to “fix” above mentioned compromise that was committed with single DSD by moving the noise shaper from 20kHz up to around 40kHz, well above our standard audio band of 20kHz. It also allowed for gentler and simpler output filters on the DAC. Life became a bit easier and every time that happens in audio we can expect better sonic performance. That is clearly the case with double DSD, and the price of double the data rate seems well worth it."
It could be how your DAC handles lower resolution files (poorly) or an incorrect software setting in Roon. It could also be volume or other settings (filters?). Since your DAC is also a Blueray player try to check if any processing had been turned on. Worth checking. At the end of the day there is no reason for an improvement unless the DAC does a poor job at lower rates. Did you SPL level match with an external SPL meter? Did you A to B with quick changes.
As the article below explains, there are 7 filters and the differences are most audible at 44.1KHz.
Corrected Minimum Phase fast would definitely be audibly different from the other filters at 44.1KHz. As this plot shows
Not a huge difference - just a half dB in the extreme treble but similar to what you reported.
Based on the rest of Archimago’s article, I can’t see any other reason for an audible improvement. If you don’t think it is the filter then you should definitely investigate further as either your device is faulty or you have an incorrect setting somewhere. One possibility might be inter-sample overs - possibly the file you use to test has many clipping events and by converting the file you are addressing this. There is also a software setting in Roon to fix this on all files (-3dB setting) or you could use a Benchmark DAC which can handle inter-sample overs. There are a couple of other DACs that handle inter-sample overs but I can’t recall them right now. Intersample overs are ubiquitous in pop rock digital music but unfortunately DACs that can handle this and eliminate the distortion caused are few and far between.
So for adding Roon to existing Naim equipment, is the SonoreUPnP bridge.
It runs on their hardware (low noise environment, plus you can add a low noise PSU), behaves a full Roon Endpoint but outputs into the UPnP input on your Naim network player, with input switching when playback is initiated.
Works like a treat on all formats, PCM & native DSD.
Also allows you to apply DSP to the playsream, and upsample to max rate or convert to DSD.
No format conversion to S/PDIF & the opportunity to introduce jitter into the playback stream.
I now have the UltraRendu with UpTone UltraCap LPS as a pair (so looks good) into my NDS.
Plus MQA support in the Roon Core.
See SonoreUPnP Bridge (beta) and MQA unfold in Roon 1.5 over UPnP Bridge and https://forums.naimaudio.com/topic/sonore-upnp-bridge-for-roon
I’m not quite sure I follow. I haven’t changed any settings in the Oppo, including the digital filters. All I’ve done is tried out Roon’s upsampling features. The Oppo is now receiving DSD64 from my Mac Mini. The same music sounds dramatically on the Oppo better when upsampling in this format.
I also like DSD (I upsample and send DSD128 to my Pro-Ject Pre Box S2 Digital), and I let Roon handle the MQA first unfold for MQA content (MQA is also upsampled to DSD128 instead of the second MQA unfold). The only exception is if I do something that demands a lot from the computer (like some games), in which case I disable up-sampling.
It is quite simple really. There is no magical benefit to converting files to DSD. If the same file is sounding dramatically different when upsampled then something has to happen to make it sound different. As I showed you in a plot of the Oppo filter at 44.1 KHz and the Archimago review - you can see that this output filter is very different for different filter selections and would be even more different on a file converted to DSD. Perhaps you are hearing the affect of the different filters.
At the end of the day you may not care what is causing the difference. I understand that too. You simply want to report how happy you are with DSD upsampling and that is fantastic! Good discovery.
The article is a bit outdated. Already there are commercial chipsets which can work with DSD256 (quad) and even DSD512 directly without conversion to PCM. For example, AK4490/93 and AK4497.
I’ve only changed one variable in the chain. I’ve gone from Roon with no DSP/Upsampling to using DSP/Upsampling via DSD 64. USB before, USB after. I therefore don’t understand why you are commenting on the different filters in the Oppo. I’ve not changed the Oppo’s filter.
As for measurements, yes, I’m a strong believer in data. As a matter of fact, using my RTA I perceive that the highest frequency band is more active than before. Nevertheless, what it really comes down to for me is what my ears hear. Am I going to go through the effort of double-blind testing myself? No. But am I going to discount experiences based off of what I can and can’t measure? No.
Already there are commercial chipsets which can work with DSD256 (quad) and even DSD512 directly without conversion to PCM. For example, AK4490/93 and AK4497.