Introducing Olympus & Olympus I/O - A new perspective on modern music playback

For those who just started reading up on Olympus, Olympus I/O, and XDMI, please note that all information in this thread has been summarized in a single PDF document that can be downloaded from the Taiko Website.

https://taikoaudio.com/taiko-2020/taiko-audio-downloads

The document is frequently updated.

Scroll down to the 'XDMI, Olympus Music Server, Olympus I/O' section and click 'XDMI, Olympus, Olympus I/O Product Introduction & FAQ' to download the latest version.

Good morning WBF!​

We are introducing the culmination of close to 4 years of research and development. As a bona fide IT/tech nerd with a passion for music, I have always been intrigued by the potential of leveraging the most modern of technologies in order to create a better music playback experience. This, amongst others, led to the creation of our popular, perhaps even revolutionary, Extreme music server 5 years ago, which we have been steadily improving and updating with new technologies throughout its life cycle. Today I feel we can safely claim it's holding its ground against the onslaught of new server releases from other companies, and we are committed to keep improving it for years to come.

We are introducing a new server model called the Olympus. Hierarchically, it positions itself above the Extreme. It does provide quite a different music experience than the Extreme, or any other server I've heard, for that matter. Conventional audiophile descriptions such as sound staging, dynamics, color palette, etc, fall short to describe this difference. It does not sound digital or analog, I would be inclined to describe it as coming closer to the intended (or unintended) performance of the recording engineer.

Committed to keeping the Extreme as current as possible, we are introducing a second product called the Olympus I/O. This is an external upgrade to the Extreme containing a significant part of the Olympus technology, allowing it to come near, though not entirely at, Olympus performance levels. The Olympus I/O can even be added to the Olympus itself to elevate its performance even further, though not as dramatic an uplift as adding it to the Extreme. Consider it the proverbial "cherry on top".

 

I suspect Reference Recording "filtered" the various formats to justify the price difference. Pure speculation on my part ...
View attachment 135757

In this and similiar cases the recoding engineer at NativeDSD, Tom Caulfield (a great guy btw), uses HQPlayer Pro and the DXD edit master to produce the lower resolution formats. Digital filters are necessarily involved in HQP, but nothing to intentionally degrade lower-priced offerings. If there is a difference between the resolutions, they are speaking for themselves. I have the 24-bit DXD of this recording, but I haven't compared it with others. I suspect this debate could become every bit as impassioned as DSD v. PCM and delta-sigma v. ladder. Once emotions settle, maybe we can get an accurate read of some of these things on the Olympus.

 

In this and similiar cases the recoding engineer at NativeDSD, Tom Caulfield (a great guy btw), uses HQPlayer Pro and the DXD edit master to produce the lower resolution formats. Digital filters are necessarily involved in HQP, but nothing to intentionally degrade lower-priced offerings. If there is a difference between the resolutions, they are speaking for themselves. I have the 24-bit DXD of this recording, but I haven't compared it with others. I suspect this debate could become every bit as impassioned as DSD v. PCM and delta-sigma v. ladder. Once emotions settle, maybe we can get an accurate read of some of these things on the Olympus.

Thanks. Frankly I lost interest (though cannot help jumping in now and then) in such SQ debate. All I know is that on Extreme the Bohm 7 (1977 analog in Red Book CD issue) sounds more natural and organic than the 2018 Honeck 9 DSD256 recording to this pair of ears. SQ is hugely personal like most other things in audio or life. One cannot get into the core classics (the most "learned" music) if one pays too much attention to SQ. I have around 40 different albums of Bruckner 8 alone (the greatest symphony ever composed IMHO). In the end, I love only a handful (1944 Furtwangler/VPO live mono; 1955 Beinum/RCO studio mono; 1970 Horenstein/LSO live; 1987 Wand/NDRSO Lubeck Dom live; 1986 Suitner/SKB studio; 2017 Jansons/BRSO live; and also 1979 Jochum/SKD studio for a fast and highly fluid take). The latest recordings of B8 in DSD or PCM 24/96 or 24/192 do not sound better (let alone the performance) (the likes of Nelsons, Thielemann, Janowski, Zweden, Barenboim, Gergiev, Young, Poschner etc.). If I have to choose one for my desert island at this moment, I would choose Horenstein/LSO 1970 live issued on BBC Legends label ("this is it!" upon first hearing; yet to hear more; good sound remastered in 20 bits but issued as red book). SQ is of almost the least importance in that select group for me.

 

John, I am sure that is the case but it is way too early to draw any conclusions about recording resolutions. Not enough people have their Olympus or have had time to really test differences.

And less of those who have received their Olympus, are actually after burning in completed.
We need to wait and be patient.
But it looks from early reports it is going to be a big thing.

 

I have hesitated to write something in this thread on the subject of Delta Sigma chips. I don't want to be misunderstood because I am a great enthusiast of Taiko Audio. The ingenuity of Emile and the excellent build quality of all the equipment is a role model for me.

Since the discussion about upsampling continued here, I would like to make some suggestions from a technical point of view.

Answering the question behind the question: I don’t care if the DAC chip is Delta Sigma or not, I only care how it sounds, which btw is largely dominated by power supplies and I/V in my experience so far.

Emile understood the question behind the question right away, of course.

In order not to bore readers who are not interested in technical details, here is the:

Short version

There is no such thing as bit-perfect playback from delta-sigma DACs! This is because each of these chips contains an interpolator with digital filters and a modulator, which converts each PCM file (e.g. 44.1 kHz) into a bitstream. This bitstream is then DSD.

The filters and modulators in the chip must be designed with simple algorithms due to the low computing power. Therefore, every Delta Sigma DAC can benefit from receiving PCM > 44.1k or DSD as source material in order to bypass the internal (simple) oversampling.

Long version

How does a delta-sigma chip work? I did a bit of research and it seems that the Olympus I/O XDMI DAC module contains a ROHM chip.

XDMI Digital section - that's the secret sauce, and I know absolutely nothing about it (other than it's using a ROHM DAC chip).

Let's take a look at the block diagram of a BD34352EKV chip ‘as an example’ below. PCM as input goes via Audio Function Control to a FIR filter. With this filter, for example, PCM 44.1k is up-converted to 352.8k or 705.6k. In a second step, oversampling to the modulator rate takes place. This is, for example, 5.6448 MHz (16 x 352.8 kHz). This corresponds to DSD128! The bitstream generated by the delta-sigma modulator is then converted into an analogue signal.

It is interesting to note that DSD as an input completely bypasses the internal oversampling of the ROHM chip, as it is already available as a bitstream.


Source: New 32bit D/A Converter IC for Hi-Fi Audio Equipment | ROHM Semiconductor - ROHM Co., Ltd.

In order to bypass the chip's internal oversampling, external upsampling with higher precision (e.g. floating point instead of fixed point), better filters and higher quality modulators can be used. DSD requires very high computing power. And here Emile is of course right that more power = more noise and interrupts have a negative effect on the sound. There are strategies for this. For example, you can buy high-quality DSD as source material. Or you can do the upsampling with another computer using software such as PGGP or HQPlayer.

Summary

Taiko Audio is designed for bit-perfect and noise-free signal processing. Anyone who buys this device should operate it exactly as the manufacturer recommends.

In my opinion, the Olympus ROM DAC-Chip or your DAC can also benefit from upsampling. But I would either buy DSD as the source material straight away or convert it to DSD online or offline using another computer. Because the Delta Sigma chip does it anyway, only with presumably worse algorithms. These chips are widely used, e.g. ESS Sabre DAC chips, AKM, ROHM or Burr-Brown.

Neither the Olympus nor the DAC recognises that the material is upsampled. Instead, both treat it as ‘bit-perfect’ source material.

Whether this really sounds better is something that everyone can decide for themselves. With my contribution I want to give a suggestion as to why upsampling can improve the sound and what the technical background is.

 

All theories and prior experiences become irrelevant the moment you introduce the Olympus XDMI into your system.

I'm not going to spoil the experience for you, but once you get your Olympus, the debate over high-res vs. Redbook vs. DSD is resolved within five minutes - it's as clear as night and day.

It goes something like this: When you first plug in the Olympus (and give it about a day), you're absolutely blown away. You realize that you've never truly heard digital music before. For the first couple of weeks, your brain is so overwhelmed that distinguishing any differences is nearly impossible. At that point, don't even bother comparing high-res and low-res tracks. But once your brain adjusts, it only takes five minutes to decide what you prefer.
YMMV

 

All theories and prior experiences become irrelevant the moment you introduce the Olympus XDMI into your system.

I'm not going to spoil the experience for you, but once you get your Olympus, the debate over high-res vs. Redbook vs. DSD is resolved within five minutes - it's as clear as night and day.

It goes something like this: When you first plug in the Olympus (and give it about a day), you're absolutely blown away. You realize that you've never truly heard digital music before. For the first couple of weeks, your brain is so overwhelmed that distinguishing any differences is nearly impossible. At that point, don't even bother comparing high-res and low-res tracks. But once your brain adjusts, it only takes five minutes to decide what you prefer.
YMMV

Hi nenon,

What did you prefer?

 

All theories and prior experiences become irrelevant the moment you introduce the Olympus XDMI into your system.

I'm not going to spoil the experience for you, but once you get your Olympus, the debate over high-res vs. Redbook vs. DSD is resolved within five minutes - it's as clear as night and day.

It goes something like this: When you first plug in the Olympus (and give it about a day), you're absolutely blown away. You realize that you've never truly heard digital music before. For the first couple of weeks, your brain is so overwhelmed that distinguishing any differences is nearly impossible. At that point, don't even bother comparing high-res and low-res tracks. But once your brain adjusts, it only takes five minutes to decide what you prefer.
YMMV

Pearl of the Day!

 

I will echo @nenon System burn in and brain burn in are real with Olympus. Be patient and give yourself time. What XDMI does is different than other digital levers and optimizations I've heard before. Those others levers are still there, but the XDMI impact is so outsized and impactful that it dwarfs the others.

I've had to reset many (most?) of my mental models over the last week. Lots more critical listening this week before coming to any sort of conclusions, but I am starting to get a good sense for it

As I mentioned to some folks privately, Emile and team are living a couple years into the future of digital audio, and over the last month, the rest of us are just starting to get a taste for what that means. Olympus isn't just more/better of what has come before, it is a different way of doing digital. Looking back, you can see the hints of this approach in the Extreme, but the intentionality to lean into that approach with Olympus is extremely clear.

Embrace Beginners Mind, and enjoy this ride. This is the beginning of a new wave of digital, not an improvement on what has come before. Hope to have more cogent thoughts over the next several weeks.

 

Or you can do the upsampling with another computer using software such as PGGP or HQPlayer.

Your analysis is spot on, and there are several of us "waiting for Olympus" Extreme owners who are doing exactly that (pre-upsampling with PGGB to 16FS PCM or DSD512) with our current DACs.

However, Emile has warned us to keep our minds and ears open as we move to the Olympus. While the advantage of presenting the Rohm chip with pre-upsampled 16FS PCM or DSD512 is technically clear, it does come at the cost of noise generated from simply playing back these higher rates in Roon. This comes from loading the data from storage, and transporting it over the digital interface to the DAC. In the ultra-low-noise world of the Olympus, these things now start to matter a lot.

I know one or two Olympians who are already starting to understand these tradeoffs, and I expect to do the same when my Olympus arrives.

Ultimately, our choice of sample rates and formats will be determined by what sounds best on the Olympus. This is the same approach that drove me to use PGGB DSD512 on my Extreme with my current DAC. If it now drives me to prefer native sample rates of 1FS and 2FS on Olympus, so be it.

My ears will be my guide, as always.

 

No Doubt, It will be an adjustment and an interesting exercise on a number of levels. As @ray-dude explains "resetting many (most?) mental models" This in itself will take several deep breaths...

 

What did you prefer?

I will answer indirectly - I've listened to Qobuz streaming exclusively in the past couple of weeks.

Going from XDMS back to Roon has been so good ( sorry @EuroDriver ). I've discovered a lot of new music. As I've mentioned before, even music that's not well recorded sounds amazing with XDMI. While I was excited to go from a NOS DAC to a DAC that supports native DSD and bought a bunch of DSD albums, I realized I am much more drawn to the music with Qobuz streaming than DSD albums. And that's all that matters to me.
I converted from a total geek who could not stop tweaking things around to someone who does not obsess with any of that and just enjoys the music available from streaming services. And I can tell you - life is so much better this way!

Did I say that "theories and prior experiences become irrelevant" with the Olympus XDMI? I think I did...

 

Horizon1 with Sigma-Delta chip should in theory process native DSD files better than PCM files. But in reality I get the most natural and organic sound from Red Book albums with H1. I once experimented by converting WAV files to DSD beforehand (with such software as dbpoweramp) and playing them in H1 (just as StreamFidelity recommended above) but heard no improvement at all. In the end, how an album was recorded is the most critical. I am yet to hear unmixed (with only 2 mono mikes) and uncompressed music files but trust they would sound the best. I am puzzled by the expressed need to adjust one's mental processes for the Olympus sound. The yardstick is live music (natural timbre, immediacy of sound, long decay, etc.). We need to regularly recalibrate our ears with live music. If not, our ears are conditioned by hifi sound (often with exaggerated and boom-box bass). If O sounds more like live music (experienced in near field), it should be an instant relief and met with instant recognition. Once I have heard the more natural sound from the purist 4xRCA/i2s connection (between modded DVD transport and DAC), the more artificial and veiled sound from the USB connection became constantly noticeable.

 

so I m now day 3 and as everyone says, it only gets better. Im done with USB and cannot stop listening to native XDMI. FWIW and much to my amazement I am back at output level 63 on my Horizon so I am passing the full signal and my Lamm preamp-setting is back to detente position 5. I hate to sound cliché but this XDMI is something that must be heard to be believed. Not digital, not analogue but something so extraordinary that I still need a few more days to get my thoughts together.
The belll cannot be unsung. Im glued to my seat. Im playing Roon which I never used and cannot stop listening. Everything sounds amazing

 

I m reading about bit perfect and all this upsampling stuff but I can tell you XDMI is the future and its here now. The system is settling and as stated I am able to pass full signal through my Horizon and my preamp is set where I always have it. I have a special guest here today who is smiling ear to ear. Big reveal to come later

 

I don''t even look any more at which format I am playing because everything is just so enjoyable. I'll take more time later with DSD and hires but right now I honestly don't care

 

I have hesitated to write something in this thread on the subject of Delta Sigma chips. I don't want to be misunderstood because I am a great enthusiast of Taiko Audio. The ingenuity of Emile and the excellent build quality of all the equipment is a role model for me.

Since the discussion about upsampling continued here, I would like to make some suggestions from a technical point of view.


Emile understood the question behind the question right away, of course.

In order not to bore readers who are not interested in technical details, here is the:

Short version

There is no such thing as bit-perfect playback from delta-sigma DACs! This is because each of these chips contains an interpolator with digital filters and a modulator, which converts each PCM file (e.g. 44.1 kHz) into a bitstream. This bitstream is then DSD.

The filters and modulators in the chip must be designed with simple algorithms due to the low computing power. Therefore, every Delta Sigma DAC can benefit from receiving PCM > 44.1k or DSD as source material in order to bypass the internal (simple) oversampling.

Long version

How does a delta-sigma chip work? I did a bit of research and it seems that the Olympus I/O XDMI DAC module contains a ROHM chip.



Let's take a look at the block diagram of a BD34352EKV chip ‘as an example’ below. PCM as input goes via Audio Function Control to a FIR filter. With this filter, for example, PCM 44.1k is up-converted to 352.8k or 705.6k. In a second step, oversampling to the modulator rate takes place. This is, for example, 5.6448 MHz (16 x 352.8 kHz). This corresponds to DSD128! The bitstream generated by the delta-sigma modulator is then converted into an analogue signal.

It is interesting to note that DSD as an input completely bypasses the internal oversampling of the ROHM chip, as it is already available as a bitstream.


Source: New 32bit D/A Converter IC for Hi-Fi Audio Equipment | ROHM Semiconductor - ROHM Co., Ltd.

In order to bypass the chip's internal oversampling, external upsampling with higher precision (e.g. floating point instead of fixed point), better filters and higher quality modulators can be used. DSD requires very high computing power. And here Emile is of course right that more power = more noise and interrupts have a negative effect on the sound. There are strategies for this. For example, you can buy high-quality DSD as source material. Or you can do the upsampling with another computer using software such as PGGP or HQPlayer.

Summary

Taiko Audio is designed for bit-perfect and noise-free signal processing. Anyone who buys this device should operate it exactly as the manufacturer recommends.

In my opinion, the Olympus ROM DAC-Chip or your DAC can also benefit from upsampling. But I would either buy DSD as the source material straight away or convert it to DSD online or offline using another computer. Because the Delta Sigma chip does it anyway, only with presumably worse algorithms. These chips are widely used, e.g. ESS Sabre DAC chips, AKM, ROHM or Burr-Brown.

Neither the Olympus nor the DAC recognises that the material is upsampled. Instead, both treat it as ‘bit-perfect’ source material.

Whether this really sounds better is something that everyone can decide for themselves. With my contribution I want to give a suggestion as to why upsampling can improve the sound and what the technical background is.

I have no idea and cannot comment except about what I am hearing and Nenon is right on in his comments

 

I had a too short session at Nenon’s last week. We made adjustments to amplification (Aries Cerat Aperio) but not to the digital side, which was the Olympus XDMI analog out. So I can’t offer direct compares, although the sound was vastly improved from my last visit months ago with my Horizon and his Extreme.

I concur with the comments about mental realignment of expectations. But again my time there was too short (2.5 hours) to fully acclimate. I’ll describe what I heard by saying less mental processing is required because there is more and less to contend with. More resolution. Less inherent noise. (Couldn’t believe it was Roon playing). This makes the music more present and more relaxing mentally. IMO, it is the noise component which has kept digital off of the precipice of performance. Until now. I feel validated in my choice to move forward with the Olympus XDMI. The elves just need to their part now

 

A quick comment about the BPS app. I set all devices to auto and left the default charge time to between 1200 AM and 0700. Beyond that I see no real reason to futz with any settings. Taiko has set the time according to where we livened default charge to 1200 AM to 0700. From this point on I see no reason to use it unless to trouble shoot an issue. Everyone just needs to listen,. I have still yet to hear the analogue card but I have listened using USB which is way ahead of that heard with the Extreme and to XDMI via Lampi input and having heard this I doubt if I will get to hear the analogue card until Emile and Chris visit here for the long awaited weekend open hose demo

I am also looking forward to swapping out my present XDMI horizon to the new Horizon 360. I also am upgrading my existing KBL XDMI cable to the newer upgraded KBL version....so lots of fun things to look forward to. I understand my new Horizon ships this next Monday

 

A quick comment about the BPS app. I set all devices to auto and left the default charge time to between 1200 AM and 0700. Beyond that I see no real reason to futz with any settings. Taiko has set the time according to where we livened default charge to 1200 AM to 0700. From this point on I see no reason to use it unless to trouble shoot an issue. Everyone just needs to listen,. I have still yet to hear the analogue card but I have listened using USB which is way ahead of that heard with the Extreme and to XDMI via Lampi input and having heard this I doubt if I will get to hear the analogue card until Emile and Chris visit here for the long awaited weekend open hose demo

I am also looking forward to swapping out my present XDMI horizon to the new Horizon 360. I also am upgrading my existing KBL XDMI cable to the newer upgraded KBL version....so lots of fun things to look forward to. I understand my new Horizon ships this next Monday

Steve, how many Horizons do you have by now… 15?!

 

Just one. I had an original one from a few years ago. Then with Native XDMI coming from Lukasz, I have that one at present and next week I have the H360 coming


I love tubes and my entire system is tube based . That I why I am presently so enamored with the Taiko native XDMI sound