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".

 

The only wart I can see is that many folks proclaimed XDMS was “almost ready for release” a year ago. Having spent time in the world of software I thought that was unrealistically optimistic, and here we are still knocking on the door of a beta release.

Along the way, we get snips of code added to a piece of alpha software and is so good we tolerate the pace of development.

writing commercial software is truly an enormous task on its own, but add to that the repetitive sonic evaluation of each new bit of code. I’m sure even the developers get frustrated if their new functions turn out to have negative consequences.

I imagine looking in and watching as the threads add hundred of pages could be daunting; but you are watching real-time development something most manufacturers would never let you see. The trust this creates is unique.

 

Yes correct with the exception of needing 2 complete XDMI solutions. XDMI can be completely relocated to an external chassis. On the source/server side you then only need this interface card:

View attachment 121971

Something I don't understand Emile. If you place the XDMI card on the DAC side in an external box, how do you communicate this box with the DAC, using a “long” i2s cable?

 

Bob I believe their are 2 Options For The Extreme: Option 1 Olympus I/0, Option 2 Olympus I/0 XDMI. With both one has to return USB and Network card...

Let me explain the Olympus I/O options in a bit more detail:

There's no price difference between adding an Olympus I/O to a Extreme or Olympus server.

The Olympus I/O (export) price is EURO 20.800 without Network, USB or XDMI cards.

We are offering a pre mounting service, the Olympus I/O then ships to you with a Network card AND USB card OR XDMI card pre mounted. You can then just replace your existing USB and network card by the 2 interfaces cards, connect the I/O and ship your old USB and network card back to us. We are doing this to make the Olympus I/O addition as easy as we can.

So in this scenario, you already own the upgraded USB and network cards, which you return to us AFTER you have installed your Olympus I/O, you pay EURO 20.800.

If you DON'T already own the upgraded USB and network cards the Olympus I/O will be non functional, so you'll have to add those to your purchase. This then adds 2* EURO 1600 for a total of EURO 24.000.

If you ONLY own the USB card OR not the network card, you'll have to purchase the one you're missing and pay EURO 20.800 + 1* 1600 = EURO 22.400

If you don't use USB but have a network connected DAC (there are only a handful of those), you will need 2 network cards to be able to fully utilize the Olympus I/O, this does not change the math outlined above as the price of a network card is the same as the price for a usb card. If you have zero network cards you pay EURO 24.000, if you have 1 network card you pay EURO 22.400, if you have 2 network cards you pay EURO 20.800.

If you own an Extreme and want to add XDMI you will have to replace your USB card, so this option adds 15.000 - 1.600 = Euro 13.400 (minus the EURO 5000 introduction discount while it lasts, for Euro 8.400). If you don't own the USB card, you will pay Euro 15.000 (10.000 with introduction discount).

If you add the I/O to your Olympus the math becomes a bit easier, whether you go for USB or XDMI, it will always add EURO 20.800.

 

Emile, I'd like to better understand the advantage Option 4 offers vs. Option 3. I recall in this discussion Option 4 offers a 10-15% improvement in sound performance for the reasons explained. Beyond the performance improvement are there other reasons to add the Olympus O/I such as the physical capability to accommodate future I/O options or would those be accommodated in Option 3 as well?

Those would be accommodated in Option 3 as well.

 

My two cents.

XDMI is designed to improve two things:

1. From player to digital signal out.

2. From digital signal out to DAC getting I2S.

For player to digital signal out: this step will usually need to go through

a) Windows drivers

b) Some hardware for digital output, e.g. a USB PCIE card or a AES/EBU PCIE card

Both of these are very suboptimal at the moment, since in order to manage R&D cost and due to lack of deep enough understanding of all factors that affect sound in the process there are (at least) three limiting factors:

i) Usage of off the shelf driver components/code for the driver

ii) Usage of off the shelf hardware components and schemes for the cards

iii) Non optimal way the driver and hardware work together

This whole area of player to digital signal out seems to have been reimagined from the grounds up in XDMI and it seems to have found a way to transport this signal through QSFP and DAC cable with very close to no loss in SQ.

The signal is transferred from either Extreme or from Olympus to Olympus IO through QSFP DAC cable.

Olympus IO transmits either

a) Digital out through SPDIF or AES/EBU, and even though the digital “close to lossless SQ” signal received by O IO from Extreme or O needs to go through a conversion, it benefits from everything above and the conversion itself I would bet has also been to a high extent rebuilt to minimize SQ loss. In other words, the digital signal received by the DAC is way better than previously through USB.

or

b) Analogue out, and in this case the DAC module receives immediate I2S directly from the digital “close to lossless SQ” signal O IO received from Extreme or O. So even if the engineering surrounding the DAC module itself is somewhat compact, by getting rid of the conversion steps to AES and back to I2S and by powering it with the BPS and separately enclosing it in the O IO the gain might be so huge that it exceeds the engineering of almost any DAC.

Now, if a stand alone DAC could receive the “native” QSFP digital signal, you get all the benefits of the signal transmission and the DAC engineering, but without BPS to power the DAC (of course the QSFP module on the DAC could be externally powered by a BPS). So in that case it would be a matter of the gains in DAC engineering vs the loss of one more QSFP transfer and lack of BPS.

I’m not sure anyone has really heard this yet, but this would be my bet for best SQ. And once people hear the analogue out of XDMI, DAC manufacturers should be much more incentivized to include the “native” module.

I'd say that's spot on!

 

Let me explain the Olympus I/O options in a bit more detail:

There's no price difference between adding an Olympus I/O to a Extreme or Olympus server.

The Olympus I/O (export) price is EURO 20.800 without Network, USB or XDMI cards.

We are offering a pre mounting service, the Olympus I/O then ships to you with a Network card AND USB card OR XDMI card pre mounted. You can then just replace your existing USB and network card by the 2 interfaces cards, connect the I/O and ship your old USB and network card back to us. We are doing this to make the Olympus I/O addition as easy as we can.

So in this scenario, you already own the upgraded USB and network cards, which you return to us AFTER you have installed your Olympus I/O, you pay EURO 20.800.

If you DON'T already own the upgraded USB and network cards the Olympus I/O will be non functional, so you'll have to add those to your purchase. This then adds 2* EURO 1600 for a total of EURO 24.000.

If you ONLY own the USB card OR not the network card, you'll have to purchase the one you're missing and pay EURO 20.800 + 1* 1600 = EURO 22.400

If you don't use USB but have a network connected DAC (there are only a handful of those), you will need 2 network cards to be able to fully utilize the Olympus I/O, this does not change the math outlined above as the price of a network card is the same as the price for a usb card. If you have own network cards you pay EURO 24.000, if you have 1 network card you pay EURO 22.400, if you have 2 network cards you pay EURO 20.800.

If you own an Extreme and want to add XDMI you will have to replace your USB card, so this option adds 15.000 - 1.600 = Euro 13.400 (minus the EURO 5000 introduction discount while it lasts, for Euro 8.400). If you don't own the USB card, you will pay Euro 15.000 (10.000 with introduction discount).

If you add the I/O to your Olympus the math becomes a bit easier, whether you go for USB or XDMI, it will always add EURO 20.800.

What if you have no idea what USB card or Network card you have? Then what is the price?

Asking for a friend...

 

May it be possible that the XDMI - Taiko USB work better then XDMI - AES/EBU?

We designed XDMI to be able to remove USB from the "signal path". I did not look into it but I'm sure there are ways to convert I2S to USB to enable XDMI to have an USB output but that would be wasting what we designed XDMI for, you'd just have a very expensive USB card then of which I doubt it will perform better then the already existing USB card. Either way we're not planning to find out

 

What if you have no idea what USB card or Network card you have? Then what is the price?

Asking for a friend...

Euro 24.000 with USB
Euro 37.400 (temporary Euro 32.400) with XDMI

Your friend can look at the rear of his Extreme, the upgraded cards can be identified by the copper mounting brackets in the Extreme's expansion slots:

 

As I am (trying) to understand, we have the following 4 scenarios to compare and choose from with regards to XDMI digital to analog conversion.

1. Olympus > via USB > DAC.

2. Xdmi digital > via AES > DAC

3. Xdmi digital to analog > direct to Pre amp

4. Xdmi digital > via QSFP > to XDMI board and BPS at DAC?

I seems like Option 4 could be implemented a number of different ways from relatively straight forward to complicated and expensive.

To me, Option 3 is beautiful in it’s simplicity. I would love to go this route.

Option 4 seems to be by far the most complex, ambitious and expensive proposition. Maybe, cost no object, it could be the best?

It will be interesting to see if dac manufactures and Taiko can come up with a practical solution for option 4.

Yes that's correct. Option 4 would be next on the evolutionary ladder.

 

Something I don't understand Emile. If you place the XDMI card on the DAC side in an external box, how do you communicate this box with the DAC, using a “long” i2s cable?

Well yes that way would enable perhaps the shortest possible I2S cable for an external connection option. Obviously we'd like to explore better options for this. From an engineering point of view I2S signal integrity is peanuts compared to links like UPI, PCIe, 100 or 800Gb networking. I2S is 37 years old, designed by Philips in 1986...

 

I am hopeful of that and also realize this is Taiko’s first iteration. I am sure the development isn’t stopping here.

The development will not be stopping here. And is why we went through all the extra effort of delaying XDMI's release to make it modular, so it can be updated by exchanging relatively individual cheap modules.

Technological innovation is now a part of our daily lives.

AES3 is 38 years old
I2S is 37 years old
SPDIF is 35 years old
USB audio is 12 years old

Aka this is all ancient by todays standards. Today's standards don't care much about audiophiles though, who judge innovation with even more ancient tools

 

The only wart I can see is that many folks proclaimed XDMS was “almost ready for release” a year ago. Having spent time in the world of software I thought that was unrealistically optimistic, and here we are still knocking on the door of a beta release.

Along the way, we get snips of code added to a piece of alpha software and is so good we tolerate the pace of development.

writing commercial software is truly an enormous task on its own, but add to that the repetitive sonic evaluation of each new bit of code. I’m sure even the developers get frustrated if their new functions turn out to have negative consequences.

I imagine looking in and watching as the threads add hundred of pages could be daunting; but you are watching real-time development something most manufacturers would never let you see. The trust this creates is unique.

Had we skipped the listening part of the development process it would surely have been at full release status a very long time ago. Enormous credit to Wilson, NSM entailed almost a full rewrite of all XDMS code, all that only for sound quality, it took him months. Not all of us could have mustered up the motivation to do that.

 

Agree! On this, a question on pricing for Emile

So the Olympus and I/O XDMI bundle costs €81,200 with the intro discounts. Assuming you just buy Olympus XDMI upfront for €60,400 (with the discount) and add the I/O at a later time, what would be the price for the I/O (XDMI)?

€81.200 - €60,400 = €20,800 , there are no "bundle discounts".

Also I think you mentioned somewhere that adding the I/O later is a simple change of cards that can be done by the user at home?

Correct.

 

Put more directly, and please anyone correct me if I am wrong, reading between the lines, Taiko Olympus XDMI analogue competes with, if not surpasses DACs in the DCS, MSB, Lampizator best offering realm.

Of course, until comparisons are done, we won’t know. But the hint has not been subtle. And a clear question is what will the value of those hi end dacs be once Taiko olympus releases? Will XDMI compatability on the DAC end reincrease their performance?

Obviously sensitive topics but I do think the first out of the gate re: dac hi end specialists to come forward will instill confidence and help provide some clarity.

I am grappling w this newly created tech (which no doubt is incredible) with no response yet from hi end DAC manufacturers that we all own, so far, as to how to leverage and/or how their product will compete. Obviously a huge factor in any future purchasing decisions. and the future of server/DAC compatability/performance.

Appreciating what Taiko has already delivered w upgraded network and usb cards, switch, router, power distributor, nsm, it’s INCREDIBLE. perhaps that is best for me. and , of course, Taiko pushing the envelope benefits all in the digital space in the mid to longer term. Hats off as always

For absolute clarity:

I don't presume the Analogue output board module, containing the DAC and analogue output stage, to be competitive with current top of the line DACs on it's own. It's well designed, but it's "basic". We spend perhaps 1-2% of the entire design effort on it.

The performance comes from the interface structure, the entirety which makes up XDMI, software, firmware and all the bits and pieces of hardware involved.You have heard NSM, you have heard what changes in code alone can do. Now think a complete restructuring of code working optimally in tandem with simultaneously designed hardware.

The analogue output board option just allows you to hear the full potential of XDMI, as using 35-40 year old technologies like AES and SPDIF to interface XDMI to your existing DAC sure does impose some limits. But still I'm convinced, even when using it this way, that you will still hear a sizeable chunk of what XDMI brings to the table over USB, it will be a significantly larger uptick then NSM on it's own. And if a 24/192 sample rate limit does not bother you it's already really a very substantial upgrade IMHO. If we manage to find better solutions then AES / SPDIF to interface with other manufacturer DACs, that would be very exciting! But of course we will not have that bit with a low volume market adaptation. That's how things work in commerce. Time will tell but a few manufacturers have shown true intent to work on it with us which is a very promising start.

 

I can already tell I am becoming an unpopular poster here despite 4 year support for the Extreme thru every one of its iterations. and owner of its incredible components to date. the taiko thread can sometimes feel like only a very positive self enforcing circle with no doubt full truth to it. but can alienate others from joining the conversation as a result as i hear from a number of non participants that watch this thread.
and referencing my post above for any insights from taiko, msb, dcs or lampizator

Please do keep sharing your honest views and opinions! Honesty is what serves everybody best in the long run. Much appreciated!

 

Yes that's correct. Option 4 would be next on the evolutionary ladder.

That is my secret plan.
Get Olympus XDMI now and see if TDT
(Taiko Dream Team) will improve 30% difference to higher level and get the Olympus IO with that OR earlier if my remaining funds will allow.

The buty of geeting Taiko is a top luxury to never loose money on next investment AND being able to return it if you don't like it , that fortunately never happens.

 

@Taiko Audio I think much of our concerns/hesitation/consternations/ for those of us with External DACS is the "not knowing aspect". For Lukasz/ Lampizator, Will it be an external device albeit AES/I2S to "bridge" XDMI? Will our current DACS be retrofitted internally? Will a new line of Horizon DACs come with XDMI Capability? I realize these are not questions for you to answer but by proxy Taiko is part of this. I really think this is the most important underlying aspect of this. All of us are onboard for XDMI, its the initial implementation with XDMI with external DACS...

 

@Taiko Audio I think much of our concerns/hesitation/consternations/ for those of us with External DACS is the "not knowing aspect". For Lukasz/ Lampizator, Will it be an external device albeit AES/I2S to "bridge" XDMI? Will our current DACS be retrofitted internally? Will a new line of Horizon DACs come with XDMI Capability? I realize these are not questions for you to answer but by proxy Taiko is part of this. I really think this is the most important underlying aspect of this. All of us are onboard for XDMI, its the initial implementation with XDMI with external DACS...

If you want to adopt XDMI now you can use SPDIF or AES/EBU. If you want to wait for an alternative method of connection, wait

I suspect mounting the entirety of XMDI in a remote DAC chassis may have space restraining issues for some.

Then there's I2S, should be easy to implement.

Or we just design something new, perhaps adopt our SFP/QFSP technology, with a small remote daughterboard which can replace for example the jlsounds USB board, used by many manufacturers, let's say a better way to transport I2S. That would be similar to creating something like MSB Pro ISL or Wadax AKASA, but then replacing USB on the server side with XDMI.

 

If you want to adopt XDMI now you can use SPDIF or AES/EBU. If you want to wait for an alternative method of connection, wait

I suspect mounting the entirety of XMDI in a remote DAC chassis may have space restraining issues for some.

Then there's I2S, should be easy to implement.

Or we just design something new, perhaps adopt our SFP/QFSP technology, with a small remote daughterboard which can replace for example the jlsounds USB board, used by many manufacturers, let's say a better way to transport I2S. That would be similar to creating something like MSB Pro ISL or Wadax AKASA, but then replacing USB on the server side with XDMI.

Perhaps I'm thinking more into this than need be! But that's kind of what we do! The SFP/QFSP Technology is what I was clumsily trying to allude to...We would then utilize XDMI to its full potential. Right? Thank You for explaining this...

 

Or we just design something new, perhaps adopt our SFP/QFSP technology, with a small remote daughterboard which can replace for example the jlsounds USB board, used by many manufacturers, let's say a better way to transport I2S. That would be similar to creating something like MSB Pro ISL or Wadax AKASA, but then replacing USB on the server side with XDMI.

I would say that's it