Taiko Audio SGM Extreme : the Crème de la Crème

Taiko Audio

Industry Expert
Feb 10, 2017
4,233
13,024
1,925
The Netherlands
taikoaudio.com
@Taiko Audio

i know windows can be a little strange sometimes.
can the extreme read an EXT3 or EXT4 HDD on the usb3 input?
all my computers/systems are linux and i'll want to copy all my music over
to internal storage on the extreme.

once the music is copied over the first time, i'll plan to keep my music and backup drives in sync
using BeyondCompare. i don't expect any problems there as it seems to handle everything i throw at
it on my home network.

It does not, you would need to install a 3rd party driver / utility for that. It can be removed again which I would recommend as it'll have some audible consequences.
 
  • Like
Reactions: M27HF

Taiko Audio

Industry Expert
Feb 10, 2017
4,233
13,024
1,925
The Netherlands
taikoaudio.com
For those interested in the Taiko DIY initiatives, you may want to take a look at this post:


The idea here is for this to be a non commercial community project, we supply a few quality building blocks which are currently not available in the DIY market. Support and further development would have to come from the DIY community, our input would be limited to general building guidelines and some tips & tricks.
 

shaned

Well-Known Member
Nov 7, 2019
66
26
83
51
I have a 1Gbps internet connection, faster does have some audible advantages when using Roon, we have a few customers with access to a 10Gbps connection reporting another step up. With TAS however the benefit is limited to reduced track loading times.
Would a faster internet connection also help with the Evo/Evo Light although I’m guessing because the Extreme is on another level to the Evo it might not make as much difference to the Evo as it does to the Extreme. Qobuz user here too.
 

oldmustang

Well-Known Member
Dec 1, 2012
1,057
3,124
1,445
Steve.
Please get a better internet Connection if you can.
I was on 15 mbps than 30 mbps, 50, 100, 200 .
Unfortunate the faster the better for extreme.
As far as I know Emile tested upto 10 Gbps.
I'd love to get something faster if I could, Kris. But this is the only service other than data-capped satellite internet that is available to my house. Ironic that in a semi-rural area almost in sight of Seattle, WA -- one of the major US tech centers -- DSL is the best that can be done. Cable goes past the end of my private road, but we're almost a mile away from that. So until 5G internet for the home is a reality, this is it.

Steve
 

cat6man

Well-Known Member
Feb 6, 2013
890
995
1,155
west of NYC, east of SF
I'd love to get something faster if I could, Kris. But this is the only service other than data-capped satellite internet that is available to my house. Ironic that in a semi-rural area almost in sight of Seattle, WA -- one of the major US tech centers -- DSL is the best that can be done. Cable goes past the end of my private road, but we're almost a mile away from that. So until 5G internet for the home is a reality, this is it.

Steve
good luck steve!

(a little off topic, so if you have any further questions just message me)

i doubt 5G internet can be your savior. there are two flavors of 5G wireless,
1. one that is a little faster than today's 4G/LTE (using roughly similar frequency bands and can reach customers at about the same distance)
and
2. a much faster but much higher frequency version using millimeter wave band.

option 1 would not be much different than what AT&T offers in their home internet wireless today, about 25mbps if you put up an outdoor antenna, connected to their wireless indoor router. currently they have excess charges if you use more than about 1Gb/day. also, good luck with QoS (Quality of Service, aka latency control) over that type of channel.

option 2 could actually be useful to you but has a couple of limitations. first of all, the millimeter wave band has a very short range of service and is highly impacted by foliage and rain and typically must have a clear line of sight. therefore, it must have an antenna quite close to your house to serve your immediate neighborhood, and to do that they typically run fiber and place distributed antennas along the way. the advantage to them is that they don't need to provide a service call to the home for intallation, you would just put a box in a window that is LOS. in your situation though, if they don't have a cable that runs by your house, they won't have anything to connect their 5G pico-cell to, so you're out of luck.

option 3 is that someone will come up with a buffering scheme and parameter settings that will obviate the worst effects of lower bandwidth on SQ.
 
Last edited:
  • Like
Reactions: oldmustang

oldmustang

Well-Known Member
Dec 1, 2012
1,057
3,124
1,445
Well, maybe the worlds richest man, Elon Musk will help me out. . .

Steve Z
 

cat6man

Well-Known Member
Feb 6, 2013
890
995
1,155
west of NYC, east of SF
Well, maybe the worlds richest man, Elon Musk will help me out. . .

Steve Z

still think your best bet is option 3, unless Elon lives on your street, in which case i'm sure you'll have very good service soon
:)
 

heebrog

Well-Known Member
May 12, 2018
131
51
135
Perth, WA
I can independently confirm that Steve has slow internet (sorry Steve ;)

@Koegz Are you using HQP from Roon? I've had issues with some filters in HQP where if you have a sampling rate change, track transitions don't always go well
Hi Ray,

Is the problem with track transitions an issue when using HQP with TAS?
 

ray-dude

Well-Known Member
Dec 8, 2019
107
364
150
54
Yes Geoff, if you’re using HQP for upsampling using particular filters. Some of the filters require a long setup time (buffers, etc) so the pipelines have to get set up again (dependent on filter). Also, the setup time is different for different filters, so it may only be noticeable in certain cases as well

In practice I only have the issue on playlists with mixed content or when switching albums. Within an album, almost always sample rates are the same
 

matthias

Well-Known Member
Mar 14, 2019
1,226
545
213
Germany
Hi Emile,

congrats, the Taiko DC-ATX SMPS got a rave review on AS :)

Would an ultra high specs version of this SMPS make sense for use in the Extreme for demanding upsampling procedures or simply for SQ?

Thanks

Matt
 
Last edited:

oldmustang

Well-Known Member
Dec 1, 2012
1,057
3,124
1,445
I’ve had the opportunity to do some listening comparisons (still going on BTW) between streaming and local content since Wilson installed the latest Emile/Wilson updates to TAS alpha v2 this morning. I would say that I don’t think I can now reliably distinguish between streaming and local playback. The more remarkable, that this is with my very slow “high speed“ internet connection.

I’d also offer that these new changes also raise the overall sound quality of TAS playback, regardless of source.

Chapeau to Emile and Wilson!

Steve Z
 

Taiko Audio

Industry Expert
Feb 10, 2017
4,233
13,024
1,925
The Netherlands
taikoaudio.com
Hi Emile,

congrats, the Taiko DC-ATX SMPS got a rave review on AS :)

Would an ultra high specs version of this SMPS make sense for use in the Extreme for demanding upsampling procedures or simply for SQ?

Thanks

Matt

Hi Matt,

Running out of time today, will reply in depth tomorrow.
 
  • Like
Reactions: matthias

cat6man

Well-Known Member
Feb 6, 2013
890
995
1,155
west of NYC, east of SF
I’ve had the opportunity to do some listening comparisons (still going on BTW) between streaming and local content since Wilson installed the latest Emile/Wilson updates to TAS alpha v2 this morning. I would say that I don’t think I can now reliably distinguish between streaming and local playback. The more remarkable, that this is with my very slow “high speed“ internet connection.

I’d also offer that these new changes also raise the overall sound quality of TAS playback, regardless of source.

Chapeau to Emile and Wilson!

Steve Z

congrats steve

wow, i didn't think option 3 was *that* close to realization.
theoretically, there is no reason why sufficient bandwidth cannot be as good as tons of bandwidth,
but it takes both excellent engineering chops and artistry to make it happen.

bravo Taiko
 

cat6man

Well-Known Member
Feb 6, 2013
890
995
1,155
west of NYC, east of SF
question for our advance scouting team of alpha testers, particularly with respect to TAS alpha2:

how sensitive is the SQ, with alpha2, to the network mods we are/were all making with smart switches, isolated audio path/VLANs, etc?

question for emile:

does TAS make the extreme more or less sensitive to network mods and does it still benefit more or less from the new USB card and switch?
i'm sure combination of TAS, usb card and switch will be awesome but i'd be interested in your thoughts/observations on the interactions and behavior of the combos wrt SQ (in other words, can i expect to get rid of much of my spaghetti and small boxes and LPSs in the network)?

stay safe everyone!
 
Last edited:

oldmustang

Well-Known Member
Dec 1, 2012
1,057
3,124
1,445
congrats steve

wow, i didn't think option 3 was *that* close to realization.
theoretically, there is no reason why sufficient bandwidth cannot be as good as tons of bandwidth,
but it takes both excellent engineering chops and artistry to make it happen.

bravo Taiko
Me either!

Steve Z
 

oldmustang

Well-Known Member
Dec 1, 2012
1,057
3,124
1,445
question for our advance scouting team of alpha testers, particularly with respect to TAS alpha2:

how sensitive is the SQ, with alpha2, to the network mods we are/were all making with smart switches, isolated audio path/VLANs, etc?

question for emile:

does TAS make the extreme more or less sensitive to network mods and does it still benefit more or less from the new USB card and switch?
i'm sure combination of TAS, usb card and switch will be awesome but i'd be interested in your thoughts/observations on the interactions and behavior of the combos wrt SQ (in other words, can i expect to get rid of much of my spaghetti and small boxes and LPSs in the network)?

stay safe everyone!

I can't really answer the question for you because in the interest of only evaluating one variable at a time -- TAS -- as well as the very selfish desire not to do *anything* that could possibly lower the superb sound quality, I haven't changed anything in my LAN since TAS has been installed. If it isn't broken. . .

My setup is rather simple compared to many:

ActionTec 1900C DSL modem/router > Sablon ethernet jumper > OpticalModule > Finisar 1324 SFP > fiber > Finisar 1324 SFP > Extreme

All other items in my home are connected to the DSL modem/router via ethernet jumper > StarTec FMC > fiber > StarTec FMC > ethernet jumper > Netgear GS105 switch > rest of house.

Uptone JS-2 power supply, Sonore Signature power supply and Jameco regulated LPS wall warts variously power the modem/router, Optical Module, GS105 switch and FMCs.

Steve Z
 
  • Like
Reactions: EuroDriver

EuroDriver

Well-Known Member
Sep 16, 2015
926
2,479
450
Monaco
question for our advance scouting team of alpha testers, particularly with respect to TAS alpha2:

how sensitive is the SQ, with alpha2, to the network mods we are/were all making with smart switches, isolated audio path/VLANs, etc?

question for emile:

does TAS make the extreme more or less sensitive to network mods and does it still benefit more or less from the new USB card and switch?
i'm sure combination of TAS, usb card and switch will be awesome but i'd be interested in your thoughts/observations on the interactions and behavior of the combos wrt SQ (in other words, can i expect to get rid of much of my spaghetti and small boxes and LPSs in the network)?

stay safe everyone!
Hi cat6man,

let me chime in with my observations from 10,000 feet

- TAS with its lower noise floor makes network activity and its associated noise more audibly discernable

however

- TAS software architecture and process layout design minimizes network activity during music playback, which makes the playback sound quality less sensitive to the quality of the network infrastructure

Extreme with TAS can deliver stunning results with modest network infrastructure which we are hearing with the tweaked Alpha V2

Good News for Beta Testers in waiting, we have locked the Beta code changes and are now working on the distribution infrastructure to allow for a smooth and quick installation
 

matthias

Well-Known Member
Mar 14, 2019
1,226
545
213
Germany
Hi Matt,

Running out of time today, will reply in depth tomorrow.

Thanks Emile,
please forget the "demanding upsampling procedures", the focus is "simply for sound quality" :)

Matt
 

Taiko Audio

Industry Expert
Feb 10, 2017
4,233
13,024
1,925
The Netherlands
taikoaudio.com
Thanks Emile,
please forget the "demanding upsampling procedures", the focus is "simply for sound quality" :)

Matt

Hi Matt, answer divided over 2 posts, was unaware there is a 10.000 character limit!

Thank you for raising the question, it is a very interesting topic which deserves an elaborate reply. For audio equipment power supplies we typically have 2 general mechanisms for which you have a choice between switching or linear solutions. The first is AC to DC conversion, the second is DC-DC conversion (regulation).

What is typically referred to as a SMPS (Switch Mode Power Supply) or LPS (Linear Power Supply) mainly addresses AC-DC conversion, now you could argue both LP and SM power supplies are switchers, a LPS switches at 50/60Hz or 100/120Hz (depending on the rectification circuit), with a SMPS you can choose/design the switching frequency, this is typically 50KHz or higher. A SMPS is always a more complex design, with a much higher component count, yet it can be cheaper as you can design a complete SMPS at a total cost of below that of just a mains transformer, however for audiophile applications you would want to use "audiophile quality" components which could make it much more expensive. It can also be more efficient, however you can also have a very efficient AC-DC LPS, as you can have very efficient rectifiers (active rectification for example) and toroidal transformers can have efficiencies up to 96%, combined that type of LPS rivals a SMPS for AC-DC conversion efficiency, leaving you with 2 factors in play, switching frequency and its associated noise spectrum on both the AC and DC side, and component quality which also affects the noise spectrum of the supply.

So far we've only been addressing AC-DC conversion, which turns an AC voltage into an unregulated DC voltage, this is typically not a constant voltage, it varies with AC grid voltage fluctuations and current draw. This unregulated DC voltage can be used to directly supply power to amplifier output stages, however we also need fixed voltages for which we need voltage regulation. Here we can again choose between linear and switching regulator solutions. Unlike AC-DC conversion, for DC-DC regulation efficiency there is no comparison between linear and switching regulator designs.

As an extreme example, let's look at powering a CPU. A CPU typically operates at 0.6-1.2 volts but can easily draw 100 watts or more, let's assume 1 volt for easier calculations. For 100 watts at 1V you are talking about 100 Amps(!) of current. As current drops voltage over resistance (Ohm's law), a CPU is typically supplied with a 12V voltage rail, so we need to regulate 12V down to 1V. If you would use linear regulation, and you'd have a 100 Amp current draw at 1V, you would also have a 100 Amp current draw at 12V (simplified), meaning 12V*100A=1200 watts. Then we need to feed this 12V by a 16-19V supply (again to account for voltage drop caused by current over resistance), let's assume 19V as that's a very common value to ensure broad compatibility, and we are talking 19V*100A=a shocking 1900 watts. The conversion efficiency here would be ~5%, the other 95% will just be converted to heat. A switch mode regulator is the inverse of this, it can regulate 19V down to 1V at a 95% efficiency wasting only 5% as heat. That is a 1900W versus 105W of power consumption.

As an hybrid solution, you could opt to perform the 12V to 1V conversion switched mode, and 19V to 12V linear. With 100A at 1V, and a regulation efficiency of 95%, your current draw at 12V would be about 8.8A (100*1*95%/12). You would then have about a 9A current draw (again simplified not accounting for resistive and other losses) at 19V, translating to a ~170W power consumption. Then you end up with a ~60% efficiency, burning about 70W in heat. Now you could stretch this a bit with a careful design, bring it down to say 16V, increasing efficiency to about 70%, and that is in fact a very typical efficiency for a carefully designed linear regulation circuit to serve a particular application. If you need multiple voltages, which you virtually always do, this would necessitate multiple unregulated DC power rails, meaning multiple transformers or a transformer with multiple secondaries, multiple rectification and DC filter stages. This is also where you need to start being really careful with your layout and circuit design as these rails can be supplying individual but interconnected circuits (creating loops), which is where for example your ground layout become a critical component affecting performance. RFI does not seek the shortest path to earth ground as is often assumed, it just travels everywhere it can (think antenna). In this scenario there is a potential benefit for an AC-DC SMPS as it's relatively easy to create multiple unregulated DC rails from a a single small footprint supply.

So where does this all of this leave us in our considerations to use switching or linear power supplies for audiophile purposes?

Approaching this question from a noise perspective, assuming a system's accumulated noise level directly affects sound quality, we propose the following aspects:

-higher currents increase noise
-vibration increases noise, higher currents increase vibration
-higher heat increases noise, where higher currents increase heat
-components produce noise, where we have their own operational noise plus additional noise caused by the 3 listed above
-circuit layout produces noise and is affected by it, by component parameters and circuit design including pcb traces, wiring, radiated noise and their interaction

Ideally we'd have the lowest possible currents flowing, low heat, high component quality and the best possible environment to allow these components to operate at their intended parameters or in their optimal range in order to lower distortion, and if possible a simple layout using a minimal component count where each of the components used can be of the highest possible quality with the lowest possible generated noise spectrum/levels.

Switching vs linear benefits:
-Higher efficiency, lower power waste therefor lower current flows and lower heat, DC-DC conversion efficiency also allows running higher voltage rails reducing currents even further for the same power requirements (as power (W) = I (current) * V (voltage) and therefor a lower vibration environment.

Linear vs switching benefits:
-Simplified circuit design, lower component count and a wide availability of audiophile grade (low noise and lower vibrations sensitive) components.

From a cost perspective:

For AC-DC applications: Linear design complexity is lower then switching, you typically need fewer parts, but we tend to spend more on individual parts quality in linear supplies and they are physically much larger. Design cost of linear circuits is less advanced, less complex and therefor (much) lower, the skill level requirements for designing switching mode supplies is magnitudes of orders higher then those for linear supplies. This nets out to a lower cost for Linear supplies for small quantity manufacturing, although the component and housing costs are higher, the development costs are much lower. For high volume manufacturing switch mode is cheaper with development costs divided by sales numbers.

For DC-DC applications: Switching costs are always higher when using high quality parts, the component count is higher and the design costs are higher. However this may not apply when attempting to regulate voltages in high current designs if you are going to waste dozens of watts in heat due to physical size and heatsinking requirements.

Reading back you may now think why are we using linear power supplies at all, but we have to refine that, as for AC-DC applications I currently do not see considerable advantages to using switch mode supplies for audiophile applications with the exception of lower costs for higher volume manufacturing and reduced space requirements. For DC-DC applications it depends on current needs, with some very obvious advantages for switching regulators for high current applications, in my book, high current is everything over about 1A, actually around 500mA is already borderline where you start running into issues with heat dissipation/power waste and associated noise.
 
Last edited:

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing