Emile is the only one who can tell us if his developments used mainly trial and error or science, but I would bet he used a lot of science and theory in his research ...
Taiko Audio SGM Extreme : the Crème de la Crème
- Thread starter CKKeung
- Start date
Emile is the only one who can tell us if his developments used mainly trial and error or science, but I would bet he used a lot of science and theory in his research ...
I’d argue that trial and error is a big part of science. There’s theory versus practical application and result. Application of theory does often not lead to the expected result. Then you have to investigate your application, run a few more experiments to see if you can have your result match theory. And then there is the annoying, but infinitely interesting, “does it sound better”. If it does or not is not even that important as it adds datapoints which lead to new theories and experiments.
Musical perception by the human body is not that well investigated relative to other fields of research. It is just not that interesting for large companies with significant research budgets as there is much more money to be made in other fields. Discoveries are usually not published unless a patent can be applied in markets of interest. Most high end audio companies only have 1 or 2 researchers which have to share their time with other duties. So relatively speaking, research moves painfully slow in our field.
There is a lot of research budget for food. That's why people have figured out accurately what combination is carbs, fats, and proteins to eat and have recommended a steady diet since the 80s to guard against heart issues etc. Same with audio, we know exactly what to do
Actually there are more changes to advice around diets in the last 20 years than you can possibly believe.
The issue with the science around “diets” is that the experimental modality (typically non-randomised, observational research) is highly prone to confounding. The most scientifically robust means of robustly determining such matters is via randomised controlled trials with *very* large sample sizes. Unfortunately for us all, public health (under which diet falls) is not very well funded versus drug discovery.
My post was sarcastic that with all that research spend if people keep changing the recommendations for diet what hope is there for them to get audio right based on research, given that trials population for audio is so much smaller in terms of sample size
You should give up your UK citizenship if you can't get sarcasm except that you back the team that cheated itself to the cricket World Cup
My post was sarcastic that with all that research spend if people keep changing the recommendations for diet what hope is there for them to get audio right based on research, given that trials population for audio is so much smaller in terms of sample size
You should give up your UK citizenship if you can't get sarcasm except that you back the team that cheated itself to the cricket World Cup
Ked - you missed the main point. There isn’t anything like “the research spend” you think hence the essence of my entire post.
In terms of cricket - well there are bad losers everywhere...
As soon as a series of bits is moved from one place to another, everything becomes very different compared to 'stored bits' on a HDD or SSD. It becomes dynamic, involve timing and jitter and high bit rate switching On & Off is a source of pollution.
In term of reducing "noise" and unwanted perturbations in a high-end audio server, isn't it interesting to reduce CPU and RAM supply voltage (by 5% or 10%) ?
Also, reducing RAM's running speed is usually beneficial for a better SQ (at least to my experience on a couple of audio server/NUC).
Having more CPU Cores running in parallel is not always the best configuration. I have a 6x Cores INTEL and prefer to only use 4x Cores.
Some people add EMI/RFI sheet material at some locations of the mother board. Is this something TAIKO Audio, or other guys here, believe into ?
Every noise source or pollution matters, I guess, in real time audio decoding and streaming.
In term of reducing "noise" and unwanted perturbations in a high-end audio server, isn't it interesting to reduce CPU and RAM supply voltage (by 5% or 10%) ?
Also, reducing RAM's running speed is usually beneficial for a better SQ (at least to my experience on a couple of audio server/NUC).
Having more CPU Cores running in parallel is not always the best configuration. I have a 6x Cores INTEL and prefer to only use 4x Cores.
Some people add EMI/RFI sheet material at some locations of the mother board. Is this something TAIKO Audio, or other guys here, believe into ?
Every noise source or pollution matters, I guess, in real time audio decoding and streaming.
Hello Cellindo,
I would recommend to not draw conclusions to fast but to view it as a total system concept. If you change the amount of cores used and/or adjust voltages/clock speeds, you should consider retuning your operating system parameters and power supply design to adjust for the change in environment and make the most out of your changes. At the end of the day it is a sum of all parts.
The amount of combinations is near infinite. Years of accumulated knowledge on combinations of settings and their effects can shorten time spend on tuning a system, but it will still take considerable time.
To take your post as an example, if you prefer only using 4 out of 6 cores, and subsequently change operating voltages/clockspeeds, do you still prefer using 4 cores? What if you change process allocations in your operating system to make better use of more cores? What if you adjust your power supply and/or change your EMI/RFI filtering parameters? If you cover motherboard components with EMI/RFI absorbing material, you may find yourself preferring a whole different range of settings again.
I would recommend to not draw conclusions to fast but to view it as a total system concept. If you change the amount of cores used and/or adjust voltages/clock speeds, you should consider retuning your operating system parameters and power supply design to adjust for the change in environment and make the most out of your changes. At the end of the day it is a sum of all parts.
The amount of combinations is near infinite. Years of accumulated knowledge on combinations of settings and their effects can shorten time spend on tuning a system, but it will still take considerable time.
To take your post as an example, if you prefer only using 4 out of 6 cores, and subsequently change operating voltages/clockspeeds, do you still prefer using 4 cores? What if you change process allocations in your operating system to make better use of more cores? What if you adjust your power supply and/or change your EMI/RFI filtering parameters? If you cover motherboard components with EMI/RFI absorbing material, you may find yourself preferring a whole different range of settings again.
As you said..., the number of configuration is indeed way too high to imagine to test everything in depth !!!
That's why the lucky ones who can afford the very best computer audio servers get tuning/listening/design experience from people like you.
I hope one day I will have the chance to audition the SGM Extreme. ;-)
That's why the lucky ones who can afford the very best computer audio servers get tuning/listening/design experience from people like you.
I hope one day I will have the chance to audition the SGM Extreme. ;-)
Audition and buy it just after......
I think If you own Vivaldi and Rockport you should consider getting extreme to your system for good.
I went via many DIY servers over past 7 years.
I was so tired of chenging and testing I had to quit for 5 years.
Now I am trying to get Taiko Extreme because there is no risk with their return policy .
You should consider doing the same
Yep, these days I get sometimes really bored trying all and every possible settings on the BIOS, Squeeze2UPnP software, Network card parameters, Fidelizer, power supplies etc...
The SGM EXTREME is out of my reach at this time. Maybe I will try some other good commercial products such as Pink Faun, Antipodes, etc...
Nice to read that we seem to agree that considering the extreme complexity of the problem the accumulated empirical knowledge is not enough to bring an optimal solution ...
I tested SR Black, Blue, Audio Magic Beeswax Super, Ultimate, Power House Zero,
and Brimar Graphene Source Fuse with SGM Extreme.
All fuses are 20mm / Slow / 2A.
Tested fuses have specific strong characters and all is awesome fuses.
But their characters don't always make SGM Extreme better sounds.
In Korea, SGM Extreme's smash hit's reason is that Extreme sound is very close to vinyl sound.
Of cause you seem to be same with us.
Anyway, SGM Extreme needs the fuse that makes Extreme more analogue
and higher resolution sound.
Upper 6 fuses all can make Extreme higher resolution and more dynamic sounds.
But only Ultimate, Zero, Graphene Source fuses make more analogue sounds like vinyl.
My final result is like below :
- Beeswax Ultimate : More Analogue : 8.5/10 // More HI FI : 6/10
- Zero : More Analogue : 7/10 // More HI FI : 9/10
- Graphene Source : More Analogue : 8/10 // More HI FI : 8.5/10
Price : Graphene Source > Zero > Ultimate
My choice : Brimar Graphene Source Fuse
This is my subjective opinion.
My cables are all Entreq Flagship sets.
These cables are flat sound.
In my case, Brimar Graphene Source Fuse made Extreme be great upgraded.
Because your system and your wish are very different, I can't recommend
Brimar Graphene Source Fuse, but this fuse is obviously fantastic fuse in the world.
I love this fuse and will order Power Fuses for my pre & power amps.
and Brimar Graphene Source Fuse with SGM Extreme.
All fuses are 20mm / Slow / 2A.
Tested fuses have specific strong characters and all is awesome fuses.
But their characters don't always make SGM Extreme better sounds.
In Korea, SGM Extreme's smash hit's reason is that Extreme sound is very close to vinyl sound.
Of cause you seem to be same with us.
Anyway, SGM Extreme needs the fuse that makes Extreme more analogue
and higher resolution sound.
Upper 6 fuses all can make Extreme higher resolution and more dynamic sounds.
But only Ultimate, Zero, Graphene Source fuses make more analogue sounds like vinyl.
My final result is like below :
- Beeswax Ultimate : More Analogue : 8.5/10 // More HI FI : 6/10
- Zero : More Analogue : 7/10 // More HI FI : 9/10
- Graphene Source : More Analogue : 8/10 // More HI FI : 8.5/10
Price : Graphene Source > Zero > Ultimate
My choice : Brimar Graphene Source Fuse
This is my subjective opinion.
My cables are all Entreq Flagship sets.
These cables are flat sound.
In my case, Brimar Graphene Source Fuse made Extreme be great upgraded.
Because your system and your wish are very different, I can't recommend
Brimar Graphene Source Fuse, but this fuse is obviously fantastic fuse in the world.
I love this fuse and will order Power Fuses for my pre & power amps.
Attachments
Did you compare it with master tape sound? IMHO it is a much more solid and universal reference, although most of us have only a limited number of high quality tapes.
As my anticipation, no one has tape playback devices.
In Korea, to get high quality tape sources is impossible I think.
Actually, a few Korean audiophiles have high quality turntables.
Almost Korean audiophiles have not heard real high quality vinyl sounds.
I'm one of those. Ha Ha
Hello Esotar,
You and I have the same conclusion!
Thanks for introducing great fuse~!!
Because SGM Extreme can express the tiny differences, many WBF members will agree my result.
But there is no absolute HI FI products, so all members will not order Brimar flagship fuse.
I love Brimar~!!
Now, my Brimar flagship fuse was in the SGM 2015 EVO.
EVO is super fantastic digital source device,
but EVO users must do various settings with Emile.
In this case, Brimar flagship Source fuse makes EVO much more easily set.
Brimar Graphene Source fuse makes more fantastic texture and higher resolution.
Brimar Graphene Source fuse will be more value to use it in SGM 2015 EVO.
Very Fantastic Fuse.
EVO is super fantastic digital source device,
but EVO users must do various settings with Emile.
In this case, Brimar flagship Source fuse makes EVO much more easily set.
Brimar Graphene Source fuse makes more fantastic texture and higher resolution.
Brimar Graphene Source fuse will be more value to use it in SGM 2015 EVO.
Very Fantastic Fuse.
