Jump to content


  • Content Count

  • Joined

  • Last visited

  • Country

    United States

About luisma

  • Rank
    Junior Member

Recent Profile Visitors

2259 profile views


  1. Hqplayer builds
    HQPlayer Linux Desktop and HQplayer embedded
    18 minutes ago, lmitche said:

    Just to be clear:


    Are you saying there are four different builds of Hqplayer embedded?


    1) CPU : AMD :GPU: AMD

    2) CPU : AMD :GPU: Nvidia

    3) CPU : Intel :GPU: AMD

    4) CPU : Intel :GPU: Nvidia




    1. Regular Ubuntu build with support for Nvidia and AMD GPUs (Intel specific optimizations)
    2. Ubuntu build optimized for latest AMD CPUs
    3. HQPlayer OS build for x64
    4. HQPlayer OS build for RPi4
    5. Debian Buster build for x64
    6. Debian Buster build for arm64
    7. Fedora build for x64


  2. HQP dac bits
    HQ Player
    9 hours ago, asdf1000 said:


    Measurement looks good though for ARES II :


    https://www.audio “science” review/forum/index.php?threads/denafrips-ares-ii-usb-r2r-dac-review.11166/


    in HQPlayer settings for PCM upsampling (to PCM ~1500 kHz) would you set "bits" to 18 bits? Based on:




    If using PCM mode (like in the test), I would set "DAC Bits" to 16 or 17 with LNS15 noise shaper. This keeps linearity in the very flat range, while providing 20+ bits worth of practical dynamic range in audio band.


  3. Noise shaping 01
    Any experiences with RME ADI-2 Pro DAC?
    12 hours ago, Em2016 said:

    Hi Jussi, the HQP manual states:


    "Gauss1 - Gaussian Probability Density Function. High quality flat frequency dither recommended for rates at or below 96 kHz where noise-shaping is not suitable."


    So is Gauss1 actually ok for 705/768 HQP PCM up-sampling?


    Yes, it is just flat dither like TPDF. All those can be always used at higher rates too. Flat dithers can be used at any rate. Shaped dithers are only good at higher rates where there is enough ultrasonic space where to "park" the noise moved away from lower frequencies.


    Theoretically, let's say we move 20 dB of noise from 0 - 20 kHz region up to 20 - 60 kHz region, the noise increases 10 dB in that twice as wide ultrasonic band while it drops 20 dB in the base band region.


    There are noise shapers that shovel the noise within the base band (used for RedBook mastering, etc), but I'm not so huge fan of those.


    When output is 32-bit, there's so much headroom already compared to analog world that using noise shaping doesn't make much sense. Although it is of course still possible to use.


  4. Adi-2 settings 01
    Any experiences with RME ADI-2 Pro DAC?
    11 hours ago, darkmass said:


    It seems to me that you have a decent setup, and are using it well.


    However, if you'd like to get bogged down in some technical analysis of facets of the RME ADI-2 Pro, this page will certainly burn significant amounts of your time.  :)


    I didn't have time to read through the story. But couple of notes:

     - Use external upsampling if you want to improve on the digital filter performance, ADI-2 Pro accepts up to 768k inputs

     - When measuring or using DSD output, remember to enable the DSD Direct mode (you have then two analog filter modes, 50 kHz and 150 kHz)

     - For comparable measurements set PCM volume to -3.5 dB, then the output levels between PCM and DSD Direct match


    If you use RMAA, use the ASIO driver. Other outputs go through the Windows mixer/rate-converter.


  5. Adi-2 HQP settings
    Any experiences with RME ADI-2 Pro DAC?
    Miska, what is your proposing settings to best quality sound in HQPlayer?


    If you use PCM input, just use the 705.6/768k input rate (and TPDF or Gauss1 dither) and your favorite filter. The digital filter selection at the DAC doesn't matter because at these rates it runs with the digital filter bypassed.


    For DSD256 inputs, enable the Direct DSD mode and set the DSD analog filter to 150 kHz. Your choice of filter and modulator. I use poly-sinc-short-mp and ASDM7, but I know some people like DSD7. However, this will mute the PH1/2 output because volume control disappears. PH3/4 will remain active because the DAC chip driving it cannot be switched to Direct DSD mode. So you'll need an external headphone amplifier. I'm using Fostex HP-A8C for the purpose, but I'm planning to get a Schiit Ragnarok (I need more inputs than one analog set)...

  6. New Seasonic
    HQPlayer4 EC modulator tips and techniques
    4 minutes ago, craighartley said:

    Thanks. Yes I have the Define R6 case too. What case fans do you use?


    It is the silent Fractal Design ones that come with the case. Same ones are also available separately. Noctua has also quiet PWM controlled fans.


    Seasonic has nice fanless PSU:



    The 1000W PSU I have is this:


  7. Power supplies and Fans for Server
    HQPlayer4 EC modulator tips and techniques
    22 minutes ago, luisma said:

    Hi Jussi, I now you are using Noctua (and some other brand too) for the CPU which are very quiet, for the case fans what model / brand do you use? PSU fan? thought you were using the fanless Seasonics


    These ones:




    And especially this one:



    30 minutes ago, luisma said:

    PSU fan? thought you were using the fanless Seasonics


    See my post above for the PSU's, the 1000W model (and the lower power version of same series) are hybrid and include fan that can be configured (with back panel push button) to run only when needed.


  8. Hqplayer filters StreamFidelity
    HQ Player

    I think the filter selection is great, but also confusing. I tried to use a systematic approach. Corrections are welcome.


    The results refer to an Audio PC with high computing power (Intel OctaCore i9-9900K, 3.6GHz - 5.0GHz) and with Windows 10 Pro operating system optimized for low latencies. Only the HQPlayer runs on it and nothing else.


     It should be noted that the high requirements are due to the conversion from PCM to DSD. If you only use the filters for PCM, you will need significantly less computing power! Some filters cannot be used in my system (colored orange).


    My current favorites (this can change again and is very dependent on personal taste) are marked in blue. A good solution in HQPlayer is that another filter can be stored for HiRes. For HiRes, I use an MQA filter that I like (poly-sync-mqa-lp). Otherwise, I currently prefer linear-phase filters, which bring me excellent space. For example:  poly-sync-shrt-lp.


    A reading example: Filter poly-sync-shrt-lp (fourth last on list)

    Using Convolution with 2x384kHz/64Bit,

    PCM input with 44.1kHz/16Bit (CD format)

    converted to DSD 256x48 with modulator ASDM7EC.

    The filter improves space and is particularly suitable for jazz / blues / classic (Info from manual). But now I'm listening Pop and its great. 😉 

    16 cores (hyperthreading) are active with 4.2GHz each.

    The filter works, the first initialization takes some time.

    The CPU utilization is 25% with a total power consumption of around 4.2A.



  9. Miska comment on DAC bits
    HQ Player
    1 hour ago, rickca said:

    @Miska I realize DAC bits is just for PCM.  I think you recommended setting DAC bits to 20 to @Superdad for his Holo Spring and he confirmed that it improved sound quality.  With my Nu Audio card, you said just leave it at default.


    So I'd like to understand under what conditions should DAC bits be set to a specific value and how do you determine the appropriate value?


    Holo Spring has both R2R PCM section, and DSD section. When using the R2R PCM section like he does, it is important to only utilize linear part of the DAC. This can be measured with a linearity sweep, or alternatively testing low level linearity at different word lengths to determine the optimal point.


    When sending PCM to a SDM DAC (like ESS, AKM, TI, Wolfson or Cirrus Logic), it is important to send data that utilizes full data precision it can accept, but not more or less. This is because the data goes through the on-chip DSP processing. Most of the new chips accept 32-bit data, while older accept only 24-bit.


    Another case is when the data ends up being sent through S/PDIF or AES/EBU which limits resolution to maximum 24-bit, although many USB-to-S/PDIF interfaces take in 32-bit data over USB and claim 32-bit resolution. And even if they do report the actual resolution too, it gets hidden by the driver - some time ago I added this capability to the Linux driver which was also hiding it (for example Holo Spring says it has 24-bit resolution while it takes in 32-bit samples). In addition, since S/PDIF is unidirectional, there is no way for software to determine what kind of DAC is behind the S/PDIF link. For example it could be some old DAC chip with 16-bit resolution. In such cases it is important to set the actual resolution.


    In addition there are some cases like Holo Spring 2 + macOS, where if you want to use the 32x PCM rates, you need to set DAC bits to 16. Otherwise it won't work.


    The AKM DAC chip on Nu Audio card understands 32-bit data, so that is best for it's DSP processing (in case you for some strange reason want to send PCM there).


  10. Lush^2 - Share your configuration experiences
    Lush^2 - Share your configuration experiences

    Dear John,


    Yes, your Lush^2 was shipped to you in this still (for me) best configuration.

    A: B-W-Y-R, B: B-W-R

    You name that one yourself in your last email from October 2018.


    PNF is a name given by other forum members, and stands for "Peter's New Finding". Google brings me to the page in this thread which mentions it :


    On 8/18/2018 at 11:15 AM, PeterSt said:

    Then ... since March 2019 people tend to like my new finding, commonly known now as "PNF" (Peter's New Finding). This is

    A:B-W & Y-R, B:B-W. You see this below.



    (don't be confused by the 8/18/2018, because that is the date of the first post in this thread, and the first post can be edited by its creator).

    My own description of this one (somewhere in August 2018) :


    This showed a super sound.
    It completely changes the sound from a somewhat congested (too white) highs to ever so lasting colored cymbals. Btw, this is what I had in mind with it for a change (I found the highs too profound).
    What came with it is a super fluid/liquid bass which sings and plays music. I actually never experienced the bass like that.


    February 26, 2019, I used the "PNF" again, but merely to solve a for me not the best sound with a new processor I tried. My description:


    I used it for the 12/24 processor which otherwise would not sound right at all. And now ? the very best !!!


    June 14, 2019, I changed for the last time, and back to the one "with consensus" (the one still shipped).

    After this, however, people started to tweak with the jumpers, and this came from that for myself:


    This could be the first time that all the pins were covered with pins (or wires). I mean, I was always careless and lost a jumper – and I did not care much. BUT this alters the SQ. Say that a normally assembled jumper would be denoted by a j and a 90 degree twisted one with h (half-assembled), then this comprises this config :
    A: B-W-Y-R-j-j, B: B-W-R-j-j-h
    (I may change this notation as others work on it too).
    The sound of this is a bit of “ouch” but I still managed for 3 weeks (see date below). Cymbals are very present but also very white. Not hurting, but very profound, giving the lot a clear character. It surely does not work for all music, but at least for hard-rock (like Deep Purple, J.Geils band) it worked out quite well (and my Ambient always works out).

    Keep in mind : the real change in this by now most-often used setting, is that all pins supplied with jumpers.


    I just looked what I currently use, and it is:

    A: B-W-Y-R-x-x, B: B-W-R-x-j-j

    Don't ask me the merit of this, because I don't recall making it like that and I didn't register it either. I must like the sound of it, because this must be so for months by now ...





  11. HQ Player
    HQ Player
    7 hours ago, scintilla said:


    Understood.  I may try that just to see.


    I already use acourate-built 131k tap RC filters in HQPlayer's convolution engine and it works wonderfully.  As I understood Mitch's review, this plugin works in real-time to further reduce room mode decay times based on the time-varying signal, which isn't something a static filter convolved with the signal can correct. Right?  I am limited in what I can do for room treatments because I live in a brick loft and I can't attach anything to the bricks or mortar. I could add some bass traps but the good ones (SpringTraps) are cost-prohibitive for my room volume (~6000 ft^3). So I thought this might be worth at least a trial. Thanks again, Jussi.


    To me it looks like room eq applied only to low frequencies. Which is what I'm also doing in my listening room. It is least intrusive when you stick to <500 Hz. When you attenuate the modes, they are also reduced in time. In my listening room the solution for best sound is to use stand mount speakers plus subwoofer. Cross-over between speakers and sub is acoustic, speaker ports are plugged, which makes their low frequency response roll off suitably, and then sub's low-pass frequency is tuned to match the main speaker roll-off. Then room eq is primarily applied only to the sub (I can switch between different corrections though, and now it's even easier with the matrix profiles in HQPlayer). In this scenario, I can also do the correction "in hardware" when needed using DSPeaker Anti-Mode 8033S-II (another nice Finnish product) without touching the main speaker signal. I can still do final touch-up in software when I want to, I have also different profiles for case with Anti-Mode in place.


    In my livingroom the approach is different, since there I have just Dynaudio floorstanders, so there the correction is always running purely in software. (possibility would be to use AntiMode X4 but it always runs in PCM domain to max 96k or 192k).


  12. Changing rtirq in euphony
    Euphony OS w/Stylus player setup and issues thread
    On 6/9/2019 at 4:57 PM, bobfa said:

    I have still been playing music with the custom hardware that I have assembled. I have really enjoyed learning and testing all the different hardware and software.  I am still fine tuning my system.  I have started looking at USB cards.  I have started with the Matrix Audio card:



    I have only had the card in my system for 24 hours but it really does make an interesting set of differences.  I will let it settle in some more before I try to A/B and report.




    Hi Bob,


    I have the same card in use and already wanted to write about that much earlier..

    Some food for thought: Euphony and Audiolinux give realtime priority for usb by default. That means priority for the irqs of the mainboard usb ports. When you use such a USB 3.0 PCIe card there is a different irq in use. It is easy to find out about that with Audiolinux. The Matrix card is using irq16 on my system. No matter if with my Gigabyte H110N mainboard before or now with a ASRock Z390M-ITX/ac.

    It's no special thing to change the rtirq.conf on Audiolinux. On Euphony I found a way with help of booting an Ubuntu Live USB drive and then mounting the both Euphony partitions to replace usb by i801_smb (browsing into /etc folder and opening a terminal window there: sudo nano rtirc.conf). Of course, a more convenient way is to ask Željko to change the both rtirq.conf files. I already had a conversation with him about this and he offered to make this changes via remote control.
    The not so nice thing is you must repeat these changes after each new version of Euphony. The installation of a new version overwrites the rtirq.conf files. But well it keeps the braincells awake.

    I *think* giving more rt priority to the irq16 really brings *something*. At least the good feeling that you don't waste rt priority for usb ports that are not in use. 🙂







    The Matrix card is using the irq16






    Audiolinux rt priority by default (with my former Gigabyte board)





    Euphony rtirq.conf by default





    from the top.log with the settings by default






    the changed rtirq.conf





    from top.log with the new rt pirority



  13. Ext2 vs xtr source content
    HQ Player
    6 hours ago, Yviena said:

    @Miska Just curious but is there actually any benefit to 40 bit/240db performance of ext2/xtr compared to -192db performance of poly-sinc, and if so in what situations would it be audible?

    I think I remember you saying somewhere that it would help high-rate modulators but unsure if this applies to DSD256.


    Technically it is potentially beneficial from DSD128 up and more so from DSD256 up. 192 dB comes from 32-bit PCM output resolution. While higher figures apply for SDM outputs, or noise-shaped high-rate PCM. This is still quite theoretical digital domain thing since from analog perspective either is extremely low level. But stop-band attenuation defines the ultimate boundary of reconstruction accuracy, so it should be anyway beyond DAC analog domain resolution.


    So I would say audible differences are not solely due to the stop-band attenuation, but partially about shape and steepness of the roll-off, which in turn defines many other properties. Because all of the things are mathematically interlinked.


    ext2 vs xtr is actually quite interesting comparison, because the filter lengths are not much different, nor stop-band attenuation either. Differences are elsewhere. But due to the differences, source content matters to some extent, because ext2 is strictly apodizing while xtr is not. Some older material like early Pink Floyd DSOTM and Meddle RedBook releases (not the recent remasters!) don't need apodizing function and there you can try for differences that are not due to that aspect.


  14. Room EQ HQP
    HQ Player

    REW (Room EQ Wizard) for creating impulse response for parametric EQ


    Here are the instructions I created for my own use over a year ago.  Please post any corrections, since the REW user interface may have changed since then:


    Prefs > EQ > Default Equaliser = Generic


    Prefs > View > clear “Enable mouse wheel zoom” if using a trackpad.


    Open EQ window by clicking EQ button (or Tools menu > EQ).

    • Gear button: Clear “Invert Filter Responses”.  Click Gear button again to close dialog.


    Open “EQ Filters” window by clicking “EQ Filters” button (top center of EQ window).


    Close EQ window.


    Return to “EQ Filters” window:

    • Clear “Always on top”.

    • Control = Manual for filter bank #1.

    • Type = PK for parametric EQ.

         • enter Q=1.4 for BW = 1 octave.  (0.67 for BW=2.)  http://www.rane.com/note170.html

    • Alternatively, Type = HS for high-freq shelf.  dB should be negative to shelve down above specified frequency.


    • Leave EQ Filters window open to assist in naming IR file afterward.


    Bring the main REW window to the front by clicking it or cmd-tilde.


    File menu > Export > Export filters impulse response as WAV.

    Clear “Normalize Peak”.  Mono.  32-bit (integer).  Sample rate 192.

    • No minus sign or other non-alphanumeric characters in filename.  Easier to save with default filename and then edit it immediately after saving.

  15. Hqplayer and digital conversion details
    HQ Player
    32 minutes ago, Miska said:


    Not with file output, but it of course can do conversion on the fly at playback time, no need to store conversion result in a file. Especially because such conversion is very light weight so can be done even on a less powerful computer.


    Thank you very much for the reply. But light weight? Rob Watts says the mscaler function can't be done real time (not even near) in a PC...

    Although I have my doubts on that, I am surprised you say it is light weight.

    Also I thought HQP was doing some optimizations (deviation from ideal function in order to be time efficient). The mscaler does optimizations (deviations from ideal). Offline can avoid any non-ideal optimizations. But if there are none...

    Of course one thing is the number os samples we use for the sync function, the more samples the higher the output resolution (real bits per sample), I think. Which also causes a few seconds delay with mscaler.


  16. Jussi comment on filters mods
    HQPlayer4 EC modulator tips and techniques
    1 hour ago, dminches said:

    Now that I have gotten my new build to work with ASDM7EC -> 48k / 256DSD I now need to take a step back and understand what all these modulators and filters are/do.  Is there a guide to them or are people just doing trial and error to see what sounds best?


    For filters there's some rough guidance in the manual. Filter choice depends primarily on content (mostly genre) and personal preferences. Personal preferences in a way that different people are especially sensitive to different aspects of the sound. When source material is something like RedBook, which is very tight package, there's no one size fits all filter. In addition, apodizing filters give fairly consistent results regardless of what kind of digital filters were used at the production side. While non-apodizing filters let the production side digital filter characteristics through and thus results depend more on what was used at the production side. Modern ADCs and many software rate converters leave quite a bit of aliasing band at the highest audio frequencies, which is cleaned up by apodizing filters at playback side (and at the same time allow modifying/replacing "ringing" behavior of the filter used at the production side).


    Choice of modulator is more technical, although there are many different technical aspects which are audible. If a DAC doesn't have very advanced analog filter in the output, fifth order modulator gives more relaxed noise profile by trading some SNR for that. Typical ESS Sabre DAC being example of such. Seventh order modulators provide improved SNR/performance, but the the noise profile is also more aggressive. These things matter more at lower DSD rates (DSD64 and DSD128) than at higher rates where the noise is anyway further up. Usually using either one doesn't cause problems anyway. EC ones are evolution of the earlier ones improving already very good performance significantly.


    Regarding SNR, what I've measured at DSD256+ speeds, using either fifth or seventh order modulator doesn't really change SNR of the DAC output at audio frequencies because both have already noise floor so far below analog noise floor of the DAC.


  17. SMPS thoughts and VRM's - Gavin1977
    Audiolinux Server configurations, Software, Hardware, and Listening Impressions
    18 hours ago, lmitche said:

    Hi Gavin,


    My motherboard criterion were Ryzen 8 core compatibility, an mATX or mini-ITX board, a well implemented and documented BIOS, the highest number of VRM phases possible, and lastly no cooling fan on the PCH.


    You can refer to the excellent Tomshardware AMD motherboard spreadsheet, https://docs.google.com/spreadsheets/d/1wmsTYK9Z3-jUX5LGRoFnsZYZiW1pfiDZnKCjaXyzd1o/edit#gid=2112472504 .


    To narrow things done from there, please use google to learn how the user and professional reviewers rated each board. Given the criterion and additional information above, please let us know which board you recommend.


    Many thanks,




    P.S. I am using the same power supply,  as well as the same memory sticks as the 6700k build.

    Thanks Larry - this is a very useful chart and one I was not aware of.


    I would highlight the following theory as the basis for my experiment, sourced from http://mail.personal-view.com/talks/discussion/21417/motherboard-vrm-power-guide/p1


    • First off is transient response of the inductor. If we have a single phase and very high current requirements then we will have to expand the energy storage of our components. However increasing the energy storage of the inductor also reduces the transient response of the system. So using many inductors with lower per phase inductance will allow for faster transient response. 
    • If you have a single phase the chances are that there is going to be a lot of heat there, so spreading the phases across a larger area and increasing their count can help reduce heat and increase efficiency. However it is important to note that switching losses do become greater with more parts. One must also think about the amount of current the copper in the PCB can transfer over a small area of space, a single phase as big as a normal phases in a multiphase VR can’t output 300A, but spread that over 6 phases and 6x the space and it is only 50A per the same space.
    • The interleaving effect is by far one of the most beneficial effects of having more than 1 phase. Because of how the phases operate out of phase with each other the total ripple frequency can be multiplied by the total number of phases. This huge increase in ripple frequency decreases ripple amplitude and ripple current. This helps reduce both input and output bulk capacitors requirements. If you skip down a little bit then you will see how more true phases can decrease current.


    The interleaving effect is important - so phase doubling is ok.


    It was just out of interest that I tested this theory.  I had my old Gaming PC lying around, so I thought I would see how it sounded compared to my Intel NUC7i7DNHE.  The DNHE model is apparently the 'best' NUC for audio reproduction at present according to reports on this website. So I put the Euphony USB stick into my gaming PC and...


    My gaming PC sounded signficantly better on all fronts!  


    There was significantly more detail, it was tonally lush and had huge dynamics in the presentation.


    This was not what I was expecting, as I had not optimised the PC for audio playback other than using Euphony.  Now, truth be told I was only using the stock power brick for the DNHE NUC, this should be expected to be relatively noisy with ripple likely in the 60mV or so range I would expect.  The switching mode power supply in my gaming PC is a Corsair SF450, ripple is measured at 25mV which is good, but not outstanding for a SMPS.  So the SF450 should probably sound a bit better than the NUC's stock power brick.  However, 25mV is nowhere close to a good linear power supply where ripple would be 1mV or less (and much, much lower if you go for a Paul Hynes Design).  So what's going on, and why do I hear such an improvement!?


    After some research, I believe that Voltage Regulation on high-end gaming motherboards is part of the magic sauce.  It is probably the reason why some gaming PC's can sound far better than 'lower powered' computers/streamers/renderers etc...  This is potentially one reason why very expensive servers such as Antipodes CX/EX, Sound Galleries etc... sound so good, as the cheaper, smaller, lower powered servers just don't have a lot of money spent on voltage regulation on their motherboards. The Sound Galleries Server uses a gaming motherboard from Asus... which has a high phase count.


    Another good article on Voltage Regulation is available here: https://www.maketecheasier.com/what-is-vrm/


    The motherboard I experienced these improvements in sound quality on is a mini-ITX MSI Z97I Gaming AC which is a 6 phase digital power design - I believe it's a 'True' 6 Phase design, 'True' because it doesn't use a doubler and it also has exactly 6 high quality Super Ferrite Chokes (SFC) on the motherboard.


    Of course space on an ITX motherboard is limited, but there are some older gaming motherboards out there that do have more phases - for example the ASUS Z87I-Deluxe has a power riser to achieve 12 phases.  


    ATX sized gaming motherboards can have 8, 12, 16, even 24 phases of power control.  What I have read is that a VRM with 24 phases of power is best.  Gigabyte produced one of these in a motherboard called the GA-P55-UD6 - but this is very rare and only supports older processors.  Motherboard manufacturers seem to have moved away from high phase counts for their latest designs, and have settled around 12-16 phases as perhaps they've found the sweet spot commercially.  But if you can find one of these older motherboards on the market then perhaps you have a very cost effective solution for producing a very high quality music server.


    Why is 24 phase voltage regulation on a motherboard a good thing?  See Table 3 on this datasheet - 



    With 24 interleaving phases the ripple is significantly reduced from that provided by the SMPS.


    So, my present motherboard only has 6 phases of power regulation, I have now purchased a very rare MSI Z87 XPower XL-ATX motherboard which has a staggering 32 phase voltage regulation and also a stupidly high bulk output capacitance (which I believe will further improve the dynamics of the sound).  A ueful historical list of gaming motherboard VRMs can be found here: https://web.archive.org/web/20160914210453/http://sinhardware.com/images/vrmlist.png


    So I think there might be merit is going for some of the older motherboards due to there very high phase count (we will see!)


    The very best ATX SMPS for ripple are made by Super Flower, specifically the 550W version of their Leadex Gold PSU's - no other manufacturer comes close at present.  Leadex Gold SMPS measure under 5mV of ripple (they use LLC resonance conversion, which also pops us on the DIY audio forums as well).  My theory is that 5mV of ripple will then further smoothed out by 32 phases of voltage regulation and I believe it will provide exceptional performance.


    This is a VERY cost effective solution as well - the largest expenditure is a gaming motherboard with a high phase count (some of these are £300-400 new), but an older model with a second hand processor could be had for £150 or so on eBay.


    The 550w Leadex Gold PSU is only £60-70 which is amazing for only 5mV of ripple.  Forget about Pico PSU's... I used them in prior media centre builds and they don't offer the same level of audio performance as they're still influenced by the power brick used upstream.


    This is just a different way of achieving the common goal of better sound reproduction.  I was not expecting my intel NUC to be relegated to the sidelines so quickly - I could buy a linear power supply for it, but if I'm achieving these levels of performance then I cannot see the value in spending hundreds on a bulky linear power supply.  Afterall, Chord Electronics, Benchmark and Hypex all use SMPS, so it can work given the right design approach. 


    I hope this helps others.


    The processor I am using is a 4790k - just because it's what I already had in my gaming PC.  It would be interesting for me to switch out this CPU for a lower powered variant - however (other than the 8MB L1 cache) I'm not sure it would make much difference(?  I need to do some more reading about this though) as I think the magic sauce is high phase count voltage regulation.  Underclocking the processor from stock is an option, but I have read elsewhere that some people find that some instruments are not rendered as well.

  18. Lush 5
    Lush^2 - Share your configuration experiences

    Hello, I've been lurking and experimenting since receiving my Lush^2 a couple weeks ago, using the various recipes posted here as my guide. I've tried 7 different configurations, and this one sounds best on my system.


        A: B-W-R    B: B-W


    My audio chain is UltraRendu > Lush^2 > ISORegen > Light Harmonic Lightspeed > DAC


    Like @austinpop, I expected to place the Lush^2 between the IR and the DAC, but was surprised to discover that I prefer it between the UR and IR by a considerable margin.

  19. Peter 4rd
    Lush^2 - Share your configuration experiences

    All right.

    A: B-W-Y B: B-W

    @kurb1980, I think you really have found something there.


    This config bears all kinds of oddities, and this begins with me not being able to detect a real difference on the first day I tried (day before yesterday). Maybe I was too focused, maybe I played the wrong music, maybe I had in mind that it was too soon after all (after an other change in the chain).

    Then came yesterday ...


    From of the very first notes I was a kind of shocked. There now was detail in a frequency range which brought forward sounds and voices and what not in really everything, I never heard before. But besides that, this bears characteristics which are quite uncommon - especially because they all emerged together (like changing the complete audio system). Here:


    - Super wide sound stage.

    - For a first time ever I could detect a most clear adding up of volume when walking through the middle (say sweet spot) from left to right and back. Something like "hey, in the sweet spot this is 6dB louder".

    - Depth is very shallow. All plays at the level of the speakers themselves.


    Especially this latter one intrigues me largely, already because I never experienced that. But further more and more importantly: in a fashion which did not disturb me, no matter how hard I tried to find it wrong. The most strange thing is that while the depth of stage is as flat as maybe a few feet, the speakers are totally undetectable. What adds to this is the wide sound stage (well beyond the speakers, never mind all those who claim this can't exist). So it is the whole presentation which is new to me.


    A first thing I noticed was a perceived buzzing of the bass, as if it carries too much energy and implying frequency which is too high for what's in the material. But in each track where this seemed to occur, it was a different presentation of the bass itself, and that too I never heard anywhere. Something like: at listening closer, you could hear frequency *under* the frequency you actually hear (as thus a kind of buzz) and that all together creating a sound of bass which is, well, new. Here, the angle of listening (towards the speakers) is again important (see the 6dB story). I am not used to that at all.


    The way I think about these matters is that the relatively flat layer of depth now contains all the music, while otherwise it is spread over the many feet towards you and nothing doing so much individually. Now, all being stored in a compact container, it seems to fight for freedom and pushes against each other. Think buzz and you get the idea (hopefully).


    I wrote the above like telling about a gadget. But it is G-D the best I ever ever heard. Can music sound even more real than this ?




    I would like it very much if others could confirm or debunk any or all of this. It again seems hard to imagine that people come to a same perception as just was written out, but so far there's consensus about what we perceive from the cable settings". But merely: do yourself a favor and have a brand new system once again.

    Warning: It seems impossible that after the first day I was ready to write "nothing wrong with this and maybe it takes more listening to find out what's different for the better or the worse" while the second day I was open mouthed for two hours, right from the first second of listening. In other words: it can well be that this config requires a bit of breaking in (just have it connected to powered PC and DAC for 24 hours - all idling).


    kurb1980, thank you ...



  20. Hols1st
    Lush^2 - Share your configuration experiences

    Just would like to share my early experience with the Lush 2. I received my Lush 2 last week and I burned it continuously for a week (changed the configuration a few times during the burning, majority is burning the default configuration A: B-W & B-Y  B: B-W) before I auditioned it together with my friends( all have over 30 years in this hobby). Occasional peeping through during the burning did show that the sound is improving with time. 

    I shall report according to the sequence it has turned up in our gathering when it was connecting from my Windows 10 HQplayer to Holo Spring DAC playing mainly classical  orchestral music


    (1)First that come up is the default config (A: B-W & B-Y  B: B-W) and the sound is sort of neutral nothing surprise. Not really much change from Lush 1 from my memory.

    (2)Next up is the A: B-W-Y-R  B: B-W-R and this one shocked everyone present. The orchestra and chorus suddenly falls into place with excellent layering, very lively sound and really holographic layout. Instruments accurately focused. This is the sound of a top notched USB cable one can dream of. 

    (3)Then comes the A: B-Y & W-R  B: B-Y & W-R and this another excellent presentation and seems to be on the other extreme. This one gives a very rich timbre of every instrument in the orchestra. And the music easily becomes very enjoyable. I would say that this is more like a tube gear sound while the one before is more like transistor gear sound. Almost all presnt prefer this config to the last one. But to my ears this is the one with a bit too much of mids. The piano sound is not as lively as before and one cannot hear the details as well as the last one. 

    (4)Then comes the (JSSG360 )A: B-W & Y-R  B: B-W & Y-R. The sound is sort of in between the first and second one but the musicality is significantly lower than the previous two. In other words, not so musical and then the layering and image focus is also not as good. The music does not catch the ear so much as the others.

    (5)Next comes the default again in order to verify whether it is that bad? It is quickly passed because the performance is way lower than the others. 

    (6) A: B-W &Y-R  B: B-W-Y-R  not good. quite lean

    (7) Then we changed to the Lush 1 to have a listen and everyone finds that it is not listenable. Flat sound stage, muffled sound and not enough details.


    We don't have time to go to too much detail about the other configs yet. But the second and third one described is way above all the others to our ears. And thanks Peter for giving us such a high quality USB cable and can tune the sound according to one's preference.

  21. Ddetaey1st
    Lush^2 - Share your configuration experiences
    On 9/16/2018 at 4:41 AM, PeterSt said:

    All right. I went back to this one :

    A: B-Y & W-R, B: B-Y & W-R


    Indeed this does more than the A: B-Y & W-R, B: B-Y and the major thing I readily notice again is :

    The enthusiasm on the snare drums. This may come across as strange, but think that suddenly the drummers isn't sitting on his stool, but now stands and jumps in order to get his whole weight behind the drumming stick before it lands on the snare drum. It really gives drive.

    But then again it is also not the most realistic because drummers tend to sit on their stool. Well, most do. 9_9

    There's also a hint of buzzing in the room, which to me tells that standing waves try to surface - but it is doable.


    So this one could work out for the better because we may subjectively like it better, but I don't think it *is* better. Anyway it is more lively - it energizes the room more.


    For me this configuration  A: B-Y & W-R, B: B-Y & W-R is the only one that really works, and that provides me a (of course very subjective) added quality compared to my personal reference cable ( Forza Audioworks Hybrid Series USB) .


    For me this config does NOT energize the room, as Peter decribes here. For me Peter's description here matched perfectly with the A: B-Y & W-R  B: B-Y configuration (config as delivery of my Lush2).


    Maybe important to notice that I listen exclusively to DSD,  so all PCM is also converted to DSD, using HQPlayer embedded.

    Today I listen to DSD128 using DoP ( hopefully I will be able to test DSD512 native somewhere next week).


    I also listened to Peter's newest config (up till today that is) A: B-W-Y-R, B: B-W-R.  Again my findings are different from Peter's and other forum members.  I find the stero imaging less precise than with A: B-Y & W-R, B: B-Y & W-R.


    Some examples of my experiences :


    1) Norah Jones - Sinkin' Soon (album Not Too Late)  RB upsampled to DSD128
    perfect imaging, very good clarity (without sounding harsh), strong but not exagerated bass


    2) Eric Clapton - Layla (album MTV unplugged) - 96kHz/24 bit Bluray rip upsampled to DSD128

    this configuratiion is the only one that get's the position of Eric correct, little to the right of the center from the listener's point of view.

    also the position of the piano and the second guitar player remain consistent, without displacing during the song.


    3) Laura Marling - Made by Maid (album I Speake Because I Can) RB upsampled to DSD128.

    Very good clarity, delicious voice and (almost) perfect height position of voice and guitar (very noticable with other records too !!).


    4) Roger Waters - Amused to Death (album Amused to Death) - SACD Rip (DSF) upsampled to DSD128.

    Excellent clarity, holograpic sound.  As a special notice, due to Peter's comments - the positioning of the individual drums is fantastic, in my believe because of the more correct height position).


    5) Beatles : Love Me Do (album Past Masters) RB upsampled to DSD128.

    Mono recording with perfect head tracking, as per request of FAS42.


    Last but not least:

    6) Tchaikovsky - Violin Concerto in D Major, Opus 35  by Pinkas Zuckerman & Zubin Mehta +Israel Philharmonic Orchestra - 

    RB upsampled to DSD128.

    I am far from a classical expert (max. 10% of my collection and of my listening) but his is one of my favorites. 

    However, even when resampled to DSD, Pinkas violin can sound sometimes a little harsch, or edgy of you will.

    This cable configuration however makes the sound of the violin so delicious and delicate, whilst loosing nothing in detail (at the contrary) . Also the positioning of the instuments in the orchestra is excellent, as is bass weight.



    So, I do not agree with Peter that this config is too good to be true. 


    This is my reference as of today, and I will leave it as such - no more cable config playing/testing.

    From now on, as far as cables are concerned, it is exclusively back listening to music again.






  22. P3rd
    Lush^2 - Share your configuration experiences

    This is now my best by far :


    A: B-W-Y-R, B: B-W-R


    It adheres the analogy that the middle shield is captured between the other two and really actively does something; and the source (A) it is connected to the connector and at the target (B) it is not. And now I envision something like this happening (pick your most speaking picture) :




    This thinking is based on nothing really, but since I did some work in the field of antennas and radiation patterns, for me it is workable. Anyway, such a thing would (might) happen with the middle shield (and the others) connected at one end, but not connected at the other end (the shields surrounding it, connected again). In this case the radiation pattern will be "even" like you see in any of the pictures above.




    In short : I never heard such a palpable sound which at the same time is razor sharp, which has an outrageous umpf and together with that is not disco. Mind all the contradictions in that one sentence - this is a most notable thing on its own.

    Btw, our son asked whether I was sure I had arranged for a House License, referring to the somehow for him working out (House Party) sound of music he just should be used to, while he now suddenly had to make that remark.


    Bass is as firm as I never heard it, with thus the strange property that it does not bloat or color at all. The other strange "property" for the combination, is that the highs excel in such a palpable fashion. Partly this will be because of the bass itself, but how it combines with the superb highs - no idea yet. I think I notice a somewhat "shorter cymbal" which is not a negative in my book; a too long cymbal for my system usually comes along with overly detailed sounds like spitting in the microphone and tongue clacking sounds. However, I did not test for that really, yet.

    The whole picture is one of drive - drive - drive (and is the house party thing). I can't recall a better evening, ever.


    On 9/19/2018 at 6:37 PM, zettelsm said:

    this is the one that I could keep turning the volume up and up and up and play stupid-loud.


    With a great thanks to Steve for his superb description in his last post, I will say that this configuration as of now allows for this just the same. The difference should be the sustaining presence of the highs. I think I also can tell what the real difference is : somewhat less higher mid. And yes, if I am correct I recall my own expression about the emphasis to the "lower highs" of the configuration Steve refers to - maybe my memory serves me wrong.


    There hasn't been a single stroke of which I thought "is this real ?" or "is this correct ?", which is of quite some value to myself.


    Maybe important : This was the 3rd day with this configuration with the clear (to self) notice that I hardly really listened the days prior to that. The caution now is : what if this requires a bit of burn-in. So my judgment was after three days only, and nothing prior to that. Might you (with burned-in Lush^2's) readily think "this is nothing !" then please try to hold on for two more days.

    And of course we still like to reach consensus whether the configs work out the same for everybody ... 



    A: B-W-Y-R, B: B-W-R


  23. P2nd
    Lush^2 - Share your configuration experiences
    4 hours ago, PeterSt said:

    A: B-Y & W-R, B: B-Y


    People should really try this one. It is special and sooo normal at the same time. And the most important : it is also beautiful and for that reason a rarity in the bunch. Genesis never sounded so lovely over here (I always found Genesis more difficult to play).

    Contrary to the previous one (see list) this is not flat and not "fairly mono" at all. Not lean either. And look what's actually changed to the config ...

  24. Lush^2 Peter's 1st
    Lush^2 - Share your configuration experiences
    16 minutes ago, BigAlMc said:

    I'm torn between this being fascinating discovery or just too bloody hard! ?


    Nah Alan, not bloody hard. But we must adopt some "good behavior". Learn from others (at least from me) and what they say about it - test that if seems good and confirm (hey, or debunk). If this confirmation does not happen, there is not much to learn for others, or it goes too slow.

    Then also : put up your own experiences, like I did yesterday. Thus including the failed ones. From that we may learn even more, or gain time.


    The general procedure, if (!) you're done with e.g. copying my configs :

    Try a next new one. Sounds poor ? you wil hear it in minutes. Try a next one. Poor ? same thing. Next one ... aha ! let that stay. Describe firstly to yourself what you perceive from it. Also the poor ones and the how. Now others are going to try yours and may debunk, or confirm.


    For fun, here is my document (must learn myself how to deal with it too) :




    A: B-W & Y-R, B: B-W
    This showed a super sound.
    It completely changes the sound from a somewhat congested (too white) highs to ever so lasting colored cymbals. Btw, this is what I had in mind with it for a change (I found the highs too profound).
    What came with it is a super fluid/liquid bass which sings and plays music. I actually never experienced the bass like that.


    A: B-W-Y-W, B: B-W
    Marvelous sound which keeps on being strange. Listened to it for well over a week. After that week I decided it is time for something which doesn't carry doubts.


    Up next (31-08-2018) :
    A: B-W & Y-R, B: B-W-Y-R
    Way to dry. Raspy highs. Strange. Black, but lower highs lacking ?


    Up next (31-08-2018) :
    A: A: B-R, B: B-R
    Sounds quite similar to the Lush^1. Nothing spectacular and if anything, less than the Lush^1.

    Up next (01-09-2018) :
    A: A: B-Y & W-R, B: B-Y & W-R
    Best so far ? Also see Phasure post on 02-09-2018.


    Up next : (this is due)
    A: A: B-Y & W-R, B: W-R




    Thinking back on joyful moments, I lurk to the first one mentioned. So, I can also go back to that because I kept track. Plus, maybe by now my Lush^2 has been burned in better. Plus, I changed something in my PC. Plus ...


  25. HQP Filter Description
    HQ Player
    13 minutes ago, AnotherSpin said:

    Miska, I am sure this question was answered already, but what is the the difference between apodizing and non-apodizing filters in a plain, non-technical words for those, who do not understand special terminology?) In my set closed-form gives more natural sound, while ext2 gives very sweet, but slightly artificial (in comparison with closed-form) sound. And, which other filters are non-apodizing besides closed-form?


    Since all ADCs that produce something like 44.1k output are oversampling, there's a decimation filter to convert the higher rate down to lower one. In addition, these days is quite common that recording is made for example using 96k sampling rate and then converted down to 44.1k using another decimation filter.


    Point of apodizing filter is to replace original decimation filter's impulse response with another one. This allows altering time- and frequency domain behavior of the original filter. You can get generally shorter "ringing" one, with something like poly-sinc-short or longer one with something like poly-sinc-ext2. Or you can change to a minimum-phase one.


    Another maybe more important point is to clean up aliasing band at the highest frequencies that happen due to "modern" half-band ADC and DAW decimation filters that have pretty much no suppression at the fs/2 (Nyquist) frequency, and thus content exceeding that in the higher rate source data folds down into lower frequencies. For example original CD release of Pink Floyd DSOTM didn't have this dirt band at top, but the latest remaster does. This applies primarily for source content at 44.1/48k rate, and to much lesser extent to hires content.


    closed-form is not really a filter, but instead interpolator, so it is non-apodizing due to that. poly-sinc-hb is a non-apodizing half-band filter a bit like the ones in modern ADC and DAC chips, but just better (higher precision and stop-band attenuation). minringFIR is another non-apodizing half-band filter with fairly slow roll-off, so it will let more images through too. You could compare poly-sinc-hb vs poly-sinc and minringFIR vs poly-sinc-short. Also poly-sinc-ext2 vs poly-sinc-xtr (xtr is quite a bit less apodizing than ext2).


    When you use non-apodizing filter the results largely depend on what kind of decimation filter was used for the source content, because all it's faults come through as-is. While apodizing filters give more consistent performance across the board by correcting faults of the original decimation filter and giving same impulse/frequency response across the board regardless of source content.


  • Create New...