A toast to PGGB, a heady brew of math and magic

[Zaphod Beeblebrox]

8 hours ago, astrotoy said:

Has anyone tried a Lampizator using PGGB, particularly the new Horizon?  I had a friend (on the list) send me some sample files to try.  I don't hear any significant difference through my Horizon, but it may be my settings through Roon and SQX or HQP and the Lampi.  I have a very large number of files that I ripped from vinyl and tape (over 40TB) at 192/24 (using my Pacific Microsonics Model Two and Merging Technologies Pyramix software), so it would be a big project and big storage to upconvert using PGGB.  

 

I also have an older (old) formerly pretty state of the art computer from 2017-8

CPU    $1,649.99        i7-6950X                                                        

MB              $218.86   GA-X99M-Gaming 5 (rev. 1.1)                                                         

RAM   $300.99           Savage Memory Black - 32GB Kit*(4x8GB) - DDR4 2400MHz Intel XMP CL12 DIMM

Video  $545.84           ROG STRIX-GTX1080-A8G-GAMING                                                     

PSU             $199.99   Seasonic SSR-850TD PRIME 850 W Titanium                                                        

Disk             $629.99   SSD 960 PRO NVMe M.2 MZ-V6P1T0BW                                                

Case    $137.66           Fractal Design Define R5                                                      

Cooler $89.93 Noctua NH D15                                                        

Fans     $16.99 Phanteks 140mm Cooling Fan (PH-F140SP_BK)                                                    

OS               $139.00   Microsoft Windows 10 Pro OEM (64-bit)                                                    

Apps    $129.00           Audiophile Optimizer 2.0 beta 15 (using "strip down Windows 10" service tool option)                 $120.00           Roon server                                        

                $150.78       HQplayer (sends to NAA)

 

Not sure what I would need to upgrade or do I start again with a separate unit just for PGGB conversion. 

 

Thanks,  Larry

I may be able to suggest based on the settings you use. For PGGb I would suggest 705/768kHz, 32 bits and leave everything else in default mode and make sure that noise shaping is on. If you are using both Roon and HQP, please make sure Roon has all DSP disabled and this would include headroom. Set HQP to bypass, i.e set 1x and Nx filter to None and Dither to None and volume min and max to 0 and bit depth to 32.

[Gavin1977]

Gustard X26 Pro up and running .  Little bit lack luster to start with, but then I installed the latest firmware and drivers and it took a very nice step forward.

 

One thing I can't get past though is the USB driver (either v5.0 or v5.4) on Windows LTSC only seems to allow 32 bit / 384kHz max sample rate, yet X26 Pro is capable of 32 / 768...

 

Is this a Windows issue, am I forced to use HQPlayer or Euphony to pass the PGGB files through?

 

Chord DACs do 32 / 768 no problems in Windows, but the X26 doesn't have an ASIO driver set by the looks of it.

 

@SwissBear and @kennyb123 What do you do?

[kennyb123]

1 hour ago, Gavin1977 said:

@kennyb123 What do you do?

I use a Linux-based server and it doesn’t have this problem.  

[Johnseye]

3 hours ago, Gavin1977 said:

Gustard X26 Pro up and running .  Little bit lack luster to start with, but then I installed the latest firmware and drivers and it took a very nice step forward.

 

One thing I can't get past though is the USB driver (either v5.0 or v5.4) on Windows LTSC only seems to allow 32 bit / 384kHz max sample rate, yet X26 Pro is capable of 32 / 768...

 

Is this a Windows issue, am I forced to use HQPlayer or Euphony to pass the PGGB files through?

 

Chord DACs do 32 / 768 no problems in Windows, but the X26 doesn't have an ASIO driver set by the looks of it.

 

@SwissBear and @kennyb123 What do you do?

 

Euphony Stylus doesn't support wave pack above 192k. I'm dealing with similar issues. HQPlayer does though. 

 

I know you're looking for a Windows solution but thought I'd share. 

[seeteeyou]

4 hours ago, Gavin1977 said:

Chord DACs do 32 / 768 no problems in Windows, but the X26 doesn't have an ASIO driver set by the looks of it.

 

Bypassing Windows audio mixer

https://www.audialonline.com/articles/bypassing-windows-audio-mixer/

https://addictedtoaudio.co.nz/blogs/what-we-think/in-search-of-768khz

Quote

I used JRiver Media Center to send it to the Topping E30 DAC, and it duly displayed “768.0 PCM” on its front panel display (see the photo up top). That worked using the ASIO driver, the WASAPI driver and even Windows Kernel Streaming. It would not work using Direct Sound.

 

Assuming that you're download ASIO drivers from this particular page

 

http://www.gustard.cn/?page_id=8956

 

Even V4.86.0 of Thesycon drivers supported 768 as shown below

 

https://dl.khadas.com/products/tone2/driver/Khadas_Tone_Driver_V4.86.0_201118.zip

https://forum.khadas.com/t/tone-2-pro-768khz-pcm-option/13055/3

Quote

You need a player like said above, I tried it with Foobar in ASIO and WASAPI Exclusive mode, and both work (go directly to the Tone2 Pro and avoid the Windows mixer).
The Windows setting can be what you want, it doesn’t change.

705.6 also worked just fine

 

https://community.roonlabs.com/t/wasapi-in-windows-10-21h1-update/168248/2

 

Maybe get rid of any unused devices and see how it goes?

 

https://www.uwe-sieber.de/files/DeviceCleanup.zip

https://www.thewindowsclub.com/device-cleanup-tool-windows

 

More importantly, what audio output options (namely ASIO since that could actually reach either 705.6 or 768) did you actually see once you go to the settings of your software player? Maybe capture that particular screen with relevant settings so that we could take a look.

[LowOrbit]

Not much help but I had no issues with the Gustard driver on Windows (build 1903) or Linux, 768/705k worked great and sounded sweeter.

 

Really excellent dac - and not just "for the money". As I've said elsewhere, if I didn't have the Dave, I'd be happy with the X26pro. (I had the U18 at the same time and felt that using that and IIS to the dac was gave a little more sq too, but only had the pair for a few weeks, so didnt swap stuff about too much.

 

A Sablon EVO USB cable that I received a couple of days before I boxed the Gustard kit up for Amazon return made brough a good jump too.

[Gavin1977]

14 hours ago, seeteeyou said:

 

Bypassing Windows audio mixer

https://www.audialonline.com/articles/bypassing-windows-audio-mixer/

https://addictedtoaudio.co.nz/blogs/what-we-think/in-search-of-768khz

 

Assuming that you're download ASIO drivers from this particular page

 

http://www.gustard.cn/?page_id=8956

 

Even V4.86.0 of Thesycon drivers supported 768 as shown below

 

https://dl.khadas.com/products/tone2/driver/Khadas_Tone_Driver_V4.86.0_201118.zip

https://forum.khadas.com/t/tone-2-pro-768khz-pcm-option/13055/3

705.6 also worked just fine

 

https://community.roonlabs.com/t/wasapi-in-windows-10-21h1-update/168248/2

 

Maybe get rid of any unused devices and see how it goes?

 

https://www.uwe-sieber.de/files/DeviceCleanup.zip

https://www.thewindowsclub.com/device-cleanup-tool-windows

 

More importantly, what audio output options (namely ASIO since that could actually reach either 705.6 or 768) did you actually see once you go to the settings of your software player? Maybe capture that particular screen with relevant settings so that we could take a look.

Thanks for the foobar tips @seeteeyou.  HQPlayer also allows ASIO drivers to be available (so they are packaged within the driver install, just not easily accessible).

 

Presently using HQPlayer to pass through the files without alternation via ASIO.

[Peti]

I wonder how is work going on the improved version of PGGB that requires less ram to operate...?

[Zaphod Beeblebrox]

On 10/3/2022 at 5:07 AM, Peti said:

I wonder how is work going on the improved version of PGGB that requires less ram to operate...?

It is work in progress, the new version will see improvements both in processing performance (speed and memory) and also in terms of reconstruction accuracy. Significant research went into improving the upsampling algorithms, licensing of the higher performance libraires added to the costs. So, the PGGB license for the high-performance version will be an upgrade license.

[The Computer Audiophile]

Hi @Zaphod Beeblebrox, I got a very good taste of PGGB today. @austinpopbrought over some files. The sound was delightful. Nice work!

[seeteeyou]

https://www.dbpoweramp.com/Version-Changes-DMC.htm

Quote

FLAC 1.4.1 - Encoding and decoding of 32-bit PCM is now possible, 1MHz sample rates

 

https://www.signalyst.eu

Quote

FLAC update to support higher sampling rates and up to 32-bit word length.

 

https://hydrogenaud.io/index.php?msg=1011789

Quote

This limit has been raised to 1,048,575Hz in April, which is the true limit of the format. The previous 655,350Hz was the 'streamable' limit of the format, for some reason the reference encoder (libFLAC) didn't accept anything with a higher samplerate than that.

 

https://github.com/xiph/flac/commit/24629435bb265d740bf5ff12b2e48a7fbc319977

    /* sample rate can be up to 1048575 Hz, and thus use 20 bits, so we do the multiply&divide by hand */
    FLAC__ASSERT(FLAC__MAX_SAMPLE_RATE <= 1048575);

 

---

 

Compression ratios of WavPack versus FLAC seemed to be pretty darn close, too

 

https://hydrogenaud.io/index.php?msg=1011846

Quote

I've posted some results already for the free file - now including -5 and -8p from NetRanger's build.
 

969 129 022 Let The Good Times Roll - Carmen Gomes Inc - Free 768kHz.wav
649 644 648 Let The Good Times Roll - Carmen Gomes Inc - Free 768kHz.-5.flac
642 946 998 Let The Good Times Roll - Carmen Gomes Inc - Free 768kHz.-8p.flac
641 225 096 Let The Good Times Roll - Carmen Gomes Inc - Free 768kHz.wv-hhx6.=37min=.wv
640 902 644 Let The Good Times Roll - Carmen Gomes Inc - Free 768kHz.wv480-hhx6.=40min=.wv

 

 

---

 

The encoding side should be a piece of cake since FLAC could be updated to version 1.4.1 (or above) on our own, though we've gotta ask the developers of software players to add support for the latest versions of FLAC.

 

Many of us should be able to ditch WavPack at some point once everyone is getting on board with FLAC @ 16FS, while others could stick with WavPack @ 32FS if our (R2R) DACs were able to handle that.

[Zaphod Beeblebrox]

1 hour ago, seeteeyou said:

though we've gotta ask the developers of software players to add support for the latest versions of FLAC.

This is the key

[seeteeyou]

For TAS and / or XDMS, hopefully updating FLAC or libFLAC to version 1.4.X wouldn't be all that tricky so that could be approved by Emile.

 

---

 

When it comes to Roon Server, someone else already started a new thread last month

 

https://community.roonlabs.com/t/flac-1-4-2-support/219843

 

Please feel free to cast your vote by hitting that "Vote" button over there.

 

---

 

For those music servers from Antipodes Audio, Logitech Media Server (no updates since March 2022) still came with FLAC version 1.3.4

 

https://github.com/Logitech/slimserver/blob/public/8.4/Bin/x86_64-linux/flac

Quote

Fix #773 - Add flac 1.3.4 for macOS/Linux

 

*IF* a simple drop-in replacement were actually good to go, we might consider making a backup of this file first

 

opt/logitechmediaserver/Bin/x86_64-linux/flac

 

Then grab a copy of FLAC version 1.4.2 and let's take the Arch Linux as an example

 

https://mirrors.edge.kernel.org/archlinux/extra/os/x86_64/flac-1.4.2-1-x86_64.pkg.tar.zst

 

We'll extract both files listed below and then copy them to opt/logitechmediaserver/Bin/x86_64-linux accordingly

 

https://archlinux.org/packages/extra/x86_64/flac/files/

usr/bin/flac
usr/bin/metaflac

 

If that were no go, revert all changes and then maybe ask Mark what to do about updating FLAC for LMS since the official version didn't get any updates for quite a few months already.

 

Other than that, maybe open a new issue here

 

https://github.com/Logitech/slimserver/issues

 

Simply mention a previous issue linked below and then Michael Herger should know what to do next

 

https://github.com/Logitech/slimserver/issues/773

 

---

 

HQP / NAA already got that handled, same deal with the latest releases of foobar2000

 

https://www.foobar2000.org/changelog

Quote

libFLAC 1.4.1, made possible to decode 32 bits per sample FLAC file.

 

---

 

For converting WavPack to FLAC safely, foobar2000 could be one of those viable options

 

https://hydrogenaud.io/index.php/topic,119929.msg988566#msg988566

Quote

So, using foobar2000, I'd do the following steps (lots of overnight tasks here):

1. foo_verifier, verify all files. (Re-encoding will produce "valid" files; if something is corrupted, I want to know, not to destroy evidence.
2. foobar2000 to convert to identical directory structure
3. foobar2000 with foo_bitcompare. (But first, check that you have the same number of files, same number of samples, etc.)
4. Delete old. Again, using foobar2000
5. Search old structure for WavPack files to make sure there are no-one left that weren't caught and indexed
6. If you didn't copy "other" files, now it is time to merge.

 

foo_verifier
https://www.foobar2000.org/components/view/foo_verifier

 

foo_bitcompare
https://www.foobar2000.org/components/view/foo_bitcompare

 

Foobar2000:Converter
https://wiki.hydrogenaud.io/index.php?title=Foobar2000:Converter

 

Free Encoder Pack

https://www.foobar2000.org/encoderpack

flac - Command-line FLAC encoder/decoder version 1.4.2

 

How to convert WavPack Files to FLAC
https://hydrogenaud.io/index.php?topic=120174.0

 

Foobar2000 Conversion Setup
http://www.vgmpf.com/Wiki/index.php/Foobar2000_Conversion_Setup

 

How to convert FLAC files to MP3 using foobar2000
https://captainrookie.com/how-to-convert-flac-files-to-mp3-using-foobar2000/

 

---

 

For Mac and Linux users, the author of flacon didn't state if FLAC / libFLAC were updated to version 1.4.X or otherwise

 

https://flacon.github.io

Quote

• Supported input formats: WAV, FLAC, APE, WavPack, True Audio (TTA)
• Supported out formats: FLAC, WAV, WavPack, AAC, OGG or MP3

 

Latest releases

 

https://github.com/flacon/flacon/releases/latest

https://flathub.org/apps/details/com.github.Flacon

 

We might start a new issue on GitHub and then find out if those features from version 1.4.X of FLAC were actually included

 

https://github.com/flacon/flacon/issues

 

---

 

There's always XLD as follows

 

https://sourceforge.net/projects/xld/files/

 

However, only one file was updated in September 2022 and the others seemed to be pretty old

 

https://sourceforge.net/p/xld/code/HEAD/tree/trunk/XLDFlacOutput/

 

---

 

Alternatively, Linux users might try foobar2000 via WINE instead

 

https://snapcraft.io/foobar2000

 

As always filenames and tags in plain English should be no biggie once we've converted WavPack into FLAC. Things are gonna get more complicated whenever we're dealing with double-byte characters as well as accented characters that are involved in other languages.

[LowOrbit]

It would be nice to get the new flac bersion in WTFPlay - still my player of choice for PGGB files.

[seeteeyou]

I just realized that Roon (and possibly others) could already decode FLAC @ 768 / 24 without having to touch anything at all

 

https://community.roonlabs.com/t/roon-and-ultra-highres-flac-files/112705

https://github.com/ckolivas/flac/commit/b44fcb42c9af58f04ffea82da354aa9104e7b94d

Quote

Now that many DACs support up to 768kHz, add support for these sample rates to the encoder. Files produced with these sample rates are already decodable by the reference decoder. Update documentation and tests accordingly.

Though 768 / 32 should require at least version 1.4.X for sure

 

https://github.com/xiph/flac/releases/tag/1.4.0

Quote

Encoding and decoding of 32-bit PCM is now possible

 

If our USB DACs / DDCs were actually working better with 768 / 24, maybe we're doing just fine without getting FLAC / libFLAC updated for many software players then?

 

---

 

That particular owner of MSB Reference DAC should be yet another customer of PGGB as well, I guess that could be verified by ZB then?

 

https://community.roonlabs.com/t/roon-and-ultra-highres-flac-files/112705/6

Quote

Roon’s upsampling is very simple so I’m not surprised. That’s the whole point of my approach - I use an upsampling that is so computationally complex that it would be impossible to do it on the fly.

 

His pull request (for adding support for 16FS) was submitted back in June 2020 while that wasn't merged until April 2022

 

https://github.com/xiph/flac/pull/219

 

---

 

Linux Kernel Hackers: Linus Torvalds, Andrew Morton, Hans Reiser, Alan Cox, Robert Love, Theodore Ts'o, Raymond Chen, Harald Welte, Con Kolivas

https://books.google.com/books/about/Linux_Kernel_Hackers.html?id=xg-hSQAACAAJ

 

Anyone even remotely familiar with the Linux kernel should know his name (Con Kolivas) and I didn't know that he's also an audiophile

 

https://en.wikipedia.org/wiki/Con_Kolivas

https://www.phoronix.com/news/Con-Kolivas-EOL-MuQSS-CK

Quote

Con Kolivas has worked on many patches for the Linux kernel over the past two decades and particularly focused on innovations around desktop performance/interactivity.

 

https://bitcoinmagazine.com/business/how-small-solo-miners-solve-bitcoin-blocks

Quote

Kolivas added that, since those miners made the headlines, his pool has gathered more traction and it is currently averaging over 20 PH of hash rate. But despite his vast knowledge about bitcoin mining, and the recent string of success from solo miners, he said he hasn’t mined bitcoin himself since 2017.

 

https://themarketperiodical.com/2022/02/04/what-exactly-is-ck-pool-whats-more-why-are-so-many-solo-miners-receiving-block-rewards/

Quote

Con Kolivas, a Bitcoin legend, is represented by the CK in CK Pool. The Bitcoin community erupted in applause three weeks ago when he disclosed that a single miner had received a full 6.5 BTC block reward. Then came a second announcement, followed by a third, and most recently, a fourth. Is the Bitcoin network in good working order? Is it even possible to have such odds? How can these low hash rates be winning so frequently? That’s why we’ve come to investigate.

[seeteeyou]

FLAC version 1.4.2 seemed to be fine with the Logitech Media Server + Squeeze combo then?

 

https://forums.slimdevices.com/showthread.php?116964-FLAC-1-4-2-support&p=1066343&viewfull=1#post1066343

Quote

In the process of ripping the remainder of my CDs at the moment as my trusty CD player finally bit the dust (and Naim can no longer fix them) and switched from v1.3.2 to v1.4.2 on the last batch. They are playing fine on my Touch. So I would say it seems safe enough to upgrade and the encoding time is slightly faster. Obviously YMMV as I am doing EAC rips of CDs and haven't tested with any hires audio.

 

https://audiophilestyle.com/forums/topic/62699-a-toast-to-pggb-a-heady-brew-of-math-and-magic/page/48/#comment-1200928  

On 6/5/2022 at 2:24 PM, kennyb123 said:

The software options on Antipodes are a nice plus with Squeeze sounding best for PGGB.

 

https://www.head-fi.org/threads/cx-ex-k-s-oladra-antipodes-owners-unite.932942/page-92#post-16664127

Quote

I think things will change for you when you go two-tier with Squeeze… and I mean that positively. The very lightweight Squeezelite player strikes me as the lowest latency player. Stops and starts of transients are better portrayed. It‘s also quieter with a greater sense of ease.

 

ASIO should be OK as well

 

Quote

Well, I just learned that wasapi doesn't work with Squeezelite-X. I was able to use ASIO instead, and the displayed sampling freq now corresponds to that of the music file, excellent. The bit depth still remains at 32 bit, though. I will ask in the Squeezelite-X forum about the clicks that can be heard when changing tracks or cueing in a track.

 

https://basecube.de/2018/03/01/squeezelite/

 

https://forums.slimdevices.com/showthread.php?96573-SqueezePlay-for-Windows-with-ASIO-DirectSound-WASAPI-and-WDMKS-device-support&p=861268&viewfull=1#post861268

Quote

Squeezelite supports all windows audio apis whereas squeezeplay still only works with DirectSound and MME.

 

https://forums.slimdevices.com/showthread.php?108550-Announce-Squeezelite-X&p=1019568&viewfull=1#post1019568

Quote

I was able to get FlexASIO to work. ASIO4ALL wouldn't work with my USB to Optical adapter. It was nice adding my HiRes files to the library!

 

https://forums.slimdevices.com/showthread.php?108550-Announce-Squeezelite-X&p=1019569&viewfull=1#post1019569

Quote

Excellent! FlexASIO is the most flexible/configurable, and I was going to try that next. Welcome to SLX Hi-Res!

 

Squeezelite
https://sourceforge.net/projects/lmsclients/files/squeezelite/

 

Squeezelite-R2
https://github.com/marcoc1712/squeezelite-R2/releases

 

Squeezelite-X
https://1drv.ms/f/s!AlC6xWoP5Gkwgb5YHGQ1-VHk7TQZXw

https://apps.microsoft.com/store/detail/squeezelitex/9PBHMTNP9037

 

FlexASIO

https://github.com/koord-live/KoordASIO/releases

https://github.com/flipswitchingmonkey/FlexASIO_GUI/releases

https://apps.microsoft.com/store/detail/koordasio-universal-driver/XP9CSS6NZBDV21

[kennyb123]

2 hours ago, seeteeyou said:

FLAC version 1.4.2 seemed to be fine with the Logitech Media Server + Squeeze combo then?

One challenge playing PGGB files with LMS+Squeeze is that the plug-in C-3PO is required in order to achieve PCM sample rates above 192K.  This plug-in hasn’t been updated in a while so newer release of Squeeze don’t support it.  My Antipodes box is running v1.8.4 (R2}.  We’ve tried to advance it but doing so takes away the ability to play my 16FS files.  So PGGB uses may have to miss out on whatever advancements come in terms of FLAC support.

[seeteeyou]

https://github.com/ralph-irving/squeezelite/commit/ef5fe4baf6623f57aab62da75900bc05ae627d58

Quote

Fix 24 bit 384KHz FLAC playback reported in issue #143.

 

Ralph Irving's version got 8FS handled since May 2021, let's open a new issue on his GitHub repo and see how it goes with 16FS then.

[seeteeyou]

 32 bit audio problems with Squeezebox2
https://forums.slimdevices.com/showthread.php?112848-32-bit-audio-problems-with-Squeezebox2&p=986882&viewfull=1#post986882

Quote

I installed flac and faad2 packages and copied the binaries to LMS Bin directory. Now all radios play fine

 

That's actually dated August 2020 and FLAC seemed to be OK after that's installed + copied manually.

 

Even other formats turned out to work the same way

 

https://archimago.blogspot.com/2017/11/howto-integrating-wavpack-5-compressed.html

sudo cp /usr/bin/wvunpack /usr/share/squeezeboxserver/Bin/x86_64-linux/wvunpack5

Quote

Go into "cd /etc/squeezeboxserver" and create a new custom-types.conf file

 

In other words, we could go ahead to install FLAC/libFLAC 1.4.2 ourselves.

 

---

 

https://forums.slimdevices.com/showthread.php?110642-ANNOUNCE-piCorePlayer-5-0-0&p=949774&viewfull=1#post949774

Quote

Squeezelite will default to opening the audio at 32bits. If you are playing the DAC and feeding input, they both need to be set to the same bit depth.

 

---

 

Add absurd sample rates for DSD1024 support.
https://github.com/ralph-irving/squeezelite/commit/46a16ff492f0bd6508f7bb0c09d0379cee3b07e2

https://github.com/ralph-irving/squeezelite/blob/46a16ff492f0bd6508f7bb0c09d0379cee3b07e2/pcm.c#L66

	11025, 22050, 32000, 44100, 48000, 8000, 12000, 16000, 24000, 96000, 88200, 176400, 192000, 352800, 384000, 705600, 768000, 1411200, 1536000

 

Both 16FS and 32FS were supported by Ralph Irving's version of Squeezelite since August 2021

 

https://github.com/ralph-irving/squeezelite/blob/46a16ff492f0bd6508f7bb0c09d0379cee3b07e2/squeezelite.h#L620-L621

#define MAX_SUPPORTED_SAMPLERATES 20
#define TEST_RATES = { 1536000, 1411200, 768000, 705600, 384000, 352800, 192000, 176400, 96000, 88200, 48000, 44100, 32000, 24000, 22500, 16000, 12000, 11025, 8000, 0 }

 

---

 

Interesting posts by @MrUnderhill

 

https://forums.slimdevices.com/showthread.php?106956-Streaming-Remote-FLAC-gt-PCM-amp-Local-24-192&p=875970&viewfull=1#post875970

Quote

In my system I find that PCM sounds better than flac.

 

https://forums.slimdevices.com/showthread.php?106956-Streaming-Remote-FLAC-gt-PCM-amp-Local-24-192&p=876265&viewfull=1#post876265

Quote

A few days ago I realised that in solving one problem I had changed my config and now the WAV files were being streamed as FLAC. The effect was to introduce a general feel of edge ....and cymbals lost the sound of sheen and metal but became edgy splash. Changing the config restored order .....but, it made me hope that I could listen to Qobuz streamed as PCM and hear whether this gave me a similar lift. Depending on the recording Qobuz can still sound very good, but I suspect that in my system I might be able to get a bit better - if there is a way to transcode the FLAC.

 

We could either give C-3PO a try

 

https://forums.slimdevices.com/showthread.php?105309-C-3PO-plugin-a-trnscoding-helper

 

Or simply create a file called custom-convert.conf accordingly

 

https://forums.slimdevices.com/showthread.php?111590-LMS-variable-locations-of-(custom-)convert-conf-!-!

https://web.archive.org/web/20200208104011/https://wiki.slimdevices.com/index.php/Customizing_Logitech_Media_Server_using_Slim::Utils::OS::Custom

/etc/slimserver/custom-convert.conf
/etc/squeezeboxserver/custom-convert.conf
/usr/local/slimserver/custom-convert.conf

 

https://forums.slimdevices.com/showthread.php?87260-Need-help-Some-FLACs-don-t-transcode-to-PCM&p=627320&viewfull=1#post627320

https://github.com/Logitech/slimserver/blob/65f60b6602361a1c380ec799b0d8e1bef8ce55a6/convert.conf#L181-L183

flc pcm * *
	# FT:{START=--skip=%t}U:{END=--until=%v}
	[flac] -dcs --force-raw-format --endian=little --sign=signed $START$ $END$ -- $FILE$

 

---

 

Update FLAC to version 1.4.2.

https://github.com/enzo1982/freac/commit/b42c82c317c98fda60c2c172a29e28d5da3405fc

https://hydrogenaud.io/index.php?msg=1019323

Quote

As an alternative to the suggestions offered, fre:ac works pretty well, and the current "continuous" build even supports 32bit integer FLAC, if you are into that sort of thing...

 

FLAC was updated to version 1.4.2 since October 26th

 

https://github.com/enzo1982/freac/releases/tag/continuous

[seeteeyou]

Just saw this posted on Facebook

 

https://www.facebook.com/hiroyuki.yokota/posts/5900923503273059

Quote

The precision of pi has been changed from 49bits to 196bits, and the double-precision arithmetic has been increased from 2x to 4x. The double-precision Resample operation table has the same operation performance as before, but the purity of the sound has been greatly improved. The performance of the up-sampling processing has greatly exceeded the standard for professional use and is close to the high-end standard. The number of digits deteriorates when moving data from the double-precision register to the double-precision memory variable when setting pi, and once the matter was understood, the story suddenly took off.

 

Further details linked below could be translated to English accordingly

 

https://forums.ubuntulinux.jp/viewtopic.php?pid=125180#p125180

https://archive.today/iHvA7

 

BugHeadResampler147.exe and Bug_head_Resampler_Manual.pdf could be downloaded here, they're part of the entire package

 

http://www.mics.ne.jp/~cdorya/#download

 

I wonder if there were any potential for collaborations with that developer?

 

---

 

https://www.whatsbestforum.com/threads/taiko-audio-sgm-extreme-the-crème-de-la-crème.27433/page-184#post-646651

Quote

An example of a software player with no bloat would be something like Bug Head Emperor which had a cult following once upon a time because it focused on doing only one thing well and that was excellent SQ. It had an archaic interface but the whole program was only about 60MB.

 

https://www.whatsbestforum.com/threads/taiko-audio-sgm-extreme-the-crème-de-la-crème.27433/page-443#post-770097

Quote

Strange things can happen in software, an early pioneer in this field has been the author of Bughead Emperor. The things that are coded into TAS and XDMS are grounded in the ways PC's electrically function.

 

http://jplay.eu/forum/index.php?/topic/4410-windows-11-pe-audiophile-creation-guide/?p=58068

Quote

Right now F2K and Infinity Blade are far superior. Comparing HQP (WinServCore2019 RAMDISK) against HQP (Win10-11XPE) the latter is clearly superior, the same goes for JRMC26, it sounds much better on Win10-11XPE, but both clearly worse than F2K.

Infinity Blade in "normal" Windows is much higher than F2K, on the other hand in Win10-11XPE the difference between the two is shortened enough, and if you also play the audio track from "add / send to ramdisk", the difference is still less , always, yes, in favor of Infinity Blade.

 

http://jplay.eu/forum/index.php?/topic/4410-windows-11-pe-audiophile-creation-guide/?p=61837

Quote

Thanks to F2k beta 14 I can have a W11PE Audio without SysWow 64 and with drivers (IFIAUDIO and DIYINHK), F2K 2.0 beta 14 and Bughead (Infinity Blade) I have been able to reduce the size of Windows 11 PE Audio (not XPE) up to 264 MB !!

 

Keeping everything under 300 MB, that's pretty much like as good as it gets?

[Zaphod Beeblebrox]

10 hours ago, seeteeyou said:

Just saw this posted on Facebook

 

https://www.facebook.com/hiroyuki.yokota/posts/5900923503273059

Quote

The precision of pi has been changed from 49bits to 196bits, and the double-precision arithmetic has been increased from 2x to 4x. The double-precision Resample operation table has the same operation performance as before, but the purity of the sound has been greatly improved. The performance of the up-sampling processing has greatly exceeded the standard for professional use and is close to the high-end standard. The number of digits deteriorates when moving data from the double-precision register to the double-precision memory variable when setting pi, and once the matter was understood, the story suddenly took off.

 

Further details linked below could be translated to English accordingly

 

https://forums.ubuntulinux.jp/viewtopic.php?pid=125180#p125180

https://archive.today/iHvA7

 

BugHeadResampler147.exe and Bug_head_Resampler_Manual.pdf could be downloaded here, they're part of the entire package

 

http://www.mics.ne.jp/~cdorya/#download

 

I wonder if there were any potential for collaborations with that developer?

PGGB-AP currently does 192 bits of precision (6x) end to end, this includes all processing, all constants (like pi), all filter coefficient computation, storage in memory etc,  so the precision is never changed until the final output that has to be at a lower bit depth (24, 32 etc). Here too the noise shaper operates at 192 bits.

The PGGB-256 version I am currently will do all the computation at 256 bits, but is much faster and also has much reduced memory requirements. the upsampling algorithms are also completely revamped.

[Cedric79]

Quick question - if limited by memory (16GB), what provides best performance, setting precision to 128bit and having the taps max out at 128M, or set precision to 64bit and have the max amount of taps for tracks lengths. Typically for a usual Redbook album where tracks are between 3-4:50min. Of course some tracks will fall below the 2:80min cap where taps remain below 128M by nature, but some tracks will be slightly above the 128M taps.

On 64bits with 16GB all tracks would be well below the 256M taps limit.

 

hope my question makes sense!

[Zaphod Beeblebrox]

8 hours ago, Cedric79 said:

Quick question - if limited by memory (16GB), what provides best performance, setting precision to 128bit and having the taps max out at 128M, or set precision to 64bit and have the max amount of taps for tracks lengths. Typically for a usual Redbook album where tracks are between 3-4:50min. Of course some tracks will fall below the 2:80min cap where taps remain below 128M by nature, but some tracks will be slightly above the 128M taps.

On 64bits with 16GB all tracks would be well below the 256M taps limit.

 

hope my question makes sense!

You should try both and compare on few tracks that hit the 256M limit with 64bits, you could also increase the virtual memory and set the taps to 256M (will be slower). It is subjective and hard to provide a definitive answer.

[Zaphod Beeblebrox]

There have been numerous requests to improve memory performance of PGGB, especially to allow high precision computing without the need to invest in a lot of RAM or in Server grade hardware. We have been working on addressing this while improving on the algorithms too. Many thanks to those who have been patiently waiting and many thanks to the beta testers for their feedback and encouragement, we have been working on this for the past 9 months and constantly improving. There was even an intermediate version (v4) which we skipped over to bring v5.

 

PGGB 256 Infinity and Beyond Edition

I am happy to announce PGGB 256 v5.0.42. PGGB-AP is now PGGB 256. PGGB 256 uses completely revamped algorithms, but still follows the same fundamental principle of getting as close to perfect reconstruction as possible. The new algorithms are optimized to maximize reconstruction accuracy to near theoretical limits, and it improves over what PGGB-AP did.

• PGGB 256 uses highly optimized custom built multi precision math libraries. PGGB 256 is now faster and more memory efficient than ever before.
• PGGB 256 uses a minimum of 128bit precision for resampling (even for 64bit processing), one can optionally use up to 256bit precision for all of the processing.
• The configuration menu is much simplified and there is an automatic mode that now chooses the optimal settings based on each track. More details can be found here.
• PGGB 256 does away with having to choose taps, does not use windowed sinc function and is non-apodizing.
• PGGB 256 now has a 'XDMS Mode' under the hidden menu to split output files into 1.25GB parts to aid play back in Taiko Audio's XDMS.
• PGGB 256 now supports 64bit output that is noise shaped, this is useful if you wish to process these tracks further.

 

Licensing options have changed

There are now three types of licensing in addition to the free trial. More information can be found here.

1. Trial licenses: They have been simplified, PGGB 256 will automatically start in trial mode and one need not request a trial license separately. As before it limits processing to 5 tracks at a time for 64 bits and 2 tracks at a time at 128 or 256 bits.
2. Perpetual (64): Existing PGGB-AP perpetual license holders are automatically upgraded to 'Perpetual (64)' license for PGGB 256.
   1. You can run previous version (v3 or older) side by side with PGGB 256.
   2. PGGB 256 will run without any limitations when set to 64bit processing. Even when set at 64bit processing the resampling is done at 128bits and one also gets the full benefits of the new revamped algorithms.
   3. When set to 128 or 256bit, processing is limited to 2 tracks at a time (one can restart to process couple of tracks again)
3. Perpetual (256): This allows end-to-end processing at 256 bits without any limitations.
4. Upgrade to (256): Existing PGGB-AP perpetual license holders or 'Perpetual (64)' license holders can upgrade to 'Perpetual (256)' license.

Where is my cheese

If you have been using PGGB-AP, PGGB 256 uses a different configuration file for settings. This allows PGGB 256 to coexists with previous versions. It also means, you have to setup PGGB 256 again with your default settings (which it will remember going forward).
Some of the previous settings are no longer available as they are not relevant:

1. Most notably, no more taps: PGGB 256 does away with the concept of taps. The new algorithms do not limit the quality of output based on the available resources, it uses an adaptive processing pipeline to achieve the same reconstruction quality with the available resources*. 
2. Transparency and presentation options are gone: With windowed sinc function, there is always a choicebetween time domain and frequency domain performance, and these controls provided the parameters to tweak windowing. PGGB 256 does not use windowed sinc functions and these compromises are no longer needed.
3. Apodizing check box: The check box is gone as PGGB 256 is non apodizing.
4. Hidden menu option DSD throttling: This option is no longer needed and if TDP is an issue, it can be controlled by setting the number of workers.
5. Hidden menu option Disk Caching: This option is gone because PGGB 256 is a lot more memory efficient. 

*PGGB 256 still needs enough memory available to hold input and output buffers at the desired precision, the processing however is designed to trade speed to fit into memory. The required memory proportional to the length of the track.

[CNoblet]

Thank you!!!  Thank you, thank you, THANK YOU!!!