Lukasluis Posted May 2, 2021 Share Posted May 2, 2021 1 hour ago, edwardsean said: Okay, okay, okay, I've converted to 421. I kept playing with 421 and what finally did it was that I disabled Ramroot. That alleviated the tightness I was experiencing in 421's presentation. 421 has this incredible precision and definiteness where 102 can image overly soft and diffuse. However, with Ramroot engaged, the sound feels like the screws are overtightened and it feels stiff. I don't like giving up Ramroot, but once I unloaded it, a natural sense of flow returned. I still long for more of 102's fullness, larger low end and staging. Also, I am uncomfortable losing Ramroot from a technical standpoint. I had to remind myself that the only that counts is the end result sound. Well, you can't have everything. This is still quite a good net win, and I'm glad to be going forward with Euphony! I too felt the tightness or more mechanical presentation but since I use tubes, I just reduced the bias current of the tubes and voila! Topk 1 Link to comment
Anwar Posted May 2, 2021 Share Posted May 2, 2021 My preferred Euphony version remains 217. My DAC is actually a CD/SACD player (Marantz SA-10), so every streamer or Euphony versions I tried, I compare its SQ with internal rendering by SA-10, either playing discs or from USB thumb drive attached behind, which I consider my reference SQ since there are no other external components involved, everything happened inside SA-10. 217 comes really close to matching the SQ of SA-10 rendering, but not the other Euphony versions. My streamer: DIY multi-rail LPSU powering ASRock J5040-ITX Pentium Silver with turbo boost turned off. Separate 5V rails for SSDs. Apacer consumer grade RAM, 32GB total. I use Stylus, playing music from SSD only. Music server builder Link to comment
Joerg D Posted May 2, 2021 Share Posted May 2, 2021 Hello I have reduced the Ethernet speed to 100 Mbit in the Expert Setup. The SQ has improved somewhat. The sound has become a bit finer, more relaxed and more musical. -s eth0 speed 100 duplex full autoneg off Jörg Link to comment
ASRMichael Posted May 2, 2021 Share Posted May 2, 2021 1 minute ago, Joerg D said: Hello I have reduced the Ethernet speed to 100 Mbit in the Expert Setup. The SQ has improved somewhat. The sound has become a bit finer, more relaxed and more musical. -s eth0 speed 100 duplex full autoneg off Jörg The problem is buffering tracks take so long with 100mbit. Have you tried disconnect network option? Works well for me. Add album or playlist and sit back and enjoy with network connected. Link to comment
di-fi Posted May 3, 2021 Share Posted May 3, 2021 20 hours ago, edwardsean said: Okay, okay, okay, I've converted to 421. I kept playing with 421 and what finally did it was that I disabled Ramroot. That alleviated the tightness I was experiencing in 421's presentation. 421 has this incredible precision and definiteness where 102 can image overly soft and diffuse. However, with Ramroot engaged, the sound feels like the screws are overtightened and it feels stiff. I don't like giving up Ramroot, but once I unloaded it, a natural sense of flow returned. I still long for more of 102's fullness, larger low end and staging. Also, I am uncomfortable losing Ramroot from a technical standpoint. I had to remind myself that the only that counts is the end result sound. Well, you can't have everything. This is still quite a good net win, and I'm glad to be going forward with Euphony! I appreciate your descriptions of how the sound is evolving through versions. You improved without (!) ramroot, so now sound quality depends less on RAM. To better understand, would you mind sharing what type of RAM you are (not with 421 !) using and also what other type of memory to run Stylus is attached to your mobo? In the situation above you do streaming, SSD or NAS? I looked in your setup details but you did not mention. If you did somewhere else please link to that part of the thread. Thanks. Link to comment
Popular Post edwardsean Posted May 3, 2021 Popular Post Share Posted May 3, 2021 48 minutes ago, di-fi said: I appreciate your descriptions of how the sound is evolving through versions. You improved without (!) ramroot, so now sound quality depends less on RAM. To better understand, would you mind sharing what type of RAM you are (not with 421 !) using and also what other type of memory to run Stylus is attached to your mobo? In the situation above you do streaming, SSD or NAS? I looked in your setup details but you did not mention. If you did somewhere else please link to that part of the thread. Thanks. I'm using Apacer RAM and an Optane drive, so I very much want to use Ramroot. I'm still hoping for a version that has greater fluidity and space (esp. soundstage width), and the added performance of Ramroot. The great thing about 102 is the way the sound flows freely, unrestricted into an expansive space. My ideal audio components are able to do this while occupying that space with dense and precise images. NanoSword and di-fi 1 1 Link to comment
Speedy381 Posted May 3, 2021 Share Posted May 3, 2021 I am using 421 and whilst appreciative of its detail, I too find it very etched. I tried disabling ramroot and was very surprised by how much the sound was 'dumbed down' but easier to listen to. I will have to alternate to decide which I prefer, but it does feel wrong to compromise performance to gain listenability. I am also using Apacer ram and am Optane drive. edwardsean 1 Link to comment
Popular Post edwardsean Posted May 3, 2021 Popular Post Share Posted May 3, 2021 5 hours ago, Speedy381 said: I am using 421 and whilst appreciative of its detail, I too find it very etched. I tried disabling ramroot and was very surprised by how much the sound was 'dumbed down' but easier to listen to. I will have to alternate to decide which I prefer, but it does feel wrong to compromise performance to gain listenability. I am also using Apacer ram and am Optane drive. This is similar to my impressions, though in my system, I wouldn't describe the issue as "etched" so much as "tight." 421 was definitely smoother than 216/217, but was missing 102's ease and fluidity. So, I kept experimenting and found, at least for myself, the optimum solution: I reengaged Ramroot but removed core isolation. To my ears this is the best balance of compromises. The staging still seems smaller to me, but I think that is more of a psychoacoustic side effect of 421's greater detail and precision. That improvement in performance is the highlight of 421! The clarity is breathtaking at times and worth the effort to get the other parts sorted. So, without getting into the technicalities (I don't understand), this is the way I make sense of it as an end-user: 1) On 102 the sound was expansive, flowing, but diffuse. Engaging Ramroot and pushing core isolation "tightened" up the sound, providing more precision. My core isolation was set to the extreme: 0 gstp 0-7. 2) 421 already has such good accuracy that removing Ramroot and core isolation sounds akin to 102 with both Ramroot and core isolation. So, if I go to 421 and engage both Ramroot and core isolation the sound gets so "tight" that flow is lost. 3) On 421, if I remove Ramroot and engage core isolation, the sound eases up, but still feels restricted. 4) On 421, if I engage Ramroot but remove core isolation, I do lose some accuracy but there is that natural sense of flow again. It hits really close to that nexus of "creamy clarity," that for me is "magic" because, actually, it sounds so real. I would've tried this earlier except I am an inveterate tinkerer. I'm sure you guys can't relate, but It makes me both happy and itchy when things just work out of the box. I just couldn't imagine forgetting all the time I put into comparing every permutation of CPU allocations. But, here I am, at "default," all processes sharing all cores. I'm sure if I had a different system, that could easily change. For now, I know I've found my current optimal setup. I don' have any quantifiable measurements... except the hours of sleep I lost last night just listening to, what's that thing again? Music. beautiful music, RickyV, NanoSword and 6 others 4 5 Link to comment
di-fi Posted May 4, 2021 Share Posted May 4, 2021 22 hours ago, edwardsean said: I'm using Apacer RAM and an Optane drive, so I very much want to use Ramroot. I'm still hoping for a version that has greater fluidity and space (esp. soundstage width), and the added performance of Ramroot. The great thing about 102 is the way the sound flows freely, unrestricted into an expansive space. My ideal audio components are able to do this while occupying that space with dense and precise images. Thanks. For these listening sessions are you playing local files or streaming files? Link to comment
edwardsean Posted May 4, 2021 Share Posted May 4, 2021 44 minutes ago, di-fi said: Thanks. For these listening sessions are you playing local files or streaming files? Always local. I have a Tidal account for exploring, but whenever I find something I love, I purchase the file. Link to comment
Popular Post TheAttorney Posted May 4, 2021 Popular Post Share Posted May 4, 2021 Having stopped playing about with core isolation, something forced me right back: One day, the sound seemed "off" and, on checking dot, I found the reason: My primary xhci should have been allocated to irq/131, but instead that number was taken by something called ahci (and xhci was allocated irq/132). I've since noticed that these two do occasionally swap numbers after reboot. Now, ahci seems to do nothing at all (the irq count never goes above 0), but the SQ degradation was because the all-important xhci was randomly allocated to a shared core. I fixed this by allocating both 131 and 132 to the same core, so it wouldn't matter which way the swap went next time. Whilst in the area, I took the opportunity to try out an idea inspired by others' recent posts - and I now have a new favourite 🙂 : 2-3 systemd 0 nfm 1 gstp 5-7 irq/131 4 (the full list includes the less significant irq/132 4 irq/135 3) The key thinking behind this is: Find the really important things to isolate and let the o/s sort out all the rest. So, important to completely isolate was the primary irq, systemd, and new for this time nfm, and of course gstp. And lump stylus and various daemons in with "everything else". I put "everything else" into virtual cores 2-3 simply because I wanted to put less busy processes into 0-1 because physical core 0 has traditionally run slightly hotter than the other cores. My OCD like balance. Trying slight variations of above (like dropping nfm) yielded inconclusive results, so there may well be a still better combination, but I've stuck with the one that first gave a me a Wow response. It wins at all levels: it's got all the vividness and focus as before, but now with extra fullness, flow, soul, and a deeper, more powerful, yet still-tight, bass - all helping to add some very welcome depth and spaciousness to the sound stage. What's not to like? To show that any change may impact other parameters, my previous favourite CPU speed 1.2GHz is now beaten by 1.3 or 1.4 GHz. Following edwardsean's example, I also tried no isolation (0-7) and non-ramroot. Both sounded fine on 421, but neither was as good as my new favourite on ramroot and, with the proviso that all such changes are probably objectively tiny, the subjective impact is considerable. I'm done with changes now. Really I am. edwardsean, RickyV, VoicesInMyHead and 4 others 1 5 1 Link to comment
di-fi Posted May 4, 2021 Share Posted May 4, 2021 11 hours ago, edwardsean said: Always local. I have a Tidal account for exploring, but whenever I find something I love, I purchase the file. That's what I thought, same here. Best results with local files. In my case all FLAC on a Synology NAS +LPS > switch > 30ft Cat7 > streamer. Link to comment
mrkoven Posted May 5, 2021 Share Posted May 5, 2021 Is there any issue with stylus playing pggb 32fs? Link to comment
austinpop Posted May 5, 2021 Share Posted May 5, 2021 42 minutes ago, mrkoven said: Is there any issue with stylus playing pggb 32fs? I don't know about 32FS, but a local friend has successfully run 32/16FS through Stylus to his iFi Pro iDSD DAC, running Euphony on a NUC7i7DNBE. mrkoven 1 My Audio Setup Link to comment
edwardsean Posted May 5, 2021 Share Posted May 5, 2021 10 hours ago, di-fi said: That's what I thought, same here. Best results with local files. In my case all FLAC on a Synology NAS +LPS > switch > 30ft Cat7 > streamer. Have you tried Wav/AIFF instead of FLAC? Even though flac is lossless compression, I do find uncompressed files perform better. I’m not sure whether that’s due to artifacts at the compression stage, decompression stage, or both. Some think that it’s due to the noise of the extra processing load of decompressing the files. So, if they have flacs they decompress them offline and store them as wav for playback. It’s worth the experiment unless you’ve already ascertained that it makes no difference for yourself. Link to comment
di-fi Posted May 5, 2021 Share Posted May 5, 2021 6 minutes ago, edwardsean said: Have you tried Wav/AIFF instead of FLAC? Even though flac is lossless compression, I do find uncompressed files perform better. I’m not sure whether that’s due to artifacts at the compression stage, decompression stage, or both. Some think that it’s due to the noise of the extra processing load of decompressing the files. So, if they have flacs they decompress them offline and store them as wav for playback. It’s worth the experiment unless you’ve already ascertained that it makes no difference for yourself. Thank you for pointing this out. I did try to compare FLAC vs. WAV/AIFF in the past, downloaded both versions from the same album (Bandcamp), but it was not conclusive. Also my system might not be resolving enough. I will give it a try again, my system evolved a little over time, it could be worth it. Link to comment
edwardsean Posted May 5, 2021 Share Posted May 5, 2021 Try XiSRC. As you know, not all sample rate converters/encoders are the same. XiSRC was fantastic for the low price, and there is a Mac and Win version. Now that the company, Xivero, is no longer in business, I think it may just be free. You can find download links online. It can be a bit slow compared to the common free algorithms out there, but I think more precise. di-fi 1 Link to comment
Johnseye Posted May 5, 2021 Share Posted May 5, 2021 1 hour ago, mrkoven said: Is there any issue with stylus playing pggb 32fs? With or without HQP? There were some issues at first but they were resolved. Should work fine. Audio System Link to comment
mrkoven Posted May 5, 2021 Share Posted May 5, 2021 1 hour ago, Johnseye said: With or without HQP? There were some issues at first but they were resolved. Should work fine. with HQP. good to know issues were resolved, i plan to try myself soon Link to comment
di-fi Posted May 6, 2021 Share Posted May 6, 2021 On 5/4/2021 at 11:56 PM, edwardsean said: Try XiSRC. As you know, not all sample rate converters/encoders are the same. XiSRC was fantastic for the low price, and there is a Mac and Win version. Now that the company, Xivero, is no longer in business, I think it may just be free. You can find download links online. It can be a bit slow compared to the common free algorithms out there, but I think more precise. Interesting! In my quest to optimize local files to play with Euphony Stylus I tried XiSRC (free*). The same company provides MusicScope (also free**) to examine the characteristics of files. MusicScope allowed me to visually compare the original FLAC (33Mb) vs, the WAV provided by Bandcamp (50.4Mb) and the WAV I converted myself from FLAC with XiSRC (also 50.4Mb) . In MusicScope all 3 look identical. Soundwise, at first listen, I can not confirm a difference between the versions of the selected file. This might go beyond the subject here (you can also PM me) and I would like to explore more, but are there any specific settings in XiSRC that you can refer to for optimizing/improving the WAV file vs. the original FLAC? *https://sourceforge.net/projects/xivero-audio-tools/ **https://getwinpcsoft.com/MusicScope-3870859/download/ Link to comment
di-fi Posted May 6, 2021 Share Posted May 6, 2021 * CPU utilization % and temperature are the same while playing the FLAC or WAV file in Euphony Stylus with ramroot (see previous post). I definitely give it a good listen with other tracks. Link to comment
edwardsean Posted May 7, 2021 Share Posted May 7, 2021 12 hours ago, di-fi said: Interesting! In my quest to optimize local files to play with Euphony Stylus I tried XiSRC (free*). The same company provides MusicScope (also free**) to examine the characteristics of files. MusicScope allowed me to visually compare the original FLAC (33Mb) vs, the WAV provided by Bandcamp (50.4Mb) and the WAV I converted myself from FLAC with XiSRC (also 50.4Mb) . In MusicScope all 3 look identical. Soundwise, at first listen, I can not confirm a difference between the versions of the selected file. This might go beyond the subject here (you can also PM me) and I would like to explore more, but are there any specific settings in XiSRC that you can refer to for optimizing/improving the WAV file vs. the original FLAC? *https://sourceforge.net/projects/xivero-audio-tools/ **https://getwinpcsoft.com/MusicScope-3870859/download/ It’s a really simple process. You should just keep the same sample rate and bit depth. You can leave ISP off. That’s it. I have to say, my own suspicion is that the difference has to do with the added load on the CPU to decompress the file from flac to Wav during playback. I can clearly hear a difference with flac being more smeared and mushy. But, you will need a fairly resolving/ low noise system to detect these differences. In your system, if you have to strain to hear any difference, please, don’t let this distract you from enjoying flac files! di-fi 1 Link to comment
davide256 Posted May 13, 2021 Share Posted May 13, 2021 On 3/20/2021 at 5:16 AM, flkin said: If your motherboard allows, try changing speed and voltage of the ram I had good results doing this. Just ordered this Apacer RAM, doubt I can make use of the higher speed (3200 ) in an 8i7BEH but if it behaves better for audio, thats what matters. Will report on results, this is more available than the lower speed Apacer DDR4 SODIMM (2400) https://www.mouser.com/ProductDetail/908-D22.27261S.001 Regards, Dave Audio system Link to comment
StreamFidelity Posted May 13, 2021 Share Posted May 13, 2021 29 minutes ago, davide256 said: Just ordered this Apacer RAM These are pretty high latencies: MHz: 3200 CL: 22 Calculated nanoseconds: 13.75 I prefer RAM with CL latency up to 8.8ns. Probably not available from the Apacer brand. justmy2cents Grigg Audio Solutions Owner StreamFidelitys Setup: Sonus Faber Amati Futura | T+A M10 | T+A SDV 3100 HV | fis Audio PC & Server | GigaWatt PC4-EVO+ | JCAT OPTIMO S ATX | FARAD Super10 & Super3 | Keces P8 | Afterdark Buffalo Switch | fis Audio Cables | Solidsteel HJ-3 / HY-A Link to comment
davide256 Posted May 13, 2021 Share Posted May 13, 2021 1 hour ago, StreamFidelity said: These are pretty high latencies: MHz: 3200 CL: 22 Calculated nanoseconds: 13.75 I prefer RAM with CL latency up to 8.8ns. Probably not available from the Apacer brand. justmy2cents Mehh... I'm looking for lower electrical noise feeding back into the PS bus and screwing with USB out. Speed isn't relevant unless I use HQPlayer and using HQPlayer in NUC USB out server creates its own noise issues. Regards, Dave Audio system Link to comment
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now