1
General - (fb2k) / Re: Hires Internet Radio - Buffering
Last post by boxerfan88 -Any way for me to experiment/tweak the network stream receiver thread and bump up it's CPU priority?
Notice
Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Any way for me to experiment/tweak the network stream receiver thread and bump up it's CPU priority?
2
General - (fb2k) / foobar2000 v2.25 preview not rendering DSD higher than DSD128
Last post by papavlos -It has following plugins from manisiutkin:
- foo_input_sacd
- foo_out_asio+dsd
and also the abandonned 10 yeqars old plugin "foo_upnp" from bubbleguuum with "renderer" service started, with different name: "plug-in renderer".
The computer has an USB DAC with DSD capabilities attached and configured.
When I play with foobar DSD tracks from local Media library - I can play DSD of all probing frequencies,, also DSD256. So all plugins from manisiutkin work properly.
The problem is when I wand to use this main computer as a remote UPnP renderer.
So I have setup foobar v2.25 on a second PC and configured it to read DSD audio files (foo_input_sacd).
When I stream from the second PC to the "plug-in renderer" - the first foobar plays all types of DSD,. The DAC driver shows it is in ASIO mode and processing DSD input stream, although the foobar displays in its status bar as if it was DoP-ped PCM stream.
When I stream from the second PC to the "built-in renderer" - the first foobar also properly plays DSD64 and DSD128. The DAC driver also properly shows ASIO and DSD frequency. And in the status bar foobar also reports DSD64 and DSD128, instead of PCM stream with the "plug-in renderer" selected.
But when I start streaming a DSD256 track - there is only silence from the DAC, and its driver reports that there is no input stream. Although the status bar still properly reports processing DSD256. I am also monitoring network traffic - there is an incoming data stream over the network.
Finally my report:
My diagnosis is that in the newly built-in UPnP renderer there must be something blocking proper processing of DSD streams frequencies higher than DSD128.
Please be so kind and check it.
3
Support - (fb2k) / Re: Very long image search (only with Linux / Foobar2000V2)
Last post by AldiMp3 -Result: even the 32-bit version 2.25 does not create the image in an acceptable time.
Then I checked whether it only affects the album art elements front artist back or disc or also the playlist view
Result: The playlist does not deliver any images in a reasonable time either.
To narrow down the problem even further, I only selected front.
Result: The front images are there immediately when scrolling and all album art elements now also show 4 x front immediately.
Conclusion: The problem only arises when other images and not just the front cover are to be viewed.
The next attempt to attach the pictures with batch attach picture confirms the linux time problem. After 5 minutes only 8 pictures were inserted. After a short switch to Windows 1.6.18 more than 1500 pictures were embedded within a few seconds.
After starting Linux preview 2.25 32-bit and 64-bit, these images fly into Linux at high speed.
4
Other Lossy Codecs / Re: lossyWAV 1.4.2 Development (was 1.5.0)
Last post by C.R.Helmrich -As test content, the following four CD audio (44.1-kHz 16-bit stereo) datasets were selected so as to cover a wide range of waveform material familiar to the HydrogenAudio community:
- IgorC's transform codec set (12 files, total audio duration 3:31 min) from here
- Kamedo2's audio codec set (20 files, total audio duration 7:49 min) from here
- guruboolez's LPC codec set (15 files, total audio duration 3:40 min) from here
- high-rate audio: Merzbow, annoyingloudsong, SQAM triangle (51 s), as above.
Note that guruboolez's set turns out to be very useful beyond improving or testing WavPack: it's a highly versatile and challenging set of samples to any codec, I'd say, not just WavPack.
All codec candidates chosen for evaluation were configured for maximum coding efficiency, thereby accepting slowdowns during encoding, and with default perceptual optimizations on unless noted otherwise. The chosen target bitrate in all cases was roughly 320 kbps stereo, since this is the lowest average bitrate supported by all candidates being tested. The decoded transform coded streams were stored as 24-bit PCM, and for the computations of the SNR, segmental SNR, DI, and ODG metrics, all decodings were resampled to 48 kHz and saved as 16-bit PCM (with proper rounding but no dither). Resampling and metric calculation was done per-file using the AFsp v9 r0 toolset, except for the PSNR results, which were determined per-file via Matlab/Octave on the non-resampled waveforms.
| Codec | Avg. Bitrate* | Highest Rate | Avg. SNR | Avg. opt. SNR | Avg. Seg SNR | Avg. PSNR | Avg. Dist. Index | Avg. ODG | Worst ODGs |
| LossyWAV beta 1.4.3h, b512 | 325 kbps | 722 kbps | 36.092 | 36.106 | 36.249 | 60.615 | 2.1311 | -0.3110 | -1.924, -1.209, -1.061 |
| LossyWAV feedback 0, b512 | 292 kbps | 609 kbps | 32.110 | 32.130 | 32.897 | 56.640 | 1.6850 | -0.5234 | -2.350, -1.738, -1.209 |
| FSLAC, lower-rate -2, b1024 | 351 kbps | 505 kbps | 38.025 | 38.083 | 41.367 | 62.590 | 1.6765 | -0.5045 | -2.027, -1.475, -1.130 |
All SNR related results in dB, *: as reported by foobar2000, excluding files of highest and lowest rate and all files of rates < 90 kbps
As you can see from the objective difference grade (ODG) and, to some extent, distortion index (DI) averages, modelling subjective impression of audio coding quality, lossyWAV performs roughly as well as FSLAC - in terms of average ODG for the "--feedback 0" version, and in terms of worst-case ODG for the "default feedback" version - but requires a lower bitrate than FSLAC. So well done
and with some more tweaking of Merzbow/applause-like content, you could reduce the rate even further without (probably) affecting the audio quality much. FSLAC seems to have the best rate-performance tradeoff with FLAC block-size 1024. Note that the SNR related values do not correlate with subjective audio quality - the table shows this quite clearly - and are only listed for completeness.If you want, I can send you the detailed analysis logs, allowing you to link the data to the individual test samples.
Chris
5
Audio Hardware / Re: Question in-ears magnetic driver
Last post by Porcus -How can a loudspeaker have a treble unit close to a bass unit without the bigger one demagnetizing the smaller?
Here they tried to ruin data on hard drives with quite strong magnets: https://www.kjmagnetics.com/blog/erasing-hard-drive-with-magnets . They failed. Not saying it is safe if you want to keep your data, but it is useless if you want to erase data.
This is the magnet: https://www.kjmagnetics.com/dx0x8-neodymium-cylinder-magnet . Takes a 30 kg pull to get it off a steel plate.
6
Audio Hardware / Re: Question in-ears magnetic driver
Last post by fooball -But why would you want to have them that close anyway?
7
Audio Hardware / Re: Question in-ears magnetic driver
Last post by classicaran88 -8
3rd Party Plugins - (fb2k) / Re: Library Tree Discussion
Last post by Tropoje -9
Scientific Discussion / Re: ARDFTSRC - Real Discrete Fourier Transform Sample Rate Conversion
Last post by mycroft -10
General - (fb2k) / Re: Album Information Genre drop down list
Last post by kajero -Thank you for all your help. I now have "clean" lists. I have your replies bookmarked for reference so I can fix the lists when I mess them up again. Thanks again.
One more question: How do I mark this question as "solved"?