Last post by ajinfla -
At the 2:34> mark I can hear ticks, pops and noise, including what sounds like talking or maybe a tv going in background? WTH?
This is listening with in living room with custom active 8" 2 way monitors. I'm sure the headphone guys will chime in with more detail.
Lastly, I realize its a Zoom recording in room, but the SQ is abysmal (per my >audiophile standards). Hopefully/should be much better on site
Last post by Audible! -
Realtek, the principal hardware audio CODEC manufacturer, has long produced ICs that are technically capable of exceeding CD resolution. Implementation of CODEC ICs, on the other hand, has varied. Relatively recently, motherboard manufacturers have begun to pay close attention to the implementation of the audio subsystems of their boards as a selling point, but even the 'mid-range' contemporary Realtek CODECs have 'supported' 24bit word length and 96Khz+ sampling rates for a number of years. 7.1 output has also been widely available.
So what is a poor CODEC or motherboard manufacturer to do to differentiate their high end products these days?
It's actually a bit difficult to tell what Realtek even claims their new CODEC products are capable of, because no publicly available data sheet appears to exist for their most current products. This makes some sense because the public is not their customer; the board makers are.
I have been curious as to what novel, exciting, and totally unnecessary features the ALC1220 (the current 'high end' Realtek CODEC) might be technically capable of supporting. Conveniently, it was time to update my main desktop system, and AMDs Ryzen 7 CPUs can now be found for less than $300 in retail. This being the case, I've acquired a Gigabyte X370 'Aorus gaming K5 motherboard' which features the ALC1220 with fancypants nichicon 'gold audio' capacitors in the audio stage. And das kolorful blinkenlights, for some reason.
-In the audio driver app., resolutions up to and including 32bit/192 KHz are supported by the CODEC. Laughable. Seems pointless without 384KHz sampling
-Motherboard product literature claims that (the IC maker doesn't even list the chip on their site yet) the main stereo DAC of the ALC1220 has a 120dB SNR. Quite impressive if true; these CODEC chips are definitely not particularly expensive.
-__ing DSD has now officially penetrated the onboard audio CODEC IC industry! Motherboard claims 'support for DSD128 hardware decode'. Oy vey. I wonder if this means software and supporting hardware will play the SACD layer of hybrid discs; more likely this is referring to just the data type. This means SACD rips could theoretically be supported. At long last.
-Tweaktown's review of this motherboard performed an RMAA loopback test (likely limited by the CODEC ADC) and found a dynamic range (A) of 108dB, with
THD at a shocking 0.0034% and IMD@10kHz being 0.004%. Initial listening (Paradigm mini monitor/NAD high current amp system) finds output quality to be totally indistinguishable from my Asus Xonar sound card with 16 and 24 bit FLAC audio samples. Shocking, truly. In the future I'd like to run RMAA at multiple sampling rates, with the Xonar capturing the motherboard output rather than a loopback. It may be that the Xonar ADC is better than that of the ALC1220 and could provide a clearer picture of what the DAC end of the ALC1220 is capable of.
Essentially any competent implementation of this (or nearly any current) CODEC should be absolutely excellent if the Aorus X370 K5 is representative.
This particular board has a feature Gigabyte calls 'DAC-UP' - two of the USB ports (one front, one rear) are capable of being overvolted by up to 0.3V DC in order to provide more power to external DAC/headphone solutions. Hard to imagine why you'd bother with an external solution with this board, perhaps 600 ohm headphones.
I forgot how nice Gigabyte boards are - they include a little plastic adapter to plug the LED/power/reset leads into so you have one big plug to attach to the board. Why everyone doesn't do this is beyond me.
The Ryzen 7 is really quite an unreasonable CPU - the six core/12 thread chips are a bit more sane. I'm rather irritated with AMD for making a chip like this, because I don't know how to utilize all of its resources. On some occasions I've maxed out quad core systems but that's just not going to happen here. Happily the chip is extremely energy efficient, and the Threadripper CPUs exist exist, helping to make me feel somewhat less ridiculous.
Edit: for any system builders, my board shipped with the 'F3' BIOS and works flawlessly with EVGA 'SuperSC' DDR4 at the advertised 3000/cas 15 (2x8GB).
Yes, I got it you need it for voice, which setting do you use for voice? Again, I think 64 mono and 128 stereo are more than enough.
Attached a new file:
- Added 96 kbps CBR for MP3, Marco makes good points here: https://marco.org/2016/08/15/vbr-mp3-plea and audio here https://overcast.fm/+CdTym-vo/1:24:30 and I think 96 joint stereo is also good enough for Audiobooks and Podcasts.
member frenzic has graciously offered to host the file. Thanks!
Other than processing to convert from WAV to FLAC, no other processing is employed. This is a roughly 3 minute excerpt from a 1960 Angel recording on the original red label. I have heard this LP on other systems in which the LP displayed ticks and pops.
Many ticks and pops are the result of oscillation on the part of the phono equalizer- I'm not saying that the ticks or pops is not there, just that they are normally inaudible if the phono preamp does not oscillate with each event. The other aspect is of course that the pickup be properly able to track the cartridge without issues.
Equipment used for playback: Atma-Sphere MP-1 preamp, Triplanar 9" arm, Transfiguation Orpheus cartridge.
Media damage like a scratch is another matter, but the solution there is simple- don't damage your recordings. I have many LPs that play with this sort of silence- some much better because as you can here, Angel didn't make the best of vinyl surfaces. In later years the process did improve- some of the vinyl surfaces being made today are considerably lower noise. Have fun!
Thanks again to frenzic for the space.
Last post by bennetng -
Eole skin is based on Panel Stack Splitter and JScript integrated into it.
Thank you Chamber32
All things being equal, more motor force means more efficiency. BUT...all things ain't equal.
To perhaps oversimplify for clarity, and talking about woofers, we have box size, low frequency extension, and efficiency. These three are interrelated. If you want good low frequency response from a small box you have, ipso facto, a low efficiency due to pesky and annoying laws of physics.
If you want good low frequency performance and high efficiency, they you will have a B-I-G box, again due to those pesky old laws of physics. Pay your money and take your pick.
Then I have no idea what you are asking because Monkey's is a lossless codec, Do you want a lossy codec that is more efficient than mp3 at low bitrates?
I srewed up. Meant lossy, because my files are too big.