AVS Forum banner

1 - 20 of 26 Posts

·
Registered
Joined
·
114 Posts
Discussion Starter #1
System:

Raspberry PI-3 utilizing Moode Image Release: 3.7, 2017-05-25
HIFIBerry Dac Pro -excellent BurrBrown based DAC
RCA out to Berhinger Ultragraph Pro XLR out
To Crown XLS 1002 RCA out to
20 year old Mission Argonauts Towers
small room 12X15X8 carpet no other sound properties

CD's Ripped to MP3 24 bit, 44.1 kHz, Stereo Rate: 320 kbps using Exact Copy software


I have been using the Berhinger to get the room correction and to make it sound right for me

Ok Back to the question asked

The latest version of the Moode software allows you to upsample based on my HIFIBerry DAC as seen below

Input Processing
Source: NAS/MusicDrive/Enya/Amarantine/01 Less Than a Pearl.mp3
Encoded at: Unknown
Decoded to: 24 bit, 44.1 kHz, Stereo
Rate: 320 kbps
DSP Operations
Resampling: 32 bit, 192 kHz, Stereo (SoX very high quality)
Crossfeed: off
Equalizer: off
Chip options: Low latency IIR with de-emphasis, aVol=0 dB, aPbb=0 dB
Volume: Software (MPD 32-bit float with dither)
Output Stream
Format: 32 bit, 192 kHz, Stereo
Rate: 12.288 mbps
Audio Device
Device: HiFiBerry DAC+ Pro
Chip: Burr Brown PCM5122
Architecture: Sigma/Delta Advanced Segment 24 bit 192 kHz
Interface: I2S

With this setting I have flat lined the Berhinger ( no +- Adjustments )

I like this better than before, soundstage slightly better ( greater depth ) as well as what I would describe there just seems to be more of the music to listen to.

Everything I read says if the source material does not change then more sampling or even upsampling should not make any difference

Has anyone else played with upsampling and if so did you like results

Thanks
 

·
Registered
Joined
·
941 Posts
You can't get a Rolls Royce from a Ford


Sent from my iPhone using Tapatalk
 

·
Registered
Joined
·
1,518 Posts
No I don't unless I have to process it to get it to play.
 

·
Registered
Joined
·
2,589 Posts
Why copy cd's to mp3? It doesn't make sense to use a lossy format and then upsample through your media player/dac. This upsampling does not magically restore the quality lost from the conversion to mp3.


I standardize everything to 2496 as this is the sample rate the dsp engine is working with and it is the default sample rate of the many dvda's ripped to the server.
I copied everything (CD, DVDA multichannel) to 24bit 96kHz flac files. Ripping cd with foobar using the sox upsampler plugin twice in a row: first upsample to 24bit/176.4 and then downsample to 2496.
This prevents a couple of sample rate conversions in the digital path that would have occurred anyway otherwise.


Upsampling cd's doesn't improve the audio quality.
 

·
Registered
Joined
·
1,231 Posts
IMO, you should do as best as you can to not re-sample / upsample / downsample...

You should get a DAC that supports automatic sample rate switching and also avoid using inbuilt computer audio subsystem... (Windows or Linux)

I have two RPis both installed with hifiberry digi+ connected via coax to external non-usb DACs.

As has been said already, it doesn't sound correct to first compress your audio to MP3 and then upsample playback in a DAC... I'm almost certain the DAC you're using might simply be performing a "loudness" function which you imagine sounds better than the MP3 you're playing
 

·
Registered
Joined
·
113 Posts
Well...I use Jriver to rip all my CD's, and convert/upsample to DSD128 and send to my DSD DAC. Seems like you guys frown on said activity.
 

·
Registered
Joined
·
114 Posts
Discussion Starter #9
Totally understand that mp3 format was suboptimal but at the time it was used, and I found it interesting that having the DAC upsample seemed to help listening enjoyment ( since it could not increase quality)

I may now go back and rerip all my cd's to the higher rate and different file format.

Question is - If the same CD source is used and then rip at 24/96 or higher will the net sound effect be similar or better. Basically the higher rates Only Allow for more data which may or may not be used depending on the source material it self

Bottom line I do enjoy my system as is and it still puts a smile on my face when I listen to my music so reripping is not a big priority, but probably a future project to do.

As to the comment that the upsampling was just playing louder, I really do not believe that was the case, but always possible as I listen to music way below anything close to reference levels ie seldom past 50% volume

Thanks for all your comments
 

·
Registered
Joined
·
941 Posts
Zero point ripping a CD to 24/96


Sent from my iPad using Tapatalk
 
  • Like
Reactions: rnsound and jfbar1

·
Registered
Joined
·
212 Posts
I do not, I was told by many and seen on other forums not to up sample. I have done some tests and found that listening to 44khz in 44khz and 96khz in 96khz etcetera, to sound best. That's my take anyways. If up sampling sounds better to your ears, by all means do it.
 

·
Registered
Joined
·
1,231 Posts
Totally understand that mp3 format was suboptimal but at the time it was used, and I found it interesting that having the DAC upsample seemed to help listening enjoyment ( since it could not increase quality)

I may now go back and rerip all my cd's to the higher rate and different file format.

Question is - If the same CD source is used and then rip at 24/96 or higher will the net sound effect be similar or better. Basically the higher rates Only Allow for more data which may or may not be used depending on the source material it self

Bottom line I do enjoy my system as is and it still puts a smile on my face when I listen to my music so reripping is not a big priority, but probably a future project to do.

As to the comment that the upsampling was just playing louder, I really do not believe that was the case, but always possible as I listen to music way below anything close to reference levels ie seldom past 50% volume

Thanks for all your comments
Ripping a CD to 24/96 or higher will only result in usage of more disc space... on a good system the sound should be similar, and can not be better since all that is being done is representing 16/44.1 data in a 24/96 space...

In other words, audio information that is not available in the source material cannot be made up from increasing sampling rate... otherwise, something is wrong or not correctly set in the ripping software...
 

·
Registered
Joined
·
179 Posts
Question is - If the same CD source is used and then rip at 24/96 or higher will the net sound effect be similar or better. Basically the higher rates Only Allow for more data which may or may not be used depending on the source material it self
The fidelity will remain the same. All you'd be doing is putting the same content into a larger bit bucket.
 

·
Registered
Joined
·
113 Posts
Ripping can't add information that's not there...but when ripping and up sampling a higher sample rate, you're actually padding over what's already there. So some of the more subtle nuances may become more audible. This is why you seem to hear more detail. There are many factors in hearing a difference. Different software offer different filtering....some better than others....some filters are good, some aren't. It also depends on the DAC..in my case, I use two DSD DAC's....one by iFI and the other by Marantz. I can hear the difference in 44.1 and 5.6 (128)
Digitized CD's samples are 44.1 thousand times per second, producing 16-bit PCM samples. DSD uses pulse density modulation (PDM) to encode 2,822,400 1-bit samples per second which is 64 times the CD sampling rate. Yes it uses more space....but to my ears its worth it.
 

·
Registered
Joined
·
339 Posts
I use dBpoweramp to rip to flac, which is lossless. I don't think padding with made up detail is productive or helpful. Sometimes I rip to mp3 format just so I can send a song to a friend. They are also lossless MP3 files.

I advocate that you archive your music in some lossless format so that you always can have the original in a perfect and complete way.

If next week a new file format becomes the fashion you can always bulk convert from your master archive.

Using your method at some point you will be adding detail that does not belong, artifacts, and you will have screwed your self.
 

·
Registered
Joined
·
113 Posts
Who said detail was made up? The detail is already there. Some of the small nuances aren't audible as much. The padding makes it audible. I don't do downloads....to me they are to expensive for what you get. I buy CD's from ebay and rip and store the CD. I always have them. How can you add detail that's not on the CD? I've been doing it like this from 2015 and have had no issues.....music is lovely, much better than MP3 files which you can get from Amazon down loads. When not listening to vinyl, its DSD 128. Have you ever down loaded a DSD 128 file from Acoustic sounds?
 

·
Registered
Joined
·
339 Posts
We have much in common, neither one of us downloads mp3 files. We both rip music from CDs we purchase.

To make an analogy, my TV upscale from lower resolution to 4k. It does it well and I never had any problem with artifacts of detail "made up". In general I am satisfied with the technology.

If in the same way you are satisfied with up scaling your music, in your own home I'm not the music police.

There are those who argue that normal well recorded CDs are all high quality speakers can accurately reproduce. They argue that HD CDs and SACD provide no acoustic real benefit.

When one up scales video it is a simple mathematical process when, for example, the information in one pixel becomes represented by four pixels or eight pixels. The granularity in the image becomes larger, color and contrast and detail should remain unchanged. No artifacts should ever be introduced.

I watch 480p on a 3840 screen almost every day and again in general I'm happy with the results. I would make the case that that video up scaling well is simpler than audio up scaling with the same quality expectations. Either the process changes nothing and it is padding, or it does some interpolation and it adds detail.

I would argue that neither is helpful or desirable. I am a Pittsburgh Steeler fan, I have Steeler videos in broadcast quality going back to the 1970's. I never watch them any more. I find the quality insufferable. One can't up scale sports videos well and the missing detail becomes critical. I will argue that no method exists to take those low quality Steeler video and add back the missing detail and do a good job. In that critical application I am not happy with up scaling.

Let me ask you this. Say some one gave you a poor quality mp3 file, is there any process or method to restore that file to lossless quality original? If there is then the same methods would work to take any CD quality recording and make it SACD quality; and I don't think that is possible to do well.
 

·
Registered
Joined
·
113 Posts
Tom....I'm a simple laymen.... I've never ever tried to up sample a mp3 file in my life. All I do is get a CD, (not a mp3 file)......rip it...have the software up sample/convert to DSD 128 and send that to my DSD DAC. I enjoy it.... Sounds quality is better than the CD. The proof is in the hearing. Audiophiles take CD's, rip them and up sample to SACD quality....its called ripping and up sampling to DSD 64. If you can't hear it, then there are factors to why you can't hear it. If what you are saying is true...then making and selling DSD DAC's and all that comes with that is one of the biggest audio frauds in history.
 

·
Registered
Joined
·
8,697 Posts
In any serious 2ch discussion, never mention/use any lossy codec (MP3/aac/etc) as source material, only lossless wav/aiff/flac, please ;)

That said, in the past I might have said "never" up/down convert any digital audio nor apply any processing- just play natively as-is, bit perfect, still "sound" advice.

However, when talking CD/redbook 44.1Khz, the idea would be to upsample to 88.2/96 or higher in order to avoid the "brick wall" filter applied to 44.1Khz audio at playback, moving the filter well beyond 20khz and allowing a more gentle slope.
 
1 - 20 of 26 Posts
Top