New Posts New Posts RSS Feed - Analysing hi-res files
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Welcome to the Graham Slee Audio Products Owners Forum

 

Open to all owners plus those contemplating the purchase of a Graham Slee HiFi System Components audio product and wishing to use this forum's loaner program: join here (Rules on posting can be found here)

This website along with trade marks Graham Slee and HiFi System Components are owned by Cadman Enterprises Ltd


Analysing hi-res files

 Post Reply Post Reply Page  123 6>
Author
morris_minor View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 27 Mar 2012
Location: Surrey
Status: Offline
Points: 6015
Post Options Post Options   Thanks (0) Thanks(0)   Quote morris_minor Quote  Post ReplyReply Direct Link To This Post Topic: Analysing hi-res files
    Posted: 01 Jun 2013 at 12:21pm
A short while ago I stumbled on this interesting article.

I'd long wondered if the high resolution files I'd downloaded (from sites like HDTracks) were really all they claimed. With the info from the article I downloaded the free Sonic Visualiser software (PC and Mac) and looked at two Spyro Gyra files.

The first (orange plot, below) was a known quantity since it was a rip I'd made myself of a LP at 96/24.

The second files (purple plot) is from an HDTracks download (at 96/24).

Unfortunately the horizontal axis in the software isn't labeled, but extends from 0 to half the sample rate. So for these two files it's 0 to 48Khz.

Unless I'm reading it all wrong (quite possible!) the HDTracks file has been upsampled to 96kHz because of the steep drop in the plot at around  20kHz, whereas my rip shows no such drop - rather a gentle decline in treble energy. [If I really am getting all this wrong, please can someone tell me . . Embarrassed].

Of course, knowing what I  now know about how digital recording actually works (thank you Graham!!) I won't be bothering any more with anything over redbook standards for playback. I'm sure I'll find other examples of upsampled files if I look, but I may not bother and just listen to the music instead . . . LOL

Maybe others here will find this free analysis program useful?




Bob

Majestic DAC/pre-amp
Accession MC/Enigma, Accession MM, Reflex M, Elevator EXP, Era Gold V
Solo ULDE, Novo, Lautus USB and digital, Libran balanced, CuSat50
2 x Proprius + Spatia/Spatia Links
Back to Top
Fatmangolf View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 23 Dec 2009
Location: Middlesbrough
Status: Offline
Points: 8960
Post Options Post Options   Thanks (0) Thanks(0)   Quote Fatmangolf Quote  Post ReplyReply Direct Link To This Post Posted: 01 Jun 2013 at 4:05pm
Thanks Bob, that's a good software link and I'm going to have a look at it later.

Hifi News highlighted this disturbing issue a couple of years ago and their download reviews include an energy spectrum analysis like yours to show which recordings really are 88k2, 96k, 192kHz. For example a 88k2 recording rather than an upsampled 44k1 that is labelled as 88k2, which your purple graph may be.

Simply put - why would we pay extra for a download company to upsample a CD when most of us already own DAC's that will do the same?
Jon

Open mind and ears whilst owning GSP Genera, Accession M, Accession MC, Elevator EXP, Solo ULDE, Proprius amps, Cusat50 cables, Lautus digital cable, Spatia cables and links, and a Majestic DAC.
Back to Top
Ash View Drop Down
Senior Member
Senior Member
Avatar

Joined: 18 Mar 2013
Location: Dorset
Status: Offline
Points: 4334
Post Options Post Options   Thanks (0) Thanks(0)   Quote Ash Quote  Post ReplyReply Direct Link To This Post Posted: 01 Jun 2013 at 4:58pm
Even more simply put - Why would we bother to buy any digital file "higher" than CD quality? It has no increased resolution compared to the CD. Graham and that Xiph guy have already cleared this up for us. No current digital releases address the dB stepping deficit so it seems that unless this happens, vinyl will always be King. Smile
We do not see things as they are. We see things as we are.
Back to Top
marshmid View Drop Down
Senior Member
Senior Member
Avatar

Joined: 10 Feb 2013
Location: Scotland
Status: Offline
Points: 136
Post Options Post Options   Thanks (0) Thanks(0)   Quote marshmid Quote  Post ReplyReply Direct Link To This Post Posted: 01 Jun 2013 at 9:57pm
Hi Bob,

That is a great piece of software. I have managed to reproduce your lower graph using a free download of an 88kHz sampled file. I see a gradual decline without any steep breaks.

However, the axes on the lower graph are unlabelled. Do you know what they are? Where did you see/deduce that the step you observe is at 20kHz?

I would like to try this on a number of purchased 'HD' tracks and free downloads - but I would like to know what I'm looking at!

Best regards

Marsh
Back to Top
morris_minor View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 27 Mar 2012
Location: Surrey
Status: Offline
Points: 6015
Post Options Post Options   Thanks (0) Thanks(0)   Quote morris_minor Quote  Post ReplyReply Direct Link To This Post Posted: 02 Jun 2013 at 11:39am
Originally posted by marshmid marshmid wrote:

Hi Bob,

That is a great piece of software. I have managed to reproduce your lower graph using a free download of an 88kHz sampled file. I see a gradual decline without any steep breaks.

However, the axes on the lower graph are unlabelled. Do you know what they are? Where did you see/deduce that the step you observe is at 20kHz?

I would like to try this on a number of purchased 'HD' tracks and free downloads - but I would like to know what I'm looking at!

Best regards

Marsh
Hi Marsh,

Did you mean my "upper graph" - the orange one - this is pukka 96/24 file  . . .

It's a shame the software doesn't label the axes. From that web page I linked to it seems that the horizontal scale goes from 0 to half the sample rate. So for 96kHz files it's 0 to 48KHz. Since the scale is a log scale, halfway along the scale would equate (for a 96kHz file) to 24 kHz. The sharp drop in my purple graph comes just before halfway along, so would equate to around 20-22kHz which is the limit for redbook files.

Jon mentioned Hi-Fi News doing these plots, and it was these that first got me interested in trying to find out what my download were - so finding this little program was really useful.
Bob

Majestic DAC/pre-amp
Accession MC/Enigma, Accession MM, Reflex M, Elevator EXP, Era Gold V
Solo ULDE, Novo, Lautus USB and digital, Libran balanced, CuSat50
2 x Proprius + Spatia/Spatia Links
Back to Top
Fatmangolf View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 23 Dec 2009
Location: Middlesbrough
Status: Offline
Points: 8960
Post Options Post Options   Thanks (0) Thanks(0)   Quote Fatmangolf Quote  Post ReplyReply Direct Link To This Post Posted: 02 Jun 2013 at 12:13pm
Hi Bob, you've got me thinking and I wanted to explain why I do bother with HD recordings.

I think the sample rate is really important as it determines the highest frequencies the digital recording will reproduce and what is done to those sounds. Real sounds like voices and instruments, acoustics and noise do go beyond our hearing range. A recording has capture these or reject them without adding artifacts within the sounds that are recorded.

After a quarter of a century there's a wealth of literature, evidence and opinion on CD's and the 44.1 kHz sample rate. Its 22,050 Hz "ceiling" on the sounds on CD's means low pass filters are needed to remove sounds above this. This will attenuate and alter the phase of the signals in the passband of the filter. In other words the sound that gets through will be altered a bit before it is recorded. That may be 16-22 kHz, and there's definitely musical content up there!

The choice of filter type would make a difference to the ADC process and was one of the settings included in the HDCD standard. If you have a CD player or DAC that lets you choose between different filters e.g. an Audiolab M-DAC, you are exploring a similar effect on playback.

Now imagine the cut off filter in the recording at 88k2Hz. It is at 44.1 kHz, or twice the CD level. That would mean that the sounds from 16kHz - 22kHz aren't in the filter pass band anymore. Potentially it is another octave of ultrasonic sound.

That's why I think a good recording at a high sample rate like 88k2, 96k or 192kHz would be better than the CD equivalent.
Jon

Open mind and ears whilst owning GSP Genera, Accession M, Accession MC, Elevator EXP, Solo ULDE, Proprius amps, Cusat50 cables, Lautus digital cable, Spatia cables and links, and a Majestic DAC.
Back to Top
morris_minor View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 27 Mar 2012
Location: Surrey
Status: Offline
Points: 6015
Post Options Post Options   Thanks (0) Thanks(0)   Quote morris_minor Quote  Post ReplyReply Direct Link To This Post Posted: 02 Jun 2013 at 2:25pm
I do agree with you Jon, about higher frequencies. Harmonics, if not fundamentals, exist way above the audible threshold. It's the process how we actually perceive them that seems a sticking point. 

Seems to me we're safer sticking to vinyl! Thumbs UpHeart
Bob

Majestic DAC/pre-amp
Accession MC/Enigma, Accession MM, Reflex M, Elevator EXP, Era Gold V
Solo ULDE, Novo, Lautus USB and digital, Libran balanced, CuSat50
2 x Proprius + Spatia/Spatia Links
Back to Top
 Post Reply Post Reply Page  123 6>
  Share Topic   

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.01
Copyright ©2001-2018 Web Wiz Ltd.

This page was generated in 0.125 seconds.