Absolutely correct. A 96Khz 24 bit FLAC of John Cage's 4:33 would be very small indeed. (I don't think however you can specify compression levels in FLAC - it's like ZIP, a lossless encoder)
The simple way to judge a FLAC is to look at the FLAC's properties where it's only too happy to tell you the sample rate and bit depth.
Last edited:
You can though, if you are so inclined, convert the unknown compression size flac to .wav, then convert it to flac with your chosen compression level 0 to 8
you can choose the compression level in foobar, 0 is the largest, uncompressed, just encoded within a flac container, so you can have track names embeded etc. or level 8 which is heavily compressed, yet still lossless.
you can choose the compression level in foobar, 0 is the largest, uncompressed, just encoded within a flac container, so you can have track names embeded etc. or level 8 which is heavily compressed, yet still lossless.
Last edited:
Here a 24/88 download from hdtracks ... from the SACD they tell
It's seem false 24/88 ... The part up ... noise ? antipiracy codes ?
It's seem false 24/88 ... The part up ... noise ? antipiracy codes ?
An externally hosted image should be here but it was not working when we last tested it.
Could that be a consequence of noise shaping applied during the decoding process? It certainly looks odd and there appears to be no usable output above 22kHz. SACD (DSD) conversion of a 44.1K/16 bit master, then DSD to 88/24?
My comment on open source is that the guys writing it generally are just as capable/competent and sometimes more so than the guys who write comparable closed source stuff. This is being typed on a machine that runs a quad core amd phenom processor and is running linux - an open source OS that also happens to be the basis for a lot of highly reliable things around you, not to mention the internet itself. (I also have a quad core i5 laptop running windows 7 and in a lot of respects I prefer linux - it's much faster and more stable.)
Here another very bizarre graph:
I download this album:
https://www.hdtracks.com/index.php?file=catalogdetail&valbum_code=HD827949002465
Here's the graph. WIth my soundcard Xonar it's play very very weird. I had to convert to 16/44 to play it correctly. With my audiophilleo 2, I can play correctly the 24/88.
I download this album:
https://www.hdtracks.com/index.php?file=catalogdetail&valbum_code=HD827949002465
Here's the graph. WIth my soundcard Xonar it's play very very weird. I had to convert to 16/44 to play it correctly. With my audiophilleo 2, I can play correctly the 24/88.
An externally hosted image should be here but it was not working when we last tested it.
I've got about 20 HDtracks downloads and only one of these is not real HD.
Looking at the two graphs that have been shown above, the first looks nothing like any HDtracks I've downloaded. I'm guessing the HF content is noise shaping. Not such a great thing imho. Most noise shaping moves noise to a much higher freq than that. Some systems will not like that noise there and may react badly to it.
The second one is mighty odd, for several reasons:
1) There's no music content above 12kHz - hardly HD material!! But this is just a piano recording so that might be kosher.
2) There seems to be an inverse of the music at the top of the freq range, some of which is only -20dB, which is as loud as some of the music! This is sure to mess with some systems.
I'd recommend both of these posters write to HDtracks with a copy of the graphs, and request a refund. If enough people do this then they might tighten up their QC a bit. How hard is it to check recordings before making them available??
Looking at the two graphs that have been shown above, the first looks nothing like any HDtracks I've downloaded. I'm guessing the HF content is noise shaping. Not such a great thing imho. Most noise shaping moves noise to a much higher freq than that. Some systems will not like that noise there and may react badly to it.
The second one is mighty odd, for several reasons:
1) There's no music content above 12kHz - hardly HD material!! But this is just a piano recording so that might be kosher.
2) There seems to be an inverse of the music at the top of the freq range, some of which is only -20dB, which is as loud as some of the music! This is sure to mess with some systems.
I'd recommend both of these posters write to HDtracks with a copy of the graphs, and request a refund. If enough people do this then they might tighten up their QC a bit. How hard is it to check recordings before making them available??
I agree with the assumption if both the test and test environment are such that give accurate measurements. I am not saying that this is not the case, but we must tread carefully. I am a business owner and in the modern times of of mass communication, someones misinformed opinion can spread like wildfire. I have no issue with confronting them with the issue and i have done so, but to go beyond that and move into accusations, we must have valid and repeatable proof.
Yes I am a business owner as well.
You can see that I've shown the same result with two different programs. That's all there is to the test environment - it's program-analyzes-file - simple as that. And the fourier transforms that are needed to calculate the frequencies are well known undergraduate math. (ie - It's not rocket science)
Please also note that I gave HDtracks a week to reply before I posted anything here. That should have been ample time to get a response.
I agree with your sentiments to some extent, but in the particular case I was highlighting, I don't see any question about the data. And anyone who chooses can easily reproduce the findings.
On the other hand, they have taken my money and delivered a sub-par product. They refuse (so far) to give me a refund. What other recourse do I have? And I do think I have been careful here - I feel that I understand the technical issues sufficiently to be able to explain in quite some detail what's going on in the tracks I bought (I won't make the same claim about the other tracks presented here - that would need more investigation and I don't have access to those tracks, and am loathe to give HDtracks more of my money until this issue is resolved.)
You can see that I've shown the same result with two different programs. That's all there is to the test environment - it's program-analyzes-file - simple as that. And the fourier transforms that are needed to calculate the frequencies are well known undergraduate math. (ie - It's not rocket science)
Please also note that I gave HDtracks a week to reply before I posted anything here. That should have been ample time to get a response.
I agree with your sentiments to some extent, but in the particular case I was highlighting, I don't see any question about the data. And anyone who chooses can easily reproduce the findings.
On the other hand, they have taken my money and delivered a sub-par product. They refuse (so far) to give me a refund. What other recourse do I have? And I do think I have been careful here - I feel that I understand the technical issues sufficiently to be able to explain in quite some detail what's going on in the tracks I bought (I won't make the same claim about the other tracks presented here - that would need more investigation and I don't have access to those tracks, and am loathe to give HDtracks more of my money until this issue is resolved.)
Last edited:
That one is really interesting. Is it sourced from analog tape?
If it is a digital recording, perhaps it is recorded 44.1khz PCM and then converted to 88.2khz without appropriate filtering? All the noise is sitting right at the 44.1khz sampling frequency. Something very strange is going on there.
Last edited:
Thats good news. For me, this was never about bashing HDTracks. I'm very critical of modern "remastering" techniques involving brickwalling, but that is a result of the record lables.
HDTracks has to be commended for being a pioneer of lossless audio downloads, and also what appears to be good customer service.
HDTracks has to be commended for being a pioneer of lossless audio downloads, and also what appears to be good customer service.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.