Jump to content

Compressed Codecs to MD seem to Code AGAIN therefore even lossier

Rate this topic


1kyle

Recommended Posts

I've done this enough times now to definitely say that SS (even 3.4) codes AGAIN any compressed codec going to your MD device so you lose more fidelity than if you start with a WAV file.

What I mean here is that if say you rip a CD's to ATRAC+ @352 and then transfer this to your MD vis SS it sounds not as good than if you rip to WAV first and then transfer to MD via SS at 352 kbs.

Now to me there SHOULDN'T in theory be any extra encoding if going from 352 in the library to 352 on the MD disk but I've done this enough times to say there definitely seems to be a difference in the sound (and I can tell just with headphones).

I've also tested whether using an older 74 / 80 MD in place of a 1GB disc is possibly the culprit of this but the discs don't make any difference as far as I can tell.

So although I can't say whether there is an extra encoding step or not when starting with a compressed file it definitely sounds discernably better using a WAV file and transferring that to the MD unit at the compression rate you want.

I'd suggest that if you haven't tried starting with a decent uncompressed WAV file from a good quality recorded CD you try it and you'll probably be amazed at the 352 kbs quality.

Intuitively one would have thought that a 352 kbs file on the computer being transferred to a 352 kbs file on the MD wouldn't need any more transcoding but I'm obviously wrong.

Unless you are starting with a WAV file on your computer this could be a reason why many of you can't tell the difference between Hi-SP at 256 done with Simple Burner and 352 kbs from a 352 kbs file in your library.

This really surprised me but I've done the test enough times on all sorts of music and with both standard 74 / 80 discs and 1GB discs,

I've even used different recorders NH1 and RH10 - same result.

(If trying this do start with an original uncompressed WAV file, don't convert a Lossy compressed file to WAV first as you'll still get "Double" or multiple Transcoding effects).

Cheers

-K

Edited by 1kyle
Link to comment
Share on other sites

Although I wasn't sure for sure...I had suspected as much.

This is the reason I have my entire music collection on an external hard drive in wav format.

I rip all my cds to wav using 'Exact Audio Copy' and then import the folders into Sonic Stage....works really well.

I am sure audio codecs will come and go in the future...having all my tunes in wav lets me move with the times too, should I desire...whilst my 'starting point' will always be the same (wav)

Plus when I run my tunes through my main stereo, I certainly didn't want to be listening to compressed audio of any kind....saving in wav was really my only option here (I didn't want to be tied down to atrac lossless or any other lossless for that matter....wav will always be...and if you have the space, why not?)

Cheers -_-

Link to comment
Share on other sites

Hmm.

Well, the way it's supposed to work is like this:

* if you have "as is" downloading on, things should only be re-encoded if they are originally [in your library] in a format that can't be played back by your device

* if you have manual bitrate/forced mode turned on, things will always be re-encoded to your chosen bitrate unless they already are that bitrate

That said, you can easily see the difference between when things are being transcoded during download or not; two progress bars go by for each track that does so, and it takes extra time to transcode. When no transcoding is taking place, things transfer directly asis and go pretty quickly.

I have not noticed a significant difference between things that are already encoded to a specific a3+ bitrate in my library and the same tracks once downloaded. There are differences during playback that are caused by the headphone amp, the unit's EQ settings, your headphones, &c. but otherwise the same 1st-generation encoding done by SS should be exactly the data on your disc, so there should be no further artifacting caused by encoding generations.

I *have* noticed that the digital amp on the RH10 tends to make artifacting slightly more obvious than the analogue amp on the NH700. This might be because the RH10 appears to slightly "sweeten" the sound by positively high-shelving everything by a couple of dB [i have actually measured this]. It may seem odd, but for playback I actually prefer the way my NH700 sounds over the RH10.

One way to rather unscientifically test this [this ain't no ABX] would be to play back the tracks in question through SS via USB .. compare the track from the library with the track on HiMD using SS, using the same playback device [your computer, sound card, and speakers or headphones].

You could also actually ABX the digital track by converting your library track to WAV and then uploading and doing the same with the track after downloading it to your player.

Another possibility would be to compare the WAV converted track from the library with an analogue-copied version from your player [which would help determine whether it's the output of the player that causes the difference].

In summary: I doubt there's another generation of encoding taking place [everything I've seen argues very firmly against this]; it's more likely that the output of the player sounds sufficiently different from that of your computer that you notice the difference.

Link to comment
Share on other sites

Hi there I wasn't using the computer's gear for testing. I rarely listen using a computer.

I can definitely say also when starting with WAV files in the library transferring to the MD device @ 352 is QUICKER than transferring 352kbs files in the library to 352 kbs files in the device.

So it would appear the software IS doing something but what I haven't a clue.

If there really WASN'T any extra processing the file should be transferred at around the device write speed (USB 1.1). There is some extra encryption (data encryption not transcoding) but the processing required for that is minimal and is part of the MD system and would be the same whether we are starting from our uncompressed WAV file or the ATRAC3 352 kbs compressed file.

As Hard disks are cheap enough now it makes sense to only store music in LOSSLESS formats. You can store it in a compressed format provided the compression is LOSSLESS like FLAC or whatever. Converting FLAC (Lossless) to WAV (Lossless) should restore the original WAV file to be 100% a clone of the original.

Cheers

-K

Link to comment
Share on other sites

I can definitely say also when starting with WAV files in the library transferring to the MD device @ 352 is QUICKER than transferring 352kbs files in the library to 352 kbs files in the device.

Something sounds profoundly wrong there, 1kyle.

On my machine, and other every machine I've used SS on, if you try to download a pre-encoded 352kbps track, it goes quickly. Single progress bar, no mention of conversion, it simply downloads.

If you try to transfer a WAV file in as-is mode then it copies it as PCM [which takes significantly longer since it's around 5x the amount of data]; in forced-bitrate mode, it converts [with a progress bar saying it's doing so], then downloads [with a second progress bar].

The amount of time for on-the-fly conversion isn't that long but it's noticeable that it's doing it, since it both indicates that it's doing so and takes longer.

Sounds to me like you've either got something set oddly or .. I don't know. What you're experiencing is definitely NOT what I'm experiencing, and what I've been experiencing since SS 2.1 [minor interface changes to progress bars and the implementation of as-is mode notwithstanding].

Something is definitely fishy there.

Link to comment
Share on other sites

i wonder if there's much loss when transcoding mp3s of less bitrate to 352kbps atrac. i mean is it worth doing that due to nonlinear db/freq ratio seen in himd with native mp3 playback ? yeah after all those codec holy wars mostly tell about low bitrates, and there's not much info on crosscoding mutual loss.

say, i have a 256kbps mp3, is it better to code it into atrac or just have an equalizer preset devoted to mp3 playback ? :/

Link to comment
Share on other sites

i wonder if there's much loss when transcoding mp3s of less bitrate to 352kbps atrac. i mean is it worth doing that due to nonlinear db/freq ratio seen in himd with native mp3 playback ? yeah after all those codec holy wars mostly tell about low bitrates, and there's not much info on crosscoding mutual loss.

say, i have a 256kbps mp3, is it better to code it into atrac or just have an equalizer preset devoted to mp3 playback ? :/

Transcoding an MP3 to ATRAC3plus @ 256kbps sound pretty good to me. I transcode all my MP3s to ATRAC3plus 256 because it preserves more of the sound quality and the higs are restored. I only do this when transferring to my RH10. Try it!

Link to comment
Share on other sites

Transcoding an MP3 to ATRAC3plus @ 256kbps sound pretty good to me. I transcode all my MP3s to ATRAC3plus 256 because it preserves more of the sound quality and the higs are restored. I only do this when transferring to my RH10. Try it!

What do you mean with: "the highs are restored"? MP3 is a lossy format, so if you convert a wav file to mp3 you can't recover some quality converting it to wav again or to any other superior format (than mp3). The mp3 convertion process doesn't "zip" the file, it modifies it so you lose quality with the convertion.

Link to comment
Share on other sites

What do you mean with: "the highs are restored"? MP3 is a lossy format, so if you convert a wav file to mp3 you can't recover some quality converting it to wav again or to any other superior format (than mp3). The mp3 convertion process doesn't "zip" the file, it modifies it so you lose quality with the convertion.

MP3 playback on Hi-MD units lacks "highs" or treble response. Converting them to ATRAC3plus @ 256 restores the treble response. To me, it sounds very good.

Link to comment
Share on other sites

ohh that! Now I understand. Totally right. But converting the mp3 to atrac actually doesn't "restore" the highs, they were always there in the original mp3 . You can say that when you transfer a mp3 to himd without convertion it "lacks" highs.

Edited by sebastianbf
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...