PDA

View Full Version : Is this lossy?


liveman
2005-11-29, 01:17 AM
Hello to all,
I have just downloaded a SHN set of the Beatles' A/B road session but I'm really not sure it is pure lossless. Though I don't hear any gap between tracks when playing in foobar, I suspect it's mp3 sourced. I've attached my spec/freq results... Thanks in advance!

AAR.oner
2005-11-29, 11:25 AM
liveman...can you zoom the SA to ~2 secs...its hard to tell when zoomed out to the whole track...

liveman
2005-11-29, 01:53 PM
Ok, here's a zoom of the spectral analysis:

AAR.oner
2005-11-29, 02:55 PM
thanks liveman...my impression is definitely lossy [don't know if its mp3 or some form of MD compression in the lineage?]...i'd let some of the others confirm this tho first

Five
2005-11-29, 05:34 PM
:nono: although it looks very much like lossy due to the absence of anything whatsoever above 14kHz, there are no blocks visible when zoomed in to 2 seconds.

there was much discussion about this last year at stg, especially vigotone's 30 Days release which was deemed lossless but heavily noise reduced. the purple chick set (by reputation) is a little less noise reduced, but still the nr is quite heavy.

another note about the early a/b road sets is that they were seeded in shn and since they are pretty much mono they compress down to 50% of the size if converted to flac. the conversion should be done using a secure method (see my signature).

I have three versions of these sessions (twickinham sessions box, 30 days, a/b road) and can pull them out and post screenshots from here if necessary. its quite a bulk of material (around 100 audio cdrs decompressed :eek: ). I plan to sort it all out and reseed/vine it somehow next year.

if you tell me the date/track of this a/b road session I can pull mine up and make sure that it looks the same as yours if you're still worried about it.

liveman
2005-11-29, 06:42 PM
Thank you AAR and Five for your replies I really appreciate it!

FIVE: The track I used for the screenshot above is from the January 23rd 1969 session, disc 1, track12. Files are SHN and presumably the same as:

--==<< Lossless Post Information >>==--
___________________________

Album Posted ============> The Beatles - AB-Road 1-23-69 Disc 1
Label & Catalog Number ==> Purple Chick
Number of Tracks ========> 27
___________________________

Name of Poster ==========> Hexer
Date Posted =============> 3/9/2004
Ripped Media ============> Original SHN files from Purple Chick
Program Used To Rip =====> Unknown
Number Of Discs Needed ==> 1
Posting Size ============> 332 Megs
Number of Files Posted ==> 29


Originally posted to a beatles newsgroup, the seeder told me he got the set from DIME some time ago...


Thanks again for your help!!! :)

liveman
2005-11-30, 02:57 AM
Ok, I've decoded the SHN to WAV and then encode it to MP3 (vbr extreme, setting 0). Then I've decoded the MP3 to WAV and here's the spectral view. I think there is a very small difference with the scan I've posted above but I can't interpret it.

MP3 spectral view:

range_hood
2005-11-30, 04:10 AM
I don't have a sample but playing around with colours and contrast of the negative reveals blocks. A black/white picture would be helpful. Looks mp3 to me.

liveman
2005-11-30, 02:24 PM
Hi range_hood, I've attached black & white spectral view from both files (1=shn>wav / 2=shn>wav>mp3>wav)

Five
2005-11-30, 02:49 PM
oh wow the plot thickens

both look lossy. you can't get those kind of block artifacts without lossy encoding. if anybody knows a way, show me how.

I'll break out my a/b road asap to see if it is the same as yours

could it be the entire a/b road is mp3-sourced?? :eek:

liveman
2005-11-30, 04:53 PM
Thank you again Five for your reply!
I'll stop downloading the set right now. By chance I've only downloaded the 23 January session and almost finished the 25th.

ssamadhi97
2005-12-02, 08:31 AM
Ok, I've decoded the SHN to WAV and then encode it to MP3 (vbr extreme, setting 0). Then I've decoded the MP3 to WAV and here's the spectral view. I think there is a very small difference with the scan I've posted above but I can't interpret it.
MP3 encoding always changes the spectrum. The result of your procedure pretty much doesn't tell you anything.