Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: How not to interpret waveforms... (Read 7355 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

How not to interpret waveforms...

In the 'Vinyl' forum there's some discussion  right now about errors in deriving dynamic range from waveforms.  Here's an interesting case for that of a CD -- the Mobile Fidelity remaster of The Yes Album, which has been decried on some audiophile forums as being 'too compressed', based on misinterpretation of waveforms.  Rob LoVerde of MFSL sets the story straight:

https://www.facebook.com/photo.php?fbid=505...e=1&theater

How not to interpret waveforms...

Reply #1
Quote from there:
Quote
Ernie Longmire Yeah, waveform-peeking is a bit of a cult at this point. They're great for helping you figure out why you hear what you hear (e.g. the insanely-clipped peaks on my late-'80s copy of The Lexicon of Love, whose sound had annoyed me for years), but not so much for telling you what you WOULD hear if you actually listened to the damned thing.

IOW, there's more variables to sound quality than whether things were mastered with compression or not.

Here are the DR Database entries for the album in question. The 2010 version is the MFSL. I suppose it'll be better-sounding than the 1994 remaster of similar DR for advances in brickwall limiters alone. Whether it beats out the HDCD version or original CD issue would probably have to be assessed by ear.

The interesting question would be whether an uncompressed version of the MFSL master wouldn't sound even better - not sure what kind of brickwall limiters they would have been using at the time, but oversampling ones didn't really hit the mainstream until a year or two later. (Unfortunately I did not find a verbose DR Meter log which could have given clues in the form of peak amplitudes.) If they feel the need to get defensive about it (something that always makes me suspicious), chances are they knew that they were giving up some SQ but felt pressure to keep up with contemporary loudness levels. That being said, we're talking about a relatively moderate level of compression still, presumably nothing that should bother the average music listener too much.

How not to interpret waveforms...

Reply #2
The interesting question would be whether an uncompressed version of the MFSL master wouldn't sound even better
I thought Rob said there wasn't any compression applied, or I'm misreading your question.

There's a long thread on SHTV about this and I've only recently started reading it. To sum it up IMO, waveform junkies suddenly stopped liking the MFSL reissue after seeing the DR values. If we take the engineers word for it though, he did his best to properly transfer the tapes and mostly left the EQ as it was on the tapes.

Quote
Once consumers began seeing waveform graphics on the Internet, they either changed their previously positive opinions to negative or voiced their opinions based solely on the graphics, never having actually heard the product. I have never, nor will I ever be inclined to apply compression to a MoFi product. The practice of mastering music with full dynamic range intact has been an MFSL cornerstone since the company’s inception in 1977. We have maintained that practice, always. There would be no reason for me to stray from that practice now, and certainly not for a recording like “The Yes Album”.

How not to interpret waveforms...

Reply #3
The interesting question would be whether an uncompressed version of the MFSL master wouldn't sound even better - not sure what kind of brickwall limiters they would have been using at the time, but oversampling ones didn't really hit the


Stop.  You need to carefully re-read the article Rob wrote.

How not to interpret waveforms...

Reply #4
I looked up this album on the DR database.  One track got a DR6, which is supposed to be pretty dreadful based on all the religious zealotry I've been reading about what is and what is not acceptable.

How not to interpret waveforms...

Reply #5
I looked up this album on the DR database.  One track got a DR6, which is supposed to be pretty dreadful based on all the religious zealotry I've been reading about what is and what is not acceptable.


I have 3 albums listed there myself and the numbers ain't as pretty as I'd like them, no compressor/limiter was used either. I haven't looked into the way the TT DR is calculated...
But myself I prefer if the RMS (sinewave) to peak is 10dB to 30dB (called the crest factor?). If its' less than 10dB then I usually aren't far from listening fatigue eventually. If I make music and see less than 10dB I tend to think I messed something up somewhere, surely it can't be that "flat" can it?



How not to interpret waveforms...

Reply #7
If you are starting with the premise that since it's a vintage master it should be inherently super dynamic, your burden of proof has not been met. When has an engineer said he didnt patch in a stereo compressor to the two-buss before it went on the master tape?

How not to interpret waveforms...

Reply #8
When has an engineer said he didnt patch in a stereo compressor to the two-buss before it went on the master tape?
Well, most classical recording engineers never use compression. Nevertheless various DR values are found on classical recordings. The DR value is not synonymous to "artificial dynamic range manipulation by the engineer(s)". The composition and performance play a role too.

How not to interpret waveforms...

Reply #9
If you are starting with the premise that since it's a vintage master it should be inherently super dynamic, your burden of proof has not been met. When has an engineer said he didnt patch in a stereo compressor to the two-buss before it went on the master tape?



The claims and counterclaims here center around digital compression applied during remastering-- -not  (analog) compression used to create the original master tracks.

How not to interpret waveforms...

Reply #10
Stop.  You need to carefully re-read the article Rob wrote.

Dang. I hate it when stuff like that happens.

Looking up the actual waveforms of several editions (pic), there is no brickwalling apparent, though the louder sections do look a tad flattened out compared to the others. Assuming the transfer really was as good as they say, however, this may just be the (tape) compression applied way back in the day, with the resulting shape put back together correctly at the right levels etc. Incidentally, out of the editions compared, the Rhino one looks like it's better avoided. Almost looks like someone forgot to apply deemphasis or something. All things considered, I'd probably put the MFSL first, with the '94 remaster or older CD issues being the value for money picks.

How not to interpret waveforms...

Reply #11
All this analysis from simply looking at waveforms?

I hope not, because I don't know how much more irony I can handle.

How not to interpret waveforms...

Reply #12
The claims and counterclaims here center around digital compression applied during remastering-- -not  (analog) compression used to create the original master tracks.

My point is the former could make the later a moot point.