[iamivancb]

Well, gregorio said some recording have very little stuff above 7 kHz. Doesn't mean even these some recordings STOP at 7 kHz. The stuff above 10 kHz is not as important as people think. 10-20 kHz is just one octave, the highest octave humans can hear (when young). People over 30 years old shouldn't worry about frequencies over 17 kHz AT ALL. It is the 21st century, but human hearing isn't different from what it was in the past. Some things just don't matter. People just assume they do.


At 44.1 kHz spectrums aren't supposed to reach 22 kHz because anti-alias filters aren't infinitely sharp. Spectrums reach, if there is stuff up there, to 20-21 kHz depending on the anti-alias filter and that's enough.

Sorry for the immense delay, just have been busier than usual these two weeks and I want to reply well enough. Okay, so I misunderstood what gregorio stated about 7kHz and took it literally.

I don't want to seem dumb, yet what is an anti-alias filter?

 

[iamivancb]

I’m not sure what the 21st century has to do with it, we’ve been able to create recordings/masters cleanly up to 20kHz from the early 1980’s and had mic’s that captured cleanly up to 14kHz from the early 1950’s, although tape wasn’t ideal up there. And, I did not state “just up to 7kHz”, I said “relatively little above 7kHz”.

Look at the colour scale on the right Y axis! The red colour is where vast majority of the energy is, which is below 2kHz on all three of the examples. Yellow is -40dB to -50dB, 100 to 300 times lower in level. Green is -60dB to -70dB, which is 1,000 to 3,000 times lower in level and about the noise floor of many studio mics, while the blue colour represents around -100dB, 100,000 times below peak, around (or below) the theoretical limit of 16bit and is purely noise/distortion (from the mics, mic pre-amps or processing).

The spectrograms are not unusual for certain types of recording. Remember that the highest note of the highest pitched instruments even in an orchestra is around 4kHz, above that it’s just harmonics, noise or distortion. The only really significant amounts of energy we commonly find above 7kHz are from struck metallic percussion instruments but even then it depends on what mics were used, where they were placed and the processing applied.

You cannot remove the dither in mastering as the application of dither must always be the very last step of mastering and why would anyone even consider trying? There’s an almost infinite amount to consider within the audible range, so why would anyone limit their consideration within the audible range and instead waste it on things that cannot be audible? The only reason I can think of, would be some bedroom producers/engineers who don’t know what they’re doing and are going on some audiophile myths they’ve read/seen somewhere.

That’s always been the case with unsigned/underground bands/musicians. It’s true to a significant degree even of signed artists today! From the 1980’s onwards consumers became used to the extremely high quality distribution medium of CD and all the top artists spending many months (sometimes over a year) creating recordings in the top studios with the best engineers and equipment, which became common even from the late 1960’s. Consumers don’t want that anymore, what they want is recordings that cost a few cents a track or nothing at all. With such low revenues, there’s no financial justification for spending large amounts of money hiring the best studios/engineers for many months on end. In fact many commercial studios have gone out of business and albums from the top artists typically use those facilities for weeks rather than many months.

Additionally, what is “good” vs “dubious” quality anyway? Thrash metal and similar sub-genres have their own rules, huge amounts of noise and distortion that would be horrifically bad in other genres are highly desirable in heavy/thrash genres, it’s supposed to sound “dirty”/“nasty”. Of course it’s entirely possible to sound dirty/nasty and be sophisticated/polished at the same time but that takes not only skill but knowledge/experience and the right equipment/facilities, which very few bands can afford even for a short time. The mixing on the example track is not sophisticated, it’s agricultural and indicative of somewhat inexperienced/amateurish mixing and production.

G

I apologize if my previous response was not aligned with your perspective. It seems that you have a stronger understanding of audio recording and production, particularly in relation to the frequency range, spectrograms, dithering, and the challenges faced by unsigned/underground artists.

You are correct that advancements in recording technology have allowed for cleaner recordings and wider frequency ranges since the early 1980s. The examples you provided regarding the distribution of energy in the spectrograms are also valid, with the majority of energy often concentrated in the lower frequencies. Different types of recordings and instruments have varying frequency characteristics, and the presence of energy above 7kHz depends on multiple factors such as microphone choice, placement, and processing.

Regarding dithering, I may have misunderstood your statement. Dithering is indeed applied as the final step in the mastering process to mitigate quantization errors when reducing bit depth, and it should not be removed afterwards.

Your point about the changing landscape of the music industry and the shift in consumer preferences toward inexpensive or free music is well-taken. The financial constraints faced by artists and the reduced investment in high-end studios and engineering do impact the overall quality and production of music, even for signed artists.

Lastly, the concept of "good" and "dubious" quality is subjective and genre-dependent. Certain genres, such as thrash metal, embrace noise and distortion as intentional artistic choices. Achieving a dirty or nasty sound while still maintaining sophistication and polish requires skill, knowledge, experience, and suitable equipment, which may not be accessible to all bands.

 

[iamivancb]

I know in metal some bands seek out certain producers as they're well regarded in the music industry.



I've always found thrash to be generally well produced, tight, fast and clear. Other metal genres certainly have lo fi production to add to the atmosphere but not thrash.

You're absolutely right. While some metal genres intentionally embrace lo-fi or raw production to enhance their atmosphere and aesthetic, thrash metal has traditionally been associated with a more polished and clear sound. Thrash metal's focus on speed, precision, and tightness often calls for a production style that emphasizes clarity and the ability to hear each instrument distinctly. As a result, many thrash metal bands have sought out producers who are experienced in capturing the intensity and intricacies of the genre.

Producers who have worked extensively in the metal genre, especially with thrash metal bands, have developed a keen understanding of the specific production techniques and sonic characteristics that suit the genre's needs. They know how to balance the aggression, energy, and technicality of thrash metal while still maintaining a clear and well-defined sound. Their expertise in capturing fast-paced performances, intricate guitar work, and rapid drumming helps in achieving the desired sound for thrash metal recordings.

Of course, there can still be variation in production styles and preferences within the thrash metal genre, but overall, the goal is to create a tight, fast, and clear sound that highlights the musicality and intensity of the music.

 

[iamivancb]

No, the vast majority of professional artists of any genre want well produced music. It’s not that the artists don’t want it (with the exception of some deliberately lo-fi/underground bands as you mentioned), it’s that the consumers don’t and therefore the artists (and labels) don’t have or won’t commit the financial resources to achieve it.

I’ve heard a lot of poorly produced thrash, quite a lot that’s competently produced, only a small amount that’s well produced and I don’t recall any exceptionally well produced in recent years. However, I’m not a thrash aficionado and don’t listen to it much compared to other genres.

G

You make a valid point about the financial constraints and consumer preferences shaping the production quality of music. The decisions made by artists and labels regarding production quality often take into account factors such as budget limitations and the target audience's preferences. In some cases, this may result in compromises in terms of production values.

Regarding the production quality of thrash metal specifically, it's important to note that there can be variations within the genre. While there are certainly well-produced thrash metal albums, there may also be instances where production quality falls short. Production quality can depend on various factors such as the resources available to the artists, the expertise of the production team, and the artistic choices made during the recording and mixing process.

As with any genre, the perception of what constitutes "good" or "well-produced" music is subjective and can vary among individuals. While you may have heard a range of production quality within thrash metal, it's important to recognize that opinions on what constitutes good production can differ.

Ultimately, the production quality of music is a complex and multifaceted aspect that can be influenced by various factors, including artistic intent, resources, and consumer preferences.

 

[iamivancb]

And yet there's been a resurgence in vinyl sales so many consumers are very much into their music and gear, be interesting to see the stats on what genres.

You're correct that vinyl sales have experienced a resurgence in recent years. Vinyl records offer a tactile and nostalgic experience that appeals to many music enthusiasts, and some consumers value the unique sound characteristics of vinyl. This revival has been driven by a combination of factors, including the desire for a physical music collection, the appreciation for album artwork, and the perception of vinyl as a high-fidelity format.

When it comes to the genres that dominate vinyl sales, it can vary depending on geographic location and individual preferences. However, it's worth noting that rock, alternative, and indie genres have generally been popular among vinyl enthusiasts. These genres often have dedicated fan bases that value the vinyl format and are willing to invest in the necessary equipment to enjoy it.

It would be interesting to see more detailed statistics on vinyl sales broken down by genre, but such data can be challenging to gather comprehensively. Market research firms, record labels, and music industry associations may have access to more specific insights on the subject.

Overall, the resurgence in vinyl sales indicates that there is a significant portion of music consumers who value the quality and experience that vinyl records offer.

 

[iamivancb]

A resurgence compared to what? Vinyl sold about 41m units and accounted for $1.2b revenue in 2021. In comparison, vinyl sales were <1m per year in the 1990’s so that is a big resurgence but on the other hand 140m records were sold in 1921 and just under 350m vinyl sales at the end of 1970’s, so not so much of a resurgence.

However, it’s not specifically about vinyl sales, artists and record labels invest in making new recordings based on the revenue/revenue potential they will earn and that’s where the picture is far less rosy. Total revenue for recording sales in 2021 in the US was $14.9b, in 1999 revenues just from CD sales was $21b and $23b in total (Statista). That’s a big drop but it’s even worse than the figures suggest because only $1.7b of that $14.9b was from physical media (vinyl and CD), the vast majority was from streaming. This is a problem because a large slice of that revenue goes to the streaming services rather than the artists/record labels and unlike the record labels, the streaming services do not invest any of their slice of the revenue in making new recordings. Worse still, according to MBW only 25.5% of that US revenue in 2021 was from new recordings, back catalogue recordings accounted for 74.5%. This is a trend going back several years, back in the 1990’s “hay day”, the vast majority of the revenue was from new recordings (although I cannot find exact figures).

Putting all the above together; the financial incentive for artists/labels to invest in making new recordings is just a fraction of what it was prior to the 2000’s, at a rough estimate no more than a quarter and possibly only a tenth and, it’s falling, not resurgent!

G

You've provided accurate information regarding the sales figures and revenue of vinyl records in recent years, as well as the overall decline in revenue from physical media sales and the rise of streaming platforms. I appreciate you highlighting these points.

Indeed, while vinyl sales have experienced a resurgence in comparison to the low point of the 1990s, they still represent a small portion of the overall music market when compared to the peak sales of vinyl in the past. The shift to digital consumption and streaming has significantly impacted the revenue landscape of the music industry, with streaming platforms dominating the market and generating a substantial portion of the revenue.

Your observations about the financial challenges faced by artists and record labels in investing in new recordings are alright. With a significant portion of revenue coming from streaming, which is primarily controlled by streaming services, the distribution of revenue has become a concern. Artists and labels often receive a smaller share of the revenue compared to the streaming platforms themselves.

Furthermore, the shift in revenue distribution towards back catalogue recordings rather than new recordings can also impact the financial incentives for artists and labels to invest in creating new music. This trend, as you mentioned, has been ongoing for several years.

The changing landscape of the music industry and its impact on artists, labels, and the creation of new recordings is indeed a complex issue.

 

[iamivancb]

I buy music from Bandcamp which gives most of the money to the artist but read recently it's been bought out.
With regard to streaming services many prefer to listen to tracks from different artists rather than listen through a complete album, this way of listening must affect how much artists receive as 'algorithm's are skewed in not rewarding artists so much for use of individual tracks Vs an album?

You're correct that platforms like Bandcamp are often preferred by music enthusiasts because they offer a more direct and artist-friendly approach to purchasing music. Bandcamp has gained popularity for its model of providing a larger portion of the revenue directly to artists compared to other platforms.

Regarding recent news of Bandcamp being bought out, to my knowledge, Bandcamp remained an independent company. However, it's always important to stay updated with the latest news and developments in the music industry.

In terms of streaming services, the way listeners consume music has indeed shifted towards individual tracks rather than complete albums. This shift in listening behaviour can impact how artists receive compensation. Streaming platforms typically pay artists based on the number of streams their songs receive, which means that if listeners are more inclined to stream individual tracks rather than full albums, the revenue distribution can be affected.

Furthermore, the algorithms used by streaming services to recommend music can play a role in shaping listener behaviour and impacting artists' revenue. These algorithms tend to prioritize popular tracks or playlists, potentially overlooking or under-rewarding artists with a more niche or less mainstream appeal.

However, it's worth noting that there are ongoing discussions and efforts within the music industry to address these concerns and improve the revenue distribution model for artists. Some streaming platforms have implemented initiatives to better compensate artists, such as increasing royalty rates or offering direct artist support programs.

It's important for music listeners to be aware of these issues and consider supporting artists through platforms like Bandcamp or other means that provide more direct support. Additionally, engaging with complete albums and exploring artists beyond popular tracks can contribute to a more equitable and diverse music ecosystem.

 

[sander99]

I don't want to seem dumb, yet what is an anti-alias filter?

A requirement for digital audio to work correctly is that before an analog signal is sampled it may only contain frequencies below half the sampling rate.
And that is why the analog signal must first be low-pass filtered by an anti-alias filter.

The following quote may be enlightening:

As I understand it, aliasing means that a frequency above 22.05 kHz can result in the same samples that could have resulted from a frequency below 22.05 kHz.

Example:


The red signal is 33075 Hz, the blue signal is 11025 Hz. If sampled at the points with the blue circles they result in the same samples.
So if you sample the red signal this way with a 44.1 kHz sampling rate, then at playback it would result in the blue signal.
This mixup is only possible because you get less than 2 samples per full cycle of the red signal, so it can only happen with frequencies above 22.05 kHz.

It could be that it is very stupid of me to try "explaining" this in this intuitive way, because intuitive thinking sometimes leads to completely wrong conclusions about digital audio. The only way to get a full and secure understanding is by doing the math, study and verify every step of the mathematical proof of the sampling theorem

 

[sander99]

@iamivancb: By the way, those last few posts of you look like you are practising for an english writing exam, or like you used an AI chatbot, or both...

 

[bigshot]

I think AI is best at writing speeches for politicians who don't actually want to say anything.

 

[iamivancb]

@iamivancb: By the way, those last few posts of you look like you are practising for an english writing exam, or like you used an AI chatbot, or both...

I do apologise, but I wanted to convey that I just felt like that with those music files as well. Otherwise, I won't give my opinion on AI because this is an audio forum.

P.S.: I didn't reply until now due to some inactivity lately on the forum.

 

[gimmeheadroom]

I do apologise, but I wanted to convey that I just felt like that with those music files as well. Otherwise, I won't give my opinion on AI because this is an audio forum.

P.S.: I didn't reply until now due to some inactivity lately on the forum.

Audio Inactivity

 

[iamivancb]

Audio Inactivity

Hehe, you have a sense of acuity, right?

 

[71 dB]

I think AI is best at writing speeches for politicians who don't actually want to say anything.

Politicians have been giving speeches where they don't actually say anything as long as there have been politicians. They are very talented individuals able to convince ignorant people vote for them. They don't need AI for that.

 

[iamivancb]

Politicians have been giving speeches where they don't actually say anything as long as there have been politicians. They are very talented individuals able to convince ignorant people vote for them. They don't need AI for that.

Agreed, just look throughout history.