Skip to main content
Topic: 33 vs 45rpm -- Technical Differences (Read 6041 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

33 vs 45rpm -- Technical Differences

My understanding is that differences in their sound quality can be attributed to the following:

  • 45s are typically shorter. This means that music needs to be less dynamically-compressed to avoid skipping as the grooves can be cut more deeply. This could have an effect on bass frequencies.
  • 45s are often mastered 'hot' for use as singles or for clubs. Listeners typically have a bias towards louder music, and thus dynamic-range compression can be interpreted as higher quality.

Am I missing something here? I have seen claims that 45s have a greater 'resolution' or lower SNR, but I don't immediately see how these points follow from groove density or playback speed of the recording -- surely this is related to the source and mastering of the music?

Let's say, for example, that I produce a short 2 KHz sine wave with consistent peak amplitude. I press it to 45 and 33rpm 12" records with no additional mastering. When played back, should the sound quality of these records differ -- and if so, how/why?

Re: 33 vs 45rpm -- Technical Differences

Reply #1
It's the same as with magnetic tape -- the more physical material you present to the read/write head in a given amount of time, the higher your resolution will be. Slow-running tape systems -- such as those used for surveillance and recording phone conversations -- have noticeably worse sound quality (more noise and worse frequency response) than faster-running tape systems designed for music recording/playback. In the tape world, tape speed is measured as IPS (inches per second). Higher IPS means higher sound quality, for the same given type of tape.

Similarly for LPs, you increase the resolution by increasing the speed, trading running time for quality.

Re: 33 vs 45rpm -- Technical Differences

Reply #2
Could you expand on why this is the case, please?

I understand how more frames/samples per second in a digital recording would (Nyquist factors not-permitting) produce a more accurate representation of the full frequency range of the audio signal. But when analogue represents a continuous and interrupted signal irrespective of playback speed, why does spreading that recording over more tape have the effect of reducing noise or increasing resolution?

For example, is there something in the nature of recording to a physical medium that means more of its inherent noise is picked up if the medium is read slowly?

Apologies if I'm being dense here.  :(

Re: 33 vs 45rpm -- Technical Differences

Reply #3
45rpm / 7" records are always playing at "the end of the record" so if your stylus is not very accurate then you will have a lot of inner groove distortion. Therefore, if you see a 33 1/3rpm 7" you should run but if it is 45rpm you can relax a little.

Re: 33 vs 45rpm -- Technical Differences

Reply #4
Could you expand on why this is the case, please?

I understand how more frames/samples per second in a digital recording would (Nyquist factors not-permitting) produce a more accurate representation of the full frequency range of the audio signal. But when analogue represents a continuous and interrupted signal irrespective of playback speed, why does spreading that recording over more tape have the effect of reducing noise or increasing resolution?

For example, is there something in the nature of recording to a physical medium that means more of its inherent noise is picked up if the medium is read slowly?

Apologies if I'm being dense here.  :(

Analog media isn't "continuous" as such, everything has a certain resolution, due to physical characteristics.

On tape it comes down to the size of the magnetic "grains". On LPs it is determined by the particle size of the vinyl, which again is influenced by the quality of feedstock used. If you use a lot of reground vinyl (old records or faulty pressings), that can affect the particle size, when compared to fresh "virgin" vinyl.

Re: 33 vs 45rpm -- Technical Differences

Reply #5
http://www.endino.com/graphs/

According to this site a lot of tape machines tends to have reduced bass response when running at 30IPS, so I guess there there is an optimal range of ideal operation in such analog media.

Re: 33 vs 45rpm -- Technical Differences

Reply #6
Faster should be better...

My quick & dirty calculations say the outer groove on a 12-inch 33 has higher linear velocity than 7-inch 45.   But of course the 45 is faster on the inner grooves. 

But in my experience 45's always had worse quality.   I believe this is the result of production differences...   Maybe hotter mastering/cutting, less attention to quality, and I've read it was common to use "regrind" vinyl.  The record companies didn't seem to care about quality and the customers didn't seem to care about quality either.

I don't remember ever hearing an LP that didn't sound better than the single...  Of course that's totally non-scientific..  It was  a long time ago, it was a small sample of cases where I owned the single and the LP, and it was not a blind test.  

Quote
◾45s are often mastered 'hot' for use as singles or for clubs. Listeners typically have a bias towards louder music, and thus dynamic-range compression can be interpreted as higher quality.
I wouldn't use the words "higher quality", but perhaps "more intense",  "more exciting", or "more appealing to teenagers", or something like that...

Quote
According to this site a lot of tape machines tends to have reduced bass response when running at 30IPS, so I guess there there is an optimal range of ideal operation in such analog media.
Probably just a characteristic of the particular tape machine(s) being tested.

Re: 33 vs 45rpm -- Technical Differences

Reply #7
Both of you have made the association 45 RPM = 7-inch. 
Such association is erroneous, even if 7-inch might usually be 45RPM.

In the DJ world, many LPs used to have one side at 45RPM with a single song (the iconic one), and the other side at 33RPM with two or at much three more tracks/remixes of the track.


As for the reasoning why 45RPM is expected to sound better than 33RPM:

KozmoNaut has already put some important detail:  analog is not continuous in the infinitesimal sense of it. He has already mentioned the fact that tapes are made of magnetized ferrite fragments (or whatever other component used for them) and that vinyl is a type of plastic that can have imperfections.
 
Now, the higher speed on vinyl generally allows for a stronger signal on the disc, which by itself already means an improvement in  the S/N ratio.
Then, of course higher speed means wider frequency band. That should come naturally by observation: when slowing down the playback speed, the sound is more bassy , and if speeded up, it sounds with more treble (strictly speaking, the bandwidth compresses or expands).
The exact same phenomenon can be experienced when playing a wav file on a computer with the incorrect sampling rate.


Re: 33 vs 45rpm -- Technical Differences

Reply #8
I've seen 45 rpm LPs with the same playing time (30 minutes- raga played by Ali Akbar Khan on sarod) as the 33 rpm version. OTOH, the same version of the album on 45 rpm might take two disks as seen with some reissues.

When we are mastering 45 rpm projects, we try to cut it has hot as possible so that it takes up the entire side of the project (be it 7, 10 or 12"). Cut hotter means more impervious to ticks and pops. It does not mean more bandwidth, as that is determined by the cutter (which is limited to 42KHz). I imagine that the speed does have an effect on bandwidth, but you would have to have a cutter than can go higher than ours to find out.

45s on 7" usually have the same groove spacing as at 33 rpm, but 10" and 12" can have wider groove spacing and therefore a hotter cut (which involves more amplitude in the groove modulation). There is an upper limit though- and that is the ability of the playback apparatus to track the groove. If the cartridge is not loaded properly (and most moving magnets cartridges in use are not) they can ring and make harmonic distortion; its my surmise that this has a lot to do with the idea that 45 rpm records have a hotter sound. IME they do not, other than the S/N ration is better because you usually have more signal to work with.

Re: 33 vs 45rpm -- Technical Differences

Reply #9
But when analogue represents a continuous and interrupted signal irrespective of playback speed
Analog media isn't "continuous" as such, everything has a certain resolution, due to physical characteristics.

Analog-y in digital:
https://en.wikipedia.org/wiki/Floppy_disk#Sizes,_performance_and_capacity
Why on earth spend 8" on 80 kilobytes of data when everyone above a certain age have had 1.44 megabytes fit on a much smaller area using the same principle? Because it is much more technically challenging to put the same signal on the same area (assuming you "read the surface", it is area).
Now fix the technology: area is proportional to the signal size you can write.
Audio playback is naturally constrained to 1 second of music per second (surprise!), so for a given tape or vinyl groove, lower speed means less area per second to carry the information, and less quality.
Memento: this is Hydrogenaudio. Do not assume good faith.

Re: 33 vs 45rpm -- Technical Differences

Reply #10
http://www.endino.com/graphs/

According to this site a lot of tape machines tends to have reduced bass response when running at 30IPS, so I guess there there is an optimal range of ideal operation in such analog media.


This site is high on the sad truth about analog domain frequency response. The last plot tells the story by means of comparison.

Even scarier is a problem that I have seen personally, but have not seen a lot written about.

(1) Record a 10 KHz tone at a suitable level, say -10 or -20 dB.

(2) Play it back.

(3) Note that it is  typically modulated with a low-frequency random envelope

Re: 33 vs 45rpm -- Technical Differences

Reply #11
This site is high on the sad truth about analog domain frequency response. The last plot tells the story by means of comparison.

Even scarier is a problem that I have seen personally, but have not seen a lot written about.

(1) Record a 10 KHz tone at a suitable level, say -10 or -20 dB.

(2) Play it back.

(3) Note that it is  typically modulated with a low-frequency random envelope

I don't know about this test method. Is this issue related to unstable capstan performance?

Re: 33 vs 45rpm -- Technical Differences

Reply #12
My understanding is that differences in their sound quality can be attributed to the following:

  • 45s are typically shorter. This means that music needs to be less dynamically-compressed to avoid skipping as the grooves can be cut more deeply. This could have an effect on bass frequencies.
  • 45s are often mastered 'hot' for use as singles or for clubs. Listeners typically have a bias towards louder music, and thus dynamic-range compression can be interpreted as higher quality.

Am I missing something here? I have seen claims that 45s have a greater 'resolution' or lower SNR, but I don't immediately see how these points follow from groove density or playback speed of the recording -- surely this is related to the source and mastering of the music?

Let's say, for example, that I produce a short 2 KHz sine wave with consistent peak amplitude. I press it to 45 and 33rpm 12" records with no additional mastering. When played back, should the sound quality of these records differ -- and if so, how/why?

To me cleaner has two dimensions - nonlinear distortion and noise.  I tend to characterize frequency response variations as *tone*, and so it is a different topic.

Distortion in LP playback has a geometric origin. Making a cutter and a player stylus follow the identical same path is not trivial, and becomes less trivial the shorter the recorded frequencies' wavelength (higher frequenceis). That all said, 2 KHz is generally too low for wavelength effects to dominate.

Noise in LP playback has several sources, including tics and pops, mechanical vibration, electronic noise, and physical imperfections in the playback media. All but electronic noise appear to be sensitive to speed of rotation and groove radius.  Faster rotation, more noise and more noise spectra at higher (typically more audible) frequencies.  Working against that is the fact that higher effective linear speed of groove tracking makes the same groove produce more electrical signal with velocity-sensitive pickups.

IME most single play 45's are hastily made with quick delivery in high volumes being of the essence. LP's are very sensitive to care during production and sacrifice that, and SQ quickly follows it. The performance of EP and LP 45s could follow their technical advantages more closely.




Re: 33 vs 45rpm -- Technical Differences

Reply #13

IME most single play 45's are hastily made with quick delivery in high volumes being of the essence. LP's are very sensitive to care during production and sacrifice that, and SQ quickly follows it. The performance of EP and LP 45s could follow their technical advantages more closely.


These days 45s are usually attempts to make a higher quality recording so you see most 45s in a 12" format; and often a full length project rather than a single. The days of the 7" 45 are long gone. Most of the singles I see these days are 12" 33 rpm.

Re: 33 vs 45rpm -- Technical Differences

Reply #14
Faster rotation, more noise and more noise spectra at higher (typically more audible) frequencies.  Working against that is the fact that higher effective linear speed of groove tracking makes the same groove produce more electrical signal with velocity-sensitive pickups.

Interesting. Do you know of any research/texts looking at relative performance of the two speeds? I'm particularly keen to see how the factors you've mentioned typically net -- that is, is it safe to assume that most music pressed at 45rpm will, ceterus parabis, yield less audible noise than if they were pressed at 33rpm?

This thread was first prompted by my seeing several 'audiophile' releases of albums, where a typical LP had been spread across two 45rpm records on account of its purported benefits to sound quality. Audiophile claims are nebulous at the best of times, but knowing more about digital audio, I wasn't sure if there might be something to this one.  :D

Re: 33 vs 45rpm -- Technical Differences

Reply #15

This thread was first prompted by my seeing several 'audiophile' releases of albums, where a typical LP had been spread across two 45rpm records on account of its purported benefits to sound quality. Audiophile claims are nebulous at the best of times, but knowing more about digital audio, I wasn't sure if there might be something to this one.  :D

There could be, if the groove modulation is also higher. There is certainly the room for it. Quite often though on such reissues, the master tape has degraded as they do over time. But it might be worth a listen, especially if the original is not available.

Re: 33 vs 45rpm -- Technical Differences

Reply #16
This thread was first prompted by my seeing several 'audiophile' releases of albums, where a typical LP had been spread across two 45rpm records on account of its purported benefits to sound quality.
Ah, my good old https://www.discogs.com/Metallica-Master-Of-Puppets/release/632458 . Which did sound better than my Master of Puppets LP sure ... and that could have quite a lot of different explanations, like e.g. length (MoP side B is 28 minutes, was that maybe a bit long?), the other LP is a picture disc, and it was bought second-hand (wear - the joy of vinyl!).
And, the 2x45 was direct metal mastered; I am sure you will still find discussions on whether DMM makes for a "harsh, too digital" sound (weasel words, moi?), but different masterings is always a point in itself: you cannot expect a "remastered" 2x45 to sound identical to an LP pressed thirty years earlier. Whether or not the "audiophile" pressings avoids using the phrase "remaster" on the sleeve.

But then I guess the 7" vs 12" format discussion was not a part of your question?
Memento: this is Hydrogenaudio. Do not assume good faith.

Re: 33 vs 45rpm -- Technical Differences

Reply #17
But then I guess the 7" vs 12" format discussion was not a part of your question?

I understand there are differences in distortion between the inner and outer rings -- presumably this would influence any comparison of 7" and 12" records, above and beyond any differences in their playback speeds.

As you allude, there are a great many other factors which influence sound quality such as DMM and tape sources. I'm curious about whether, all other things being equal, there is any meaningful positive effect on sound quality of a higher playback speed. In particular, I'm not certain whether the IPS/magnetic tape comparison is justified given some of the additional sources of noise raised by Arnold (reply #12).

In short... Does a clear technical case exist for using 45rpm over 33rpm, and has this been demonstrated in any research?

Re: 33 vs 45rpm -- Technical Differences

Reply #18
With 45's you have to get off the couch more often.  That's good for your health.  Following that line of reasoning, digital is bad for your health, LOL.

Re: 33 vs 45rpm -- Technical Differences

Reply #19
http://www.endino.com/graphs/

According to this site a lot of tape machines tends to have reduced bass response when running at 30IPS, so I guess there there is an optimal range of ideal operation in such analog media.


This site is high on the sad truth about analog domain frequency response. The last plot tells the story by means of comparison.

Even scarier is a problem that I have seen personally, but have not seen a lot written about.

(1) Record a 10 KHz tone at a suitable level, say -10 or -20 dB.

(2) Play it back.

(3) Note that it is  typically modulated with a low-frequency random envelope

Might as well add the rest of the sad story.  Yes, higher tape speed always results in less extended LF response, happens on all machines.  While 30IPS has a better noise characteristic, 15ips is actually the sweet spot for any machine, with minimal HF saturation with better high output HF response, reasonably low noise, and reasonable LF response.  Bass response at 7.5ips, 2 track 1/4" could actually be pretty flat to 20Hz, but the EQ caused HF saturation at higher levels.   Narrower tracks result in the "head bumps" you see on the graphs, which are cannot be equalized out using the eq adjustments in the recorders themselves.  Wider tracks result in less head bump, and lower noise because more tape area is being used. However, it's never a free ride with analog tape.  Wider tracks make azimuth more critical, which in turn makes guidance more critical.  Azimuth misalignment rapidly kills HF response.  So narrow track multitracks had lots of big head bump problems.

In addition to the amplitude modulation noise cited, which is caused by another trade off of high output tape vs high calendering (polishing) which moves the oxide away from higher output.  Higher output tape types are polished less, and thus have worse short-term drop-outs, which results in LF amplitude modulation noise.  Longitudinal tape vibration caused by an unsupported length of tape over static guides causes high frequency FM (the analog version of digital jitter) called scrape flutter, heard as modulated noise around pure tones.  Faster tape worsens the problem, roller guides is the only fix, but precision bearings are expensive making scrape flutter a "feature" of less expensive transports.  Bias purity and symmetry, or the lack thereof, also inject a form of LF noise. 

And that's just analog tape, which is then used to cue a lacquer 45 or LP.  As was mentioned, traditional 45s were usually cut hot so the tune stood out in play with other hot 45s.  All else being equal, louder presents as a positive "better" quality.  Hot 45s became such a problem that juke boxes employed compressors to smooth out differences.  Stereo 45s were rare until the early 1970s.  Lesser quality vinyl and higher modulation resulted in faster wear deterioration.  I can't speak to the audiophile tweaks of cutting 12" 45s, but it's still vinyl, still has physical limitations, notably the maximum HF modulation before the dimensions of the cutting stylus become the limiting factor, and LF maximums that vary in practicality depending on the pickup cartridge.  I guess it might be possible to cut a slightly better sounding 45 rpm 12" disc, but it seems quite impractical given play time limitations, and the fact that it is still vinyl.  But it's also highly unlikely anyone outside the mastering lab could ever make a meaningful comparison given the variables in mastering for different vintage releases.

 
SimplePortal 1.0.0 RC1 © 2008-2018