Deciding How Long To Make Each Side Of Your Vinyl Record

Photo: Echoes.in.side

A vinyl LP can hold over  40 minutes of music a side.  But, the sound quality isn’t good at all.

So, to make a great sounding record I have to be concerned about the playing time, and how the music itself sets the parameters for sound quality.  It’s complex, but I will break it down:

There are some guidelines, but it’s important to note that the length of the side, the level, and the bass response are all very closely related.

I have to look at the overall picture, and factor in many details.

Is the music very compressed or limited?  

Compression and limiting increase the average level, while lowering the peak level.  There are plenty of good reasons to do this for the benefit of the mix.  But, taken to the extreme—as is the case for most CD releases—that extra limiting increases the average level so much that I have no choice but to lower the cutting level.

Does the side have long passages of very quiet material?

It’s a numbers game.  You can make the louds really loud if you have soft sections where you can conserve space on the disk .  So, the more dynamic range you have, the louder the louds sound and the quitter the softs sound.  This isn’t ideal for high energy rock or pop, but this helps many other styles of music.

What is the overall bass “impression” of the side? 

If it’s a full bass sounding vibe, the record groves are literally deeper and wider, and take up more room. So I have to lower the level to make them fit. And there is “good bass” and “bad bass”  with regard to vinyl.  Bass that is dead mono and panned center doesn’t use as much space on the disk as “stereo bass”.   Keep an eye on the how much bass your mix has on the side.  Stereo bass makes the grove cut deeper. And, since deeper is wider (think trigonometry) that means that each groove requires more space.

How bright or aggressive are the high frequencies? 

Everything else being equal ,(and we know they aren’t, but for the sake of this discussion…) a record side that has bright vocal moments might not be able to be cut cleanly.  Let’s look at…

The Four Limiting Factors  

Compression (or, average level) 
The higher the average level, the lower the cutting level has to be. On vinyl, the effect of limiting the music is that peaks get lowered, but the average level can not be raised.

Bass level and Stereo placement
If a recording has loads of bass, that will use up space and cause us to lower the level.

High Frequency Content
If a recording has very bright moments, those will cause playback distortion and will cause us to lower the level.

Duration
If the side is “long,” then the level will have to be lowered to accommodate the extra grooves.

These four factors are always in play, working with us and against us in making an ideal album side. There are a few corollaries that can be helpful.

Vocals 
On a long side, the level will be lower. This means the high frequency vocal will not be a problem. Conversely, on a short side, bright vocals might not let me raise the cutting level to ideal.

Bass
Just a tiny compensation in the bass, either a sub bass roll-off or an elliptical equalizer, can be very effective at helping the cut.  They both affect the bass sound quality though, and must be used with good judgement.

Dynamics
Let the dogs run. Don’t use limiters on your mixes and masters for vinyl.  The effects of limiters on a vinyl cut are very different then they are on a digital master.  Hard to believe, but this is one place where softer is louder. Trust me on this one.

That said, here are your guide lines:

Under 18 minutes, I could cut almost anything.  As loud as wanted up to the groove distorting.  Full bass, metal, rock—it’s all good.

Between 18-22 minutes, I’m going to depend on a ballad or acoustic song to help with the length. If not, I’ll have to lower the level 2 or 3 db at 22 mins compared to 18 min.

From 22-26 minutes, I might have to lower the level more, or roll off a little bass. This depends on the client’s preference for bass/full/warm sounds vs level.

Over 25 minutes, it’s very likely—even for jazz—that i’ll have to lower the level, maybe clean up the low bass, and maybe more.

I can cut 30 minutes on a side, but the level is down 8db or more from the level I can cut on a 18 minute side. Thats a lot. But, the sound is still good (or can be) even though the record noise becomes more apparent since the music is quieter.  If pressed at a high-quality plant with careful consideration, a long side can sound really good.

The “right” answer to the question, How much music fits on a side of a record? comes from a decision made early on in the production. Are you making a nice sounding, good looking keepsake album? Or are you making a high-fi listening experience on vinyl?

If the prime purpose is promotional, don’t worry too much about the level. The result is, you will hear more click and pops, because the audio was cut at a lower level.  Put the saved money into great artwork or packaging.

If the purpose is a great sound and vibe, then keep the side below 22 minutes. Maybe 24 for traditional jazz that isn’t very compressed.

I hope this helps you make a great record!

Vinyl: How To Listen To Test Pressings

Masterdisk VMS-80The record pressing plant will press 5 -10 copies (test pressings) and send them to you.  I strongly recommend that you send a couple of these test pressings to your cutting studio to review. Your cutting engineer is the only person who knows how your record should sound.  I mean, you know how your music should sound, but the cutting engineer knows how the record should sound. And even more importantly, what the test pressings should sound like.

Here’s the rest of the story.: Test pressings are not perfect.  Even when everything else in the process is done perfectly, sometimes test pressings will exhibit flaws.  You have to be able to tell the difference between serious defects in the process, and ordinary flaws that are associated with the test pressings.
Almost every pressing plant utilizes a manual type press to make test pressings.  They are not often made in  ideal conditions, and always seem to be more noisy than the final records are.  There are several reasons for this:  It’s a different press, the temperature might not be optimum, and the quality of the vinyl may not be the same as they will use for your final run.

So, with all these potential flaws, why are we paying for test pressings?  You must listen to your metal work before pressing up thousands of pressings.  If there is a serious flaw, it’s much better to catch it before the big press run and all of the packaging costs. The main problem is, it takes so many steps to get from master to lacquer to father to mother and stamper and shipped back to you.

So, when you get your test pressings what are you supposed to listen for?

1. Any major flaw that would ruin the record.  Skips, locked groves (repeats)

2. Other audible flaws that would detract from the listening experience.   Once around or swishing sounds.  Harshness,  excessive high frequency distortion called Sibilants.

3. Surface flaws that may be annoying, and may be part of making records.  Clicks and pops: Small ones are normal, but, really loud ones that are plainly audible need to be investigated.

4. Oh yeah, and don’t forget to check the running order, gaps and any inscriptions in the “dead wax.”

Here is the big issue:  All test pressings have noises. Making a super short run on a manual press, it’s just going to be noisy.  The good news is, your final records should be quieter than your test pressings.  But there is a catch, and this is why we want to have more than 1 or 2 test pressing available.

If a noise—even a loud pop—is not audible on every test pressing, then the stamper is likely good and the process can be continued.  But, if that noise is on every test pressing, then plant should be contacted, alerted to a potential flaw in the stamper, and you should wait their reply.

When you (we) contact the pressing plant this is what they are ask for:

1. Is the defect visible? If it is measured with a stick ruler from the center hole to the defect,  tell the plant “at 5 1/2 “ from the center there is a hashy streak that’s visible.  If it’s a visible flaw but doesn’t have any annoying sound associated with it, you can safely ignore it.

2. Is the defect a repeated or cyclic noise?  If so, mention when you first hear it and when it stops.   Some defect sounds are only audible between songs. A particularly bad one is a squealing sound.  It might just be one squeak, or it could be a sustained whistle-like sound.  This isn’t good and needs to be brought to the plant’s attention.

This is a good time to remind you that the pressing plant doesn’t listen to the records that they make.  There are exceptions, but I don’t think any plant listens to the entire record.  They spot check it only, and often miss some issues.  Also, the playback environment within a pressing plant are often too noisy to have good play back listening.

The moral of the story is, you (we) have to evaluate your test pressings to make sure that errors and molding defects are not going to spoil your records.  And, you did leave yourself some extra time in case the test pressings don’t sound right?   Add an extra 2 to 3 weeks to the delivery time in case you have to reject the test pressings.

To document the disk, this is what i like to do.  I sit down in a QUIET room without interruptions, and I have my headphones, a stopwatch and a pad of paper.   As the record plays, if i hear anything that is loud enough to be disturbing, I note the time.  I reset the stopwatch to zero and start it again at the top of every song.  So, your list might look like this (if the test pressing is bad):

Side A  (test pressing #3 of 5)
Noise in lead in – swishing sound. once around,  noise floor is not consistent.

Song 1
0:30 loud pop (this is also audible on TP 4 but not on tp 5 or 6)
1:12  5 quick clicks in a row.

(between)
Noise floor is steady – slight surface noise, not swishy

Song 2
1:02 – 1:19  several loud pops – these are all audible on all of the test pressings, but not all as loud as TP3.
etc.

Also, note the surface noise between each song and at the end.  Some noises are ok.  Only the cutting engineer and a trained ear at the pressing plant can truly tell if a noise is bad enough to be rejected.  Records have noises… we like that about records.  But they shouldn’t have a lot of really loud noises.

It takes practice to listen to test pressings and make sense of it all.  Try to not make any guesses about what is causing the defect. It may be impossible for you to know – it’s difficult enough for us to know. But use the best onomatopoeia words that you can think of to describe the noise or defect.   Scratch, click, pop, scrape, bump, thump, ring, sizzle, crackle, snap, rattle, rumble…   remember the sounds these words make when you say them resemble the sound you are trying to describe. 🙂

Compare notes with everyone that received test pressings. Call your cutting engineer  (you do know who cut your record, right?)  You didn’t just send off a digital file to the plant to be cut?  Your cutting engineer, if he or she is really dedicated to helping you make great sounding records, will review your test pressings for little or no cost.  This is just simply a service we provide to help our clients wade through the confusing and frustrating parts of making a record. There isn’t any way you are going to get a great sounding record without the assistance of a cutting engineer. Because we can talk “record” with the plating plant.

Be patient, you can’t rush a great record. And do yourself an huge favor and do not book your record release event until you have approved the test pressings and have a confirmed delivery date from your plant. Even then, add two weeks for problems in the plant, printing problems, and shipping issues.

There is a reason why mastering engineers and producers love great sounding records.  We know how hard it is to get everything to happen perfectly.  But, when it does, like magic, your music is transferred into a completely different listening experience.