Discussion Topic |
|
This thread has been locked |
Ksolem
Trad climber
Monrovia, California
|
|
God damn...
Walleye steals the show!!!
Nut, judging by the reviews you've got some nice speakers. I know what you mean about flat pro audio speakers vs. home stereo gear. Mine are out of the mastering studio, and their flat, full range, and not so much fun to listen to. So I add a tasteful amount of sub, and voila.
|
|
phylp
Trad climber
Upland, CA
|
|
Generally speaking, among serious listeners Bose products are not well regarded Kris, that's what my husband used to say too.
|
|
Wayno
Big Wall climber
Seattle, WA
|
|
I recently upgraded my integrated amp. I have a decent pair of older JBL speakers and I still have a turntable but I don't need a tuner so the integrated amp instead of a receiver met my needs. Since a lot of my music comes from digital sources a good DAC and a variety of input formats was also desired. I ended up with a Peachtree unit with a tube pre-amp. No bass or treble nobs.
https://www.peachtreeaudio.com/
In one song I listened to I can hear violins that I never heard before.
|
|
Ksolem
Trad climber
Monrovia, California
|
|
A lot of people don't understand that the best level to listen to recorded music at is the level at which it would sound live. Most people play classical music way too loud. I go for the feel of being at the podium, mind you that can get pretty big at times but conductors aren't exactly losing their hearing every day...
|
|
Risk
Mountain climber
Olympia, WA
|
|
We had a dryer fire and in the laundry room was a massive old Onkyo receiver that didn't work and a high quality cassette component I never used. So, insurance would pay me for reasonable replacements. I opted for a Yamaha A-S301 integrated amp (stereo "hi-fi" with modern ad-ons) and a Music Hall USB1 turntable (who needs a cassette player?). Then, it was time to replace the woofer foam on my JBL 4412a's, so I had to send them off to Portland to get done. I went on Craigslist and spotted a pair of Sansui SPX-9000 speakers for $100 bucks and decided to get them while my JBL's were in the shop; massive 16" woofers and classic old-school wooden grills.
When I got the JBL's back and hooked them up, I compared them. Each were great, but surprisingly, the Sansui's are slightly better overall. I have them both on all the time (A and B).
My basic, low-end setup is friggin fantastic. The Music Hall turntable is more than adequate and the Yamaha 301 (lowest on the lineup, 60 watts x 2) is really like an old school stereo amp. If you're going back to vinyl, two-channel hi-fi is the way to go and the Yamaha 301 matched with the JBL and Sansui speakers crank as loud as the front row at The Who. It sounds best sitting back about 15 feet. I think the JBL 4412's are actually designed for listening 3 to 5 feet away.
|
|
NutAgain!
Trad climber
South Pasadena, CA
|
|
Kris, still relevant to this thread but slight drift...
What you said about preferred playback volume to match original source “being there”... does that mean you would master rock and roll and beyond type music at very high listening levels? Did you apply different EQ to account for Fletcher-Munson effect (less highs and lows, more mid emphasis for loud stuff) depending on source material and probable playback volume?
Is there a “sweet spot” dB level to balance background home or car listening vs concert emulation?
|
|
Ksolem
Trad climber
Monrovia, California
|
|
Okay folks, looks like NutAgain and I are heading for some serious geeking. It'll probably be the death of this thread.
One trend in the mastering scene beginning around 1990 or so has been to slam the levels on the CD, so in a digital editor the waveform looks like a brick, as if no one’s smart enough to turn up the volume if they want to hear it loud. This is especially true for rock music, rap, dance/club, and even R&B and gospel. And more recently even jazz. When I finished a master the first thing the producer or A&R person would do is put the CD reference copy in a player and make sure it was as loud as other music in the genre, even if it would have sounded better mastered at a lower level and played louder by using the knob that says "volume."
I was fortunate to have a big array of gear, so I could usually get the perceived loudness without simply crushing the dynamics to death. A typical scenario with a digital mix coming from a studio ran like this: Run the digital source through a dbx L2 set for about 2db limiting, but with the output level down a db below clipping. Then use a high end D to A (Pacific Microsonics) and run the analog audio through whatever I wanted, choice of Sontec, Manley, Focusrite eq’s, Manley compressor and limiter, etc. Then that processed (still dynamic) analog audio went through a separate Pacific Microsonics converter doing A to D. The resulting digital audio was recorded to hard disc on a Sonic Solutions workstation. If the master was analog tape, I'd go directly to analog gear and use the L2 later in the chain. And if the mix was already "slammed" there was little I could do.
Then I’d do all of the editing. At this point the digital audio is still 24 bit 96 kHz. So out goes the edited music through a Weiss digital console, loaded with eq’s, dynamics, efx., etc. The console is programmable, so I’d get the whole album dialed in, and the play it out through the console to a Pacific Microsonics converter set to dither down to 16 bit 44.1 kHz, and record that digital output to a second Sonic Solutions system as the finished master. So rather than having to slam the music up against the wall to make it loud, I had the luxury of being able to sort a massage it in various stages to get it sounding right.
One thing a lot of mastering engineers and mixers don’t understand is that a too much level below say 40 Hz eats up headroom at bass (50-150), mid, and high frequencies. Digital can only faithfully record 96 db at 16 bits, so if you’re clipping down at 30 Hz, your higher frequencies are going to start sounding really crunched. That Weiss console had a very cool 12 or 24 db/octave high pass which I almost always used at about 30 Hz to get more headroom higher up (the digital filter has no phase shift.)
I also found that a lot of mixes were a little fat in the low mids, and a little cutting there, 250-500 hz, would really open up the sound. I never gave much thought to the Fletcher-Munson curve, just tried to do what sounded good.
You’ll get a kick out of this. The rappers could never get enough bass when they were listening in my room, but as above, I couldn’t put all the bass they thought they wanted on the master. Since the main monitors had a programmable equalizer, I had a preset just for those sessions which sent 3db more lows to the speakers than to the master. I paid no attention to what I was hearing, just went by a spectrum analyzer. They’d come back saying “See, I told you it needed that extra bottom!”
Is there a “sweet spot” dB level to balance background home or car listening vs concert emulation?
Do you have a stereo pair of VU meters? Set them so -14 dB digital reads 0 VU. Then only exceed 0 or +1 on the meters on rare occasions. That should put you in the pocket. Old analog VU's are really cool, and shouldn't be too hard to find. Just split the monitor out from your mixer to feed the meters.
This is the type, There's less expensive one's too.
|
|
NutAgain!
Trad climber
South Pasadena, CA
|
|
Thanks for the details Kris... I'm a bit of a sound geek, some amateur digital recording experience of past band songs, but recently discovered how much deeper the hole can go with state of the art digital modeling of analog guitar amplifier components and speaker impulse responses. More nerd knobs to tweak in the products now than ever before.
Key takeaways I got from your recording process:
 Beware too much compression - the "brick" waveform :)
High pass (low cut) filters below 30Hz to give more room for frequencies more central to our hearing range (and within the range that average speakers can reproduce)
Cutting out some girth in the 200Hz to 500Hz range to "open up" the sound.
I've found very similar stuff while crafting guitar sounds:
Compression is tempting, because louder tends to sound "better" to a certain extent, but then you lose the punchiness and dynamic range. And for a full mix of instruments, lots of compression just makes my ears tired quickly.
With guitar, I typically have 12dB/octave cut at 100Hz, sometimes lower to 80Hz if I want more bottom end, but sometimes higher to 180Hz if I need to leave a lot of space for bass or keyboards.
I've seen that cutting some in the 200Hz - 500Hz range can be like "taking a blanket off" the sound, but I haven't worked out yet precisely which frequencies within that, or a consistent rule to apply. I always need to fiddle with it, but it seems that very low Q on a boost or cut tends to sound more natural, less boxy, than having a spike in the EQ graph. And I often add in a bit of 1900Hz or 2200Hz to make it "pop" more. But these details are what vary day-to-day in how much I like the result, and these are the things where the speakers you listen to make a big difference.
There, back on track to speakers!
The other key thing to mention- the room.
If folks want to upgrade to fancier speakers and specialized gear, the physical geometry of the room and composition of the surfaces might have a bigger impact. It will change which frequencies resonate and create standing waves, which acts like another EQ on the signal after it leaves the speakers.
Edit: Jim, I'm probably a bad judge of analog vs digital amp for music listening. I am not a cork sniffer. My recent "upgrade" of home music listening was a Sonos 3, through which I stream my mp3 collection. I'm not snobby about FLAC which I regret sometimes, but I am more pragmatic to enjoy music when/where I can in the format I have available (and I'm lazy so still use iTunes to organize my music). So basically, I enjoy geeking on sound stuff but I am not super-discriminating in my personal listening. Heck, I used built-in laptop speakers or apple earbuds for most of the last few years. I was disappointed with it, but accepted it.
Here is the most useful critical/comparative comments I can make for analog/digital in general:
1. Digital gear can produce sounds at a fidelity that human ears can't distinguish from analog gear.
2. A lot of what people find pleasing is a result of slight imperfections in analog equipment. Different old products have characteristic colorations or distortions that are well liked, and difficult to emulate with digital gear because it requires very complex modeling to capture the subtle ways the analog signals are distorted.
3. Any shortcomings or perceived differences of digital technology have to do with how much effort people have put into emulating in the digital domain what is really happening with the analog gear. The technology gets really deep really quickly in this space. It is evolving but it has gotten pretty damn good. Still not 100%, but plenty good enough for all but the most picky and competitive who are trying to find problems.
Talking now about guitar amp and speaker emulators, which I know more about than hi-fi stereo stuff:
1. Many modern musicians are fully stoked with the sound of digital emulated stuff- seems more so in the less discriminating range of heavy metal and very distorted players, but picking up up more classic rockers and even bluesy players too. The vast majority of people including musicians cannot tell the difference in A/B comparisons with well-adjusted sounds. But getting a "well adjusted" sound can be complicated. It is getting easier with each new release of software by the product vendors.
2. My personal experience, being in a band with an old-school really good lead guitar player (blues to 60s rock to steely dan and modern stuff), he was sometimes satisfied but often disappointed with my digital emulated tone. That has to be taken with a grain of salt because my lack of time commitment to dial it in properly, my dialing in sounds at home and no time to dial it in the band mix, and my lack of musicianship was also a factor. He was a highly complaining type that liked to pick on digital just on the principle. There is definitely a certain something great about a well-dialed analog guitar rig- a beautiful roar on stage. I think if I used my gear through a real guitar cabinet (which has a very limited and shaped frequency range) as opposed to the full-range flat-response speakers, it would have made it easier for me to dial in my sound. But the point of using flat speakers is so I can go from emulating a driven Fender Champ with a 10" speaker in one song, then next song emulating a big Marshall stack with a 4x12" Celestion G-30H speaker cabinet, or emulating a Hi-Watt with 4x12" Fane speakers. So committing to a real guitar cabinet (still with the emulated amp) might sound better on stage, but it loses the flexibility of different sounds.
3. For my home use, I can produce sounds that are sonic bliss with all digital gear through pretty "flat" (meaning a flat frequency response curve) speakers. The challenge is to get it dialed so that it still sounds blissful played louder in a band mix, opposite another guitar player using all analog gear and a loud drummer and loud bass player, etc.
|
|
NutAgain!
Trad climber
South Pasadena, CA
|
|
^^^ damn lot of words to not say much of practical value.
But I wanted to post that the audio gods are laughing at me now. I'm appreciating Neil Young Harvest album, but only can hear the left channel through my broken iPhone ear buds :)
And lots of ambient airport noise.
|
|
Ksolem
Trad climber
Monrovia, California
|
|
What are your opinions about the values and drawbacks between vacuum tube amplifiers and solid state amplifiers ?
Theoretically tube amps will sound "warmer" than solid state. But it's not fair to generalize too much because of the variety of designs out there. Take that amazing McIntosh amp Walleye teased us with up-thread. The audio circuits remain balanced through the entire amp. By balanced I mean that there are two parallel audio circuits all of which go through identical amplification, but one is out of phase with the other. At the end of the circuit the out of phase signal is inverted and the two are combined. Since any noise induced into either signal will be in phase, recombining the signals cancels every bit of added noise. So you can design an amp with lots of power without having to worry too much about adding noise because it will be cancelled out in the end. But it's very pricey to make the two parallel amplifier circuits. Then they used transformers for the outputs making it smooth like tubes.
To NutAgain, whether digital is virtually indistinguishable from analog depends on many things, most importantly the resolution. Digital audio cares about one thing: How loud am I now. Each time the question is asked, the answer is encoded in some number of bits. Asked often enough, and stored with enough bits, this information will represent the shape of a sound wave. A CD asks the question 44.1K times per second and stores each “sample” as a 16 bit binary number. Sounds like a lot? Well, many instruments have key elements of their sound up at 10 kHz. But at 44.1K we only get 4 samples to draw the waveform. That’s nowhere near enough to record or reproduce any harmonics at that frequency. The result is that as we move up though the sound spectrum, progressively the cd loses its ability to see and reproduce harmonics, the variations in the shape (not the frequency) of a wave which makes one instrument sound different from another. By the time we’re up to 6,8 and 10 kHz all of the sounds are made of the same soup, we only tell the cymbal from the piccolo by it’s dynamic envelope. A good analog system is more than capable of resolving these details.
To put it simply, the distortions in digital recording are omissions. The distortions in analog audio tend be artifacts added to the sound (hiss, ticks and pops, wow and flutter, etc.) but at the high end these distortions are almost non-existent. But the CD will never reproduce harmonics at 10kHz. Of course even the most critical listener can be fooled trying to hear what's not there. But the lack of high freq. harmonics on a CD is pretty audible once you get keyed into it.
Time for me to back slowly away from the keyboard now. Gotta go. Cheers.
|
|
NutAgain!
Trad climber
South Pasadena, CA
|
|
Kris, great point. I used to think Neil Young was a bit of a weenie for being against CDs, but your point about higher harmonics is the first argument I've heard (or the first time I heard it in a way I understood) that makes scientific sense. I hadn't properly considered it before, and I haven't done my own real observations related to high frequency distinction of instruments and different timbres when close to the sampling frequency.
Background ideas and concepts for folks who are curious:
Human hearing is at best up to 20kHz (twenty thousand waveforms per second), and typically only 12kHz or 14kHz for many adults (and progressively worse with age and exposure to loud and/or persistent sounds)
Engineers in most fields (not just audio) typically use the Nyquist frequency as the guideline for how fast you need to sample a signal. That means if we can hear sine waves up to 20,000 cycles per second, then a CD recording that captures 44,100 samples per second would seem to be good enough.
But Nyquist theorem doesn't tell the whole story for music (or any signal). That is the MINIMUM sampling frequency to avoid having "aliasing" or low frequency artifacts. It's the sonic equivalent of when you see a movie of old stage coaches and the wheels look like they are rotating backwards because the speed of the wheel spokes moving is too close to the speed of the images flickering in the movie. Google it for some interesting examples.
Many audio folks take the Nyquist sampling rate as a sort of holy grail for digital capture and playback, when it is just barely enough to avoid adding artifacts. The other consideration is what is lost.... As Kris said, it is not enough sampling to make a detailed picture of what is happening with the upper harmonics. A couple of principles here:
We can't hear the frequencies of the upper harmonics above 20kHz, but those signals can interact with each other in ways that generate lower frequency components that we CAN hear (do a google search on "beat frequency")
I'm less sure of this one, but google Virtual Pitch as a psychoacoustic phenomenon. If your brain hears a few harmonics but the fundamental frequency is missing, your brain adds in that lower missing signal. This is how small speakers can use fancy signal processing to make you think you are hearing a booming bass. I don't know if this concept still works for signals above 20kHz, where our brains might fill in a missing fundamental below 20kHz. This probably gets pretty detailed into how the Basilar Membrane works in our ear canal.
Anyhoo... harmonic content of this type (definitely from beat frequencies and maybe from virtual pitches perceived related to harmonics above 20kHz), is within a range that would be relevant for human perception. And it is chopped off even with CDs. That is the type of stuff that would affect our perception of timbre, e.g. distinguishing the tone of a piccolo from cymbals as Kris said.
Another place where missing this extra detailed sound is important is when you do sound processing in a musical recording- like to add a specific type of desirable distortion (e.g. to emulate a warm analog tube amp) or reverb or chorus or flanger effects and similar. In these situations, the signal gets degraded/distorted in an unintended way because there is not enough timing resolution in the hyper-sonic range to process the sound without missing stuff that affects the output in the range we can hear.
Some high end music products now work with 96kHz internally to avoid these problems, and some software might even do higher like 192kHz.
Any sound processing system that internally bottlenecks at 44.1kHz or 48kHz is likely to never make a "perfect" replica (from human perception perspective) of an analog sound. I wonder how much is really enough? 96kHz? 192kHz?
This could be explored theoretically by considering harmonics of instruments with high frequency components: piccolos, cymbals, snares, distorted guitars, etc... and then looking at how those harmonics interact to generate lower beat frequencies and which of those have enough power to be perceptible in an audio track.
|
|
Ksolem
Trad climber
Monrovia, California
|
|
I'm gonna have to look into that psycho-acoustic business. That's an amazing field. I had to study up on aspects of that when I was teaching a class about mastering and digital stuff for the UCLA recording program. For example MP3 uses a lot of psycho-acoustic trickery. One thing I learned, at least relative to the areas of the field I was interested in, is that psycho-acoustic perceptions vary from one individual to another.
An inconvenience with conventional digital audio is that to double the resolution with which the amplitude is encoded (bits) you just add one bit. 17 bits is twice the number of steps of resolution as 16, and so on. But to double the resolution of the sample rate requires twice as many samples.
Designers have made systems with very high sample rates, up in mega-Hertz, and two bit encoders. Basically if the sample rate is high enough all you need to encode is whether the signal is louder, softer, or the same as the previous sample. The audio simply cannot change fast enough to need any more data to encode. DBX made one for a while, I think they called it Delta-Mod. The fidelity was awesome, but the format was not robust, louder or softer does not tell you the actual amplitude so if you lose a few bits there's no way for the system to pick it back up and keep playing. Today a hard disc, vs. digital tape back then, would solve that problem. Except when you want to edit or process the signal.
edit, 96K by 24 bit is pretty darn good. For CD masters I preferred working at 88.2 by 24. The conversion to 44.1 for the product always sounded better to me. Of course that was on equipment made in 2005.
|
|
Ksolem
Trad climber
Monrovia, California
|
|
What is that thing? Judging by the wood grain it's pretty small. Is it edible?
|
|
big ears
Trad climber
?
|
|
|
Topic Author's Reply - Dec 5, 2017 - 06:56pm PT
|
Thanks for the info guys. Some of you are talking way over my head but I appreciate it. Basically, i enjoy the act of putting records on a turntable. Makes listening a more participatory experience IMO. Basically i want some gear that sounds good, and will last a while. Right now i have an old Denon middle grade amp, and a plastic audio technica table. The amp is starting to give me some trouble (the source buttons are a crapshoot, take you to different sources every time) and i can just tell by the feel of the table it doesnt have a lifetime of use in it.
This is why i chose the Sprout amp and the Project table. All in all its about 800 bucks, and eventually im gonna try to spend about 800 bucks on some speakers. Ive capped my cost at 2 grand. Is there something more practical i could spend my money on? Of course. But where is the fun in that.
🍻🍻🍻
Edit: part of the reason i went this route is that it seems like good speakers pop up used more often than a quality amp or turntable. So i bought this stuff on some black friday deals, and will scour craigslist until some nice Klipsch speakers pop up
|
|
zBrown
Ice climber
|
|
What kind of music do you enjoy listening to?
Well I like the live kind, but man, it's so expensive these days.
|
|
Ksolem
Trad climber
Monrovia, California
|
|
When looking at used speakers pull off the screen and inspect the foam surround. This is the foam bit around the diameter of the actual cone which keeps the cone centered in the magnetic "voice coil," and also allows it to move in and out. If they're cracked or separated, but the speakers are high quality you can get a deal. Just tell the seller they're shot but you'll take 'em anyway. There are plenty of places that will "recone" them for a fair price. I had a subwoofer (big driver) done locally near Pasadena by an pro for about $50.
|
|
big ears
Trad climber
?
|
|
|
Topic Author's Reply - Dec 5, 2017 - 08:29pm PT
|
Typically i am gonna be listening to rock, RnB, and any of those sub genres. Bands like The Punch Brothers, and Jazz Music somewhat regularly. Its very rare for me to listen to classical misic, but not unheard of.
Edit
What i have noticed with set ups of the caliber i will be buying have a spalshy and more alive sound to symbols. Its kind of hard to explain, but basically i want that “warmth”
|
|
|
SuperTopo on the Web
|
Let us know!