turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
I don't want to go on forever about the 224, because really, I think that between the algorithms and the main sliders - which we've already covered - it's a pretty compelling plug-in. It's rare to find a plug-in that's capable of obtaining lots of different sounds but within a defined framework.
One point I haven't really mentioned is that the parameter ranges vary depending on which program you use. For example, take predelay. With the plates, the range is from 0 to 107 ms. The Room A predelay goes from 24-255 ms, while the large concert hall goes from 24-152 ms...you get the idea. Diffusion isn't adjustable in the Acoustic Chamber, but otherwise you can adjust diffusion for all other programs, including the Chorus. Treble Decay has different characteristics in different programs, and the like.
This is a big part of why the 224 is such an interesting reverb - there are considerable variations available within each program; they're not just minor variations on a theme.
Anyway, here are some fine points about the 224.
But he pointed out two things. First, for some people, that WAS a part of the sound and therefore, for the emulation to be accurate, it needed to be able to offer that sound. Besides, as with the 224, you could disable it. Second, and more intriguingly, some listeners preferred the sound with the noise and not for nostalgic reasons - they thought it added a useful sonic character that was more interesting than a sound without noise.
Prior to that conversation, I would have used this post to say "Hey UA, why do you emulate noise when you don't have to?" But now I know better.
I presume UA's motivations for being able to add system noise are similar, but I'd be interested if the UA designers can offer any additional insights. Yes, I'm a geek, and yes, I find this kind of thing interesting.
Next, we'll describe the "hidden features" area if for no other reason than to show off the cool graphics, then get into the Chorus, then wrap up the 224 before moving on to...well, it's a tough choice actually, but I think I'm going to do the Studer tape emulation next.
One point I haven't really mentioned is that the parameter ranges vary depending on which program you use. For example, take predelay. With the plates, the range is from 0 to 107 ms. The Room A predelay goes from 24-255 ms, while the large concert hall goes from 24-152 ms...you get the idea. Diffusion isn't adjustable in the Acoustic Chamber, but otherwise you can adjust diffusion for all other programs, including the Chorus. Treble Decay has different characteristics in different programs, and the like.
This is a big part of why the 224 is such an interesting reverb - there are considerable variations available within each program; they're not just minor variations on a theme.
Anyway, here are some fine points about the 224.
- The 224 is a true stereo processor that processes each channel separately, but if the input is mono, it splits to both channels.
- There are four outputs, but I'm a little confused here. There's a Rear Outs control that swaps outs A and C (the "normal" stereo outs) and outs B and D, which can provide quadraphonic reverb. But, I have no idea if all four outs are available simultaneously, or how you'd use them...then again, I'm not particularly conversant in surround. Maybe someone from UA could describe a typical scenario for using the four outs?
- There's a feature called "Mode Enhancement" which originally, was preset to an idealized tap modulation setting for each preset. But in the 224, you have control over the amount and degree of modulation, which essentially means one of Lexicon's "secret sauce" parameters is now available for your tweaking pleasure. Frankly, I haven't figured out a way to really quantify this; it's more of a "mess with it and see what happens" control. But every now and then, I hit on something useful.
- Decay Optimization is another option that was part of the original 224 but frankly, I never even knew it was there until I read about it in the UA documentation. Anyway, according to the documentation, it was optimized for individual programs and improves "reverb clarity and naturalness by dynamically reducing reverb diffusion and coloration in response to input signal levels." Like Mode Enhancement, this is one of the "tweak until it sounds good" controls, although I found the default values nailed it most of the time.
But he pointed out two things. First, for some people, that WAS a part of the sound and therefore, for the emulation to be accurate, it needed to be able to offer that sound. Besides, as with the 224, you could disable it. Second, and more intriguingly, some listeners preferred the sound with the noise and not for nostalgic reasons - they thought it added a useful sonic character that was more interesting than a sound without noise.
Prior to that conversation, I would have used this post to say "Hey UA, why do you emulate noise when you don't have to?" But now I know better.
I presume UA's motivations for being able to add system noise are similar, but I'd be interested if the UA designers can offer any additional insights. Yes, I'm a geek, and yes, I find this kind of thing interesting.Next, we'll describe the "hidden features" area if for no other reason than to show off the cool graphics, then get into the Chorus, then wrap up the 224 before moving on to...well, it's a tough choice actually, but I think I'm going to do the Studer tape emulation next.
There are now eight music videos posted on my YouTube channel, including a cover of "We Gotta Get Out of This Place," which joins "Little Pieces", "Black Market Daydreams," and "When the Grid Goes Down" (complete with disturbing video )
)
Oh, and since the UA engineers seem pretty forthcoming about what's coming up in the future...are the rumors I'm hearing about 64-bit compatibility true? I've had no problems using the UA plug-ins with 64-bit Sonar using its built-in bit bridge, but wonder what kind of benefits native 64-bit compatibility would provide.
There are now eight music videos posted on my YouTube channel, including a cover of "We Gotta Get Out of This Place," which joins "Little Pieces", "Black Market Daydreams," and "When the Grid Goes Down" (complete with disturbing video )
)
Originally Posted by Siggidori
How about creating a Soundcloud account and upload the 24bit aiff/wav files for us to listen to? + The youtube ones of course
But, I'll look into it because if one person is asking for something, that represents a lot of people who also want it but don't take the time to ask.
There are now eight music videos posted on my YouTube channel, including a cover of "We Gotta Get Out of This Place," which joins "Little Pieces", "Black Market Daydreams," and "When the Grid Goes Down" (complete with disturbing video )
) Originally Posted by Anderton
Here's a screen shot of what happens when you open up the "secret chamber" in UA's emulation.
The two buttons on the left are pitch shift up and down, which affect the Mode Enhancement option we discussed previously. The four knobs are input and output trims - obvious enough, while the Link control sets whether the input and output level controls are ganged or not.
The little UA logo is normally lit, which indicates that original programming bugs affecting the Hall B and Chorus algorithms are fixed. But if you actually like the sound of pops in the Chorus program's right channel, be my guest...
The Hold switch is the "I'm not looking at the plug-in all the time" switch. It sorts of works like a peak hold function on a VU meter. When set to 1.5 seconds, whenever you modify a parameter, the readout displays the most recent value for 1.5 seconds before reverting to the default display (current decay time). When set to hold, the modified value is held until you either edit another parameter, or change the hold switch back to 1.5 seconds.
The power button enables/disables processing. Although disabling the 224 does not free up additional DSP, bypassing it in your host does. So why have a separate power switch? Three reasons:
The two buttons on the left are pitch shift up and down, which affect the Mode Enhancement option we discussed previously. The four knobs are input and output trims - obvious enough, while the Link control sets whether the input and output level controls are ganged or not.
The little UA logo is normally lit, which indicates that original programming bugs affecting the Hall B and Chorus algorithms are fixed. But if you actually like the sound of pops in the Chorus program's right channel, be my guest...
The Hold switch is the "I'm not looking at the plug-in all the time" switch. It sorts of works like a peak hold function on a VU meter. When set to 1.5 seconds, whenever you modify a parameter, the readout displays the most recent value for 1.5 seconds before reverting to the default display (current decay time). When set to hold, the modified value is held until you either edit another parameter, or change the hold switch back to 1.5 seconds.
The power button enables/disables processing. Although disabling the 224 does not free up additional DSP, bypassing it in your host does. So why have a separate power switch? Three reasons:
- By turning off the effect with the power button, you can bring it back in with no delay. When you "disconnect" the effect from the DSP, re-enabling it will create a slight hiccup.
- If you want to do something like cut short a really long decay, you can do so by turning off the power.
- You can use it as a creative effect by switching the reverb on and off without causing any glitching.
There are now eight music videos posted on my YouTube channel, including a cover of "We Gotta Get Out of This Place," which joins "Little Pieces", "Black Market Daydreams," and "When the Grid Goes Down" (complete with disturbing video )
)
Originally Posted by Anderton
Thanks for the suggestion! I've actually thought about that, but in some previous pro reviews, I made both downloadable audio examples and YouTube movies available. The number of YouTube views exceeded downloads by a huge factor - I think people liked the convenience of just being able to click on the video, and after watching it, didn't feel the need to listen to higher quality audio. I must say the YouTube audio is better than I expected, I upload a really high-resolution file so I guess the encoding doesn't beat it up too badly
But, I'll look into it because if one person is asking for something, that represents a lot of people who also want it but don't take the time to ask.
But, I'll look into it because if one person is asking for something, that represents a lot of people who also want it but don't take the time to ask.
The option of being able to make timed comments on Soundcloud could be interesting though.. if people bother to use it that is
Well I think we’ve pretty much covered the Lexicon 224, and I only have a few Summer NAMM videos left to upload to the HC Video Library (they’ve already all been uploaded to our YouTube channel), so let’s turn our attention to another powered plug-in—the Studer A800 tape emulator.
Everybody likes tape, right? Well, almost everybody. If I never have to record a classical guitar on analog tape again, that will be too soon. But, there are many types of music that tape flatters, as well as many individual instruments. I’ve talked to several top engineers who like to track to tape specifically for that reason, then bounce to digital to preserve the sound. It sure would be convenient, though, if you could simply take the tape recorder out of the equation, and get “that” sound in a totally digital environment. Apparently UA thought that was a good idea, too.
The Studer A800 plug-in is the result of a joint partnership between Studer and Universal Audio. UA was able to secure a “golden” unit from Ocean Way Studio, and set about doing the modeling. Studer provided the listening feedback—hey, if they don’t know what their units sound like, who does?—and gave it a final blessing.
Of course, if you’re going to model something, you’d better make sure it’s in good shape (just ask any company who does amp sims what they went through to get the best possible tubes for the amps they were modeling). As a result UA brought in consultant Jay McKnight, who’s an expert on tape machine setup and calibration, so that they’d be modeling a properly-tuned machine.
Here’s what part of the interface looks like. You can also open it up if you want to calibrate it yourself, as we’ll find out when we start getting into the audio examples and such.
But before we do, let’s touch on one more important subject: The A800 answers the question “so why do I need a DSP hardware card, given that native processing is so good?” The A800 is not shy about gorging itself on DSP power, and the whole point of a multitrack tape recorder is, well, being multitrack. If you try to put that many native processors into a virtual environment, you could get away with it up to some point; but then you’d need to bounce, or freeze virtual instruments, or make other compromises in order to accommodate processing on every track. By having dedicated DSP, you can run plenty of A800s (we’ll specify exactly how many shortly) without touching the power in your computer’s CPU.
Okay, enough background. Let’s load some tape, and fire this sucker up.
Oh, right...I don't need to load tape.
Everybody likes tape, right? Well, almost everybody. If I never have to record a classical guitar on analog tape again, that will be too soon. But, there are many types of music that tape flatters, as well as many individual instruments. I’ve talked to several top engineers who like to track to tape specifically for that reason, then bounce to digital to preserve the sound. It sure would be convenient, though, if you could simply take the tape recorder out of the equation, and get “that” sound in a totally digital environment. Apparently UA thought that was a good idea, too.
The Studer A800 plug-in is the result of a joint partnership between Studer and Universal Audio. UA was able to secure a “golden” unit from Ocean Way Studio, and set about doing the modeling. Studer provided the listening feedback—hey, if they don’t know what their units sound like, who does?—and gave it a final blessing.
Of course, if you’re going to model something, you’d better make sure it’s in good shape (just ask any company who does amp sims what they went through to get the best possible tubes for the amps they were modeling). As a result UA brought in consultant Jay McKnight, who’s an expert on tape machine setup and calibration, so that they’d be modeling a properly-tuned machine.
Here’s what part of the interface looks like. You can also open it up if you want to calibrate it yourself, as we’ll find out when we start getting into the audio examples and such.
But before we do, let’s touch on one more important subject: The A800 answers the question “so why do I need a DSP hardware card, given that native processing is so good?” The A800 is not shy about gorging itself on DSP power, and the whole point of a multitrack tape recorder is, well, being multitrack. If you try to put that many native processors into a virtual environment, you could get away with it up to some point; but then you’d need to bounce, or freeze virtual instruments, or make other compromises in order to accommodate processing on every track. By having dedicated DSP, you can run plenty of A800s (we’ll specify exactly how many shortly) without touching the power in your computer’s CPU.
Okay, enough background. Let’s load some tape, and fire this sucker up.
Oh, right...I don't need to load tape.
There are now eight music videos posted on my YouTube channel, including a cover of "We Gotta Get Out of This Place," which joins "Little Pieces", "Black Market Daydreams," and "When the Grid Goes Down" (complete with disturbing video )
)
Before getting into the individual controls, there are two “global” issues worth mentioning. The first is that the A800 doesn’t just give “tape sound” but also models the signal path and includes options for different tape formulations—a crucial aspect of emulation, as the tape you used, and how you calibrated it, had a major influence on the sound. The other is that as UA assumes you’ll be inserting the plug-in on multiple tracks, you can gang settings to maintain uniformity.
So how many instances can you insert, anyway? With a Quad UAD-2 card (the A800 won’t run on UAD-1 cards, sorry), what a coincidence—24 tracks! As you can see by the screen shot, each chip handles 6 instances.
Yes, the A800 was a 24-track machine, and now your DAW can do 24 tracks of A800 if you have a Quad card. Remember, all of this happens without loading down your computer’s CPU.
There are two control sections, the main front panel primary controls...
...and the secondary controls, which are basically the controls for which you normally needed a screwdriver and a steady hand.
Incidentally, when you’re showing the primary controls and you’ve selected to monitor the sync or repro heads, the tape reels rotate; it’s cute and all that, especially because the takeup reel changes based on the tape type you’ve selected,. But after a while, I started feeling like I was going to have a seizure or something so I’m very happy that if you click on the IPS label for the tape speed control, you can make the reels stop.
Let’s give an overview of the primary controls. Next post we’ll get into the secondary controls, and then present some audio examples.
So how many instances can you insert, anyway? With a Quad UAD-2 card (the A800 won’t run on UAD-1 cards, sorry), what a coincidence—24 tracks! As you can see by the screen shot, each chip handles 6 instances.
Yes, the A800 was a 24-track machine, and now your DAW can do 24 tracks of A800 if you have a Quad card. Remember, all of this happens without loading down your computer’s CPU.
There are two control sections, the main front panel primary controls...
...and the secondary controls, which are basically the controls for which you normally needed a screwdriver and a steady hand.
Incidentally, when you’re showing the primary controls and you’ve selected to monitor the sync or repro heads, the tape reels rotate; it’s cute and all that, especially because the takeup reel changes based on the tape type you’ve selected,. But after a while, I started feeling like I was going to have a seizure or something so I’m very happy that if you click on the IPS label for the tape speed control, you can make the reels stop.
Let’s give an overview of the primary controls. Next post we’ll get into the secondary controls, and then present some audio examples.
- For the tape forumulations, you have a choice of 3M 250, Ampex 456 BASF Studio Master 900, or Quantegy GP9.
- Available speeds are 7.5, 15, and 30ips
- The four choices of tape calibration fluxivity are from +3dB to +9dB
- Input/output controls
There are now eight music videos posted on my YouTube channel, including a cover of "We Gotta Get Out of This Place," which joins "Little Pieces", "Black Market Daydreams," and "When the Grid Goes Down" (complete with disturbing video )
) Originally Posted by Anderton
Cheers,
Billy Buck
Great comments! I was aware that you save DSP with mono, but didn't know that you MUST install mono instances on mono tracks to get the DSP savings. I've instantiated stereo versions and they work fine on mono tracks, but now I know better 
Any wonder why I like the pro review format? And I even get to play with the stuff and create audio examples!
Any wonder why I like the pro review format? And I even get to play with the stuff and create audio examples!
There are now eight music videos posted on my YouTube channel, including a cover of "We Gotta Get Out of This Place," which joins "Little Pieces", "Black Market Daydreams," and "When the Grid Goes Down" (complete with disturbing video )
)
Now let’s look at the secondary controls, as they make a huge contribution to the plug-in’s authenticity. But to fully appreciate what’s going on, we need to explain a bit about tape bias.
I used analog magnetic tape exclusively for over 20 years, and being a solderhead, got into the whole process of aligning and biasing tape. Recording to magnetic tape is anything but straightforward: you need to suspend a zillion little magnetic rust particles in plastic, tell them how to line up as they get dragged past an electromagnet, and feed them with a high-level supersonic signal just so the audio doesn’t sound like a buzzsaw (and you thought clock frequencies only related to digital...). Tape is temperamental, subject to environmental conditions, and expensive. So, I was very happy when digital recording became affordable—but the sound of tape is something else, and one of its cooler aspects is that the sound is customizable through a tape recorder's bias and EQ controls.
Like the plug-in itself, these controls were usually hidden behind a panel. You’d adjust them with a small screwdriver, and once set, you didn’t want prying hands to change them. Optimally, you’d check the bias and EQ for each reel of tape due to manufacturing differences, although some studios were lazier and readjusted only when changing brands or production runs.
Bias is the high-frequency signal that’s injected into the head to help overcome hysteresis (the tendency of tape to hold on to its magnetization until forcibly reset by the bias signal). The challenge is that bias influences frequency response, distortion, and noise level, but each requires different optimum amounts of bias. Therefore, you have to make tradeoffs.
Underbiasing gives the best high-frequency response, but is more prone to distortion. Slight underbiasing added an almost “exciter”-like effect if you got it right, as you had nice highs with a little crunch. Overbiasing reduced distortion, but resulted in less high-frequency response and if you went too far, the high frequencies would saturate sooner and you’d also compromise the signal-to-noise ratio.
All things considered, I generally went more toward over- than under-biasing to minimize distortion, and compensated for any lack of high frequencies with the EQ options that were also a part of tape calibration. Another option was to “cheat” and boost the highs going into the tape, then cut coming out, to reduce the amount of tape hiss.
Get the picture? How you calibrate tape is as much art as it is science, and one of the coolest aspects of the A800 is it brings that art to the plug-in world. Let’s look at the controls, going from left to right.
The HF control determines the high-frequency boost going into the tape, and below it, the bias control does the tricks alluded to above. Next you have separate high-frequency and low-frequency EQ for the sync and reproduction heads, and that requires a bit of explanation as well.
Early tape recorders couldn’t overdub because tape went past a record head first, then a playback head some amount of time later (the exact time depended on tape speed and physical construction; for example, the famous Sun Studios slapback echo sound was about 150ms due to picking up the echoed signal from the playback head). So if you wanted to overdub, when you heard the signal from the playback head you recorded at the record head, and playing them back at the same time caused a delay. DAW fans can think of this as the ultimate in lack of path delay compensation
Les Paul came up with a solution to this problem by creating a four-head deck with successive playback, erase, record, and playback heads. The signal from the playback head fed into Paul’s headphones, and was recorded into the record head at the same time Paul was playing—basically, sound-on-sound recording. Although Les Paul is often credited with the invention of sel-sync (the process of using part of the record head for playback, and a term trademarked by Ampex), that honor actually goes to Ross Snyder, an engineer at Ampex, who came up with the sel-sync head in the mid-50s; another engineer, Mort Fuji, did the circuit design. There’s no doubt, however, that Les Paul was an inspiration for sel-sync, and the first recorder with that technology was delivered to him.
Anyway, record heads have wider gaps than playback heads, making them less well-suited for high-frequency reproduction. Doing overdubs on early machines with sel-sync sounded like you were listening through tin cans, but the quality was good enough to use a reference for playing against. Over the years the technology improved to where the difference between the sync and repro heads was almost identical, but they always had a sonic difference to one degree or another, so both options are available in the A800.
Moving right along, we have hum and hiss controls, which of course were not variable on the original A800 but an inherent, uh, “feature.” These are for the purists who don’t think it sounds like tape unless there’s hum and hiss; I’m not a purist...next.
The NAB/CCIR buttons provide your choice of two different equalization curves. Another button turns the noise off or on (I have it taped to off, using virtual masking tape of course). Auto Cal is cool, because if you’re not into messing with the bias and EQ, you can just click “on” and you get what Studer considers the optimum calibration for whatever tape type you’ve selected. Click on “off,” and like yours truly, you too can obsess over details.
Finally, there’s the brilliant Gang Controls option, which as mentioned previously insures that if you’ve inserted a bunch of these into your tracks, tweaking one can tweak the others similarly so that you have a consistent tape sound across all tracks.
Did I hear someone say "Cool, now how about some audio examples?" Okay, we'll do that next. But first, any questions?
I used analog magnetic tape exclusively for over 20 years, and being a solderhead, got into the whole process of aligning and biasing tape. Recording to magnetic tape is anything but straightforward: you need to suspend a zillion little magnetic rust particles in plastic, tell them how to line up as they get dragged past an electromagnet, and feed them with a high-level supersonic signal just so the audio doesn’t sound like a buzzsaw (and you thought clock frequencies only related to digital...). Tape is temperamental, subject to environmental conditions, and expensive. So, I was very happy when digital recording became affordable—but the sound of tape is something else, and one of its cooler aspects is that the sound is customizable through a tape recorder's bias and EQ controls.
Like the plug-in itself, these controls were usually hidden behind a panel. You’d adjust them with a small screwdriver, and once set, you didn’t want prying hands to change them. Optimally, you’d check the bias and EQ for each reel of tape due to manufacturing differences, although some studios were lazier and readjusted only when changing brands or production runs.
Bias is the high-frequency signal that’s injected into the head to help overcome hysteresis (the tendency of tape to hold on to its magnetization until forcibly reset by the bias signal). The challenge is that bias influences frequency response, distortion, and noise level, but each requires different optimum amounts of bias. Therefore, you have to make tradeoffs.
Underbiasing gives the best high-frequency response, but is more prone to distortion. Slight underbiasing added an almost “exciter”-like effect if you got it right, as you had nice highs with a little crunch. Overbiasing reduced distortion, but resulted in less high-frequency response and if you went too far, the high frequencies would saturate sooner and you’d also compromise the signal-to-noise ratio.
All things considered, I generally went more toward over- than under-biasing to minimize distortion, and compensated for any lack of high frequencies with the EQ options that were also a part of tape calibration. Another option was to “cheat” and boost the highs going into the tape, then cut coming out, to reduce the amount of tape hiss.
Get the picture? How you calibrate tape is as much art as it is science, and one of the coolest aspects of the A800 is it brings that art to the plug-in world. Let’s look at the controls, going from left to right.
The HF control determines the high-frequency boost going into the tape, and below it, the bias control does the tricks alluded to above. Next you have separate high-frequency and low-frequency EQ for the sync and reproduction heads, and that requires a bit of explanation as well.
Early tape recorders couldn’t overdub because tape went past a record head first, then a playback head some amount of time later (the exact time depended on tape speed and physical construction; for example, the famous Sun Studios slapback echo sound was about 150ms due to picking up the echoed signal from the playback head). So if you wanted to overdub, when you heard the signal from the playback head you recorded at the record head, and playing them back at the same time caused a delay. DAW fans can think of this as the ultimate in lack of path delay compensation
Les Paul came up with a solution to this problem by creating a four-head deck with successive playback, erase, record, and playback heads. The signal from the playback head fed into Paul’s headphones, and was recorded into the record head at the same time Paul was playing—basically, sound-on-sound recording. Although Les Paul is often credited with the invention of sel-sync (the process of using part of the record head for playback, and a term trademarked by Ampex), that honor actually goes to Ross Snyder, an engineer at Ampex, who came up with the sel-sync head in the mid-50s; another engineer, Mort Fuji, did the circuit design. There’s no doubt, however, that Les Paul was an inspiration for sel-sync, and the first recorder with that technology was delivered to him.
Anyway, record heads have wider gaps than playback heads, making them less well-suited for high-frequency reproduction. Doing overdubs on early machines with sel-sync sounded like you were listening through tin cans, but the quality was good enough to use a reference for playing against. Over the years the technology improved to where the difference between the sync and repro heads was almost identical, but they always had a sonic difference to one degree or another, so both options are available in the A800.
Moving right along, we have hum and hiss controls, which of course were not variable on the original A800 but an inherent, uh, “feature.” These are for the purists who don’t think it sounds like tape unless there’s hum and hiss; I’m not a purist...next.
The NAB/CCIR buttons provide your choice of two different equalization curves. Another button turns the noise off or on (I have it taped to off, using virtual masking tape of course). Auto Cal is cool, because if you’re not into messing with the bias and EQ, you can just click “on” and you get what Studer considers the optimum calibration for whatever tape type you’ve selected. Click on “off,” and like yours truly, you too can obsess over details.
Finally, there’s the brilliant Gang Controls option, which as mentioned previously insures that if you’ve inserted a bunch of these into your tracks, tweaking one can tweak the others similarly so that you have a consistent tape sound across all tracks.
Did I hear someone say "Cool, now how about some audio examples?" Okay, we'll do that next. But first, any questions?
There are now eight music videos posted on my YouTube channel, including a cover of "We Gotta Get Out of This Place," which joins "Little Pieces", "Black Market Daydreams," and "When the Grid Goes Down" (complete with disturbing video )
)
Since you prodded me into being picky . . .
Originally Posted by Anderton
Bias and hysteresis are related, but not directly. The relation between the magnetic field strength (which is directly related to electrical current in the record head) that's trying to magnetize the tape and the magnetization of the tape has, like a vacuum tube, a range over which the relationship is fairly linear. Below the linear range, increasing the magnetic field strength makes very little change in the tape magnetization until you get to the point where it sort of "kicks into gear." When you reach the top of the linear range, the tape is saturated, meaning that no matter how great the field strength, the tape won't magnetize any further.
What bias does is centers the audio current in the linear range so that the changes in current going into the record head will result in the same changes in current coming off the play head. Bias was originally DC, a constant magnetic field which added to the alternating audio signal. The thing that made magnetic recording worth using for audio was the discovery of AC bias.
AC bias offers the improvement in linearity but with lower noise than DC bias. The reason why AC bias works is related to hysteresis. There's some "stickiness" as you move along the current vs. magnetization curve. When you're going up along the curve and change direction, the magnetization doesn't go back down along the same curve, but turns around slowly and lags behind the upgoing curve.
The combination of the upgoing and downgoing curves is called a "hysteresis loop" (unrelated to a ground loop or Fruity Loop). By using AC bias, you're constantly moving between the two curves and the average stays pretty close to a curve that follows the center of the loop.
This effect on high frequency response is known as "self erasure." By adding too much bias, you randomize too many magnetic poles, effectively erasing the tape. It takes a lot more field strength than what bias provides to effectively erase the tape, but the highs go first, and the more bias, the more high frequency erasure. You can't apply enough bias (at least not with conventional recording heads) to push the current into the upper non-linear region.
If the tape saturates, it's because of the audio, not the bias. Though you can get it closer to the saturation with more bias, the self-erasure usually takes over before you get very far into the non-linear region.
Too little bias and you don't get the audio up into the linear region of the curve. In this region, high frequencies are more effective in magnetizing tape than low frequencies, hence the appearance of increased high frequency response. The "crunch" is a result of the non-linearity at the bottom of the curve.
There's another effect of bias that's known as "modulation noise" or "bias rocks." I don't have a good explanation of what causes this, but it adds a sort of gurgling sound to the recording. It's another of the trade-offs when adjusting a tape deck.
The formulation of the tape oxide material is where the trade secrets lie. The goal, it would seem, would be to find a formula that gave the best linearity, lowest noise, and lowest distortion. This is one of the things that distinguished Agfa 468 series tape and made it popular particularly for recording acoustic music where you want the maximum dynamic range without driving the tape into saturation. Most other tapes have each of these characteristics optimized at a different bias level.
One of the things that Mike Spitz of ATR Service emphasizes and teaches in his tape recording alignment class is that tape doesn't have one particular sound, but there's a range of sounds that you can get from a particular piece of tape depending on how you adjust the recorder.
Tape equalization is a pretty deep subject. Since the current produced when you throw a magnet through a coil of wire is related to the speed that the magnetic field cuts across the coil, because high frequencies change 'direction' faster than low frequencies, assuming that recording current remains uniform over frequency, there is a natural rise in playback frequency response.
There's a gap in the head pole piece which is what the magnetization bridges, and that's where the playback action takes place. But when the wavelength of the recorded signal becomes as short as the gap is wide, the magnet no longer bridges the gap and the high frequency response drops like a rock. The head is designed so that at the highest tape speed (which, along with the frequency, determines the wavelength that's recorded on tape) the highest frequency we want to record will still bridge the gap. This is why increasing the tape speed extends the high frequency response of the recorder. High frequency EQ can't compensate for the gap loss, but it compensates for the high frequency rise with wavelength.
Low frequency EQ is necessary for a different reason. At low frequencies, the recorded wavelength approaches the length of the part of tape in contact with the head. The magnetic field takes a "short cut" through the head pole pieces which adds to the current induced when it passes through the gap in the head, resulting in some low frequency irregularities usually called "head bumps." Bass players really like this because the LF playback EQ is at a single frequency so the low frequency response is always a little irregular, and some tape machines are particularly favorable to the low end of a bass. Dan Dugan made an 8-channel parametric low frequency equallizer designed to be used to smooth out the head bump on the narrow gauge tape decks but they didn't sell very well.
Studer has always used a different head design for the record and play heads, hence needed to help out the record head with EQ when it's used for sync playback. Ampex discovered that at 15 ips, they could design a head that worked equally well for recording and playback, so the MM1000, 1100, and 1200 multitrack Ampex recorders used the same head in both positions, allowing them to use the same electronic design for both "normal" and sync playback.
TASCAM used the same design for their narrow gauge multitrack recorders, and Fostex took it a step further. They use a single head for recording and playback, saving the cost of one head, assuring that there will be no difference between sync and playback, but making adjustment difficult because you can't "see" your record adjustments in real time.
Personally, I think trying to simulate the deficiencies of a tape deck is pretty silly. There are better ways of being creative, but nostalgia has a big influence on how we think things should sound.
Originally Posted by Anderton
What bias does is centers the audio current in the linear range so that the changes in current going into the record head will result in the same changes in current coming off the play head. Bias was originally DC, a constant magnetic field which added to the alternating audio signal. The thing that made magnetic recording worth using for audio was the discovery of AC bias.
AC bias offers the improvement in linearity but with lower noise than DC bias. The reason why AC bias works is related to hysteresis. There's some "stickiness" as you move along the current vs. magnetization curve. When you're going up along the curve and change direction, the magnetization doesn't go back down along the same curve, but turns around slowly and lags behind the upgoing curve.
The combination of the upgoing and downgoing curves is called a "hysteresis loop" (unrelated to a ground loop or Fruity Loop). By using AC bias, you're constantly moving between the two curves and the average stays pretty close to a curve that follows the center of the loop.
The challenge is that bias influences frequency response, distortion, and noise level, but each requires different optimum amounts of bias. Therefore, you have to make tradeoffs.
Underbiasing gives the best high-frequency response, but is more prone to distortion.
Overbiasing reduced distortion, but resulted in less high-frequency response and if you went too far, the high frequencies would saturate sooner and you’d also compromise the signal-to-noise ratio.
Underbiasing gives the best high-frequency response, but is more prone to distortion.
Overbiasing reduced distortion, but resulted in less high-frequency response and if you went too far, the high frequencies would saturate sooner and you’d also compromise the signal-to-noise ratio.
If the tape saturates, it's because of the audio, not the bias. Though you can get it closer to the saturation with more bias, the self-erasure usually takes over before you get very far into the non-linear region.
Too little bias and you don't get the audio up into the linear region of the curve. In this region, high frequencies are more effective in magnetizing tape than low frequencies, hence the appearance of increased high frequency response. The "crunch" is a result of the non-linearity at the bottom of the curve.
All things considered, I generally went more toward over- than under-biasing to minimize distortion, and compensated for any lack of high frequencies with the EQ options that were also a part of tape calibration.
The formulation of the tape oxide material is where the trade secrets lie. The goal, it would seem, would be to find a formula that gave the best linearity, lowest noise, and lowest distortion. This is one of the things that distinguished Agfa 468 series tape and made it popular particularly for recording acoustic music where you want the maximum dynamic range without driving the tape into saturation. Most other tapes have each of these characteristics optimized at a different bias level.
One of the things that Mike Spitz of ATR Service emphasizes and teaches in his tape recording alignment class is that tape doesn't have one particular sound, but there's a range of sounds that you can get from a particular piece of tape depending on how you adjust the recorder.
Next you have separate high-frequency and low-frequency EQ for the sync and reproduction heads, and that requires a bit of explanation as well.
Anyway, record heads have wider gaps than playback heads, making them less well-suited for high-frequency reproduction. Doing overdubs on early machines with sel-sync sounded like you were listening through tin cans, but the quality was good enough to use a reference for playing against. Over the years the technology improved to where the difference between the sync and repro heads was almost identical, but they always had a sonic difference to one degree or another, so both options are available in the A800.
Anyway, record heads have wider gaps than playback heads, making them less well-suited for high-frequency reproduction. Doing overdubs on early machines with sel-sync sounded like you were listening through tin cans, but the quality was good enough to use a reference for playing against. Over the years the technology improved to where the difference between the sync and repro heads was almost identical, but they always had a sonic difference to one degree or another, so both options are available in the A800.
There's a gap in the head pole piece which is what the magnetization bridges, and that's where the playback action takes place. But when the wavelength of the recorded signal becomes as short as the gap is wide, the magnet no longer bridges the gap and the high frequency response drops like a rock. The head is designed so that at the highest tape speed (which, along with the frequency, determines the wavelength that's recorded on tape) the highest frequency we want to record will still bridge the gap. This is why increasing the tape speed extends the high frequency response of the recorder. High frequency EQ can't compensate for the gap loss, but it compensates for the high frequency rise with wavelength.
Low frequency EQ is necessary for a different reason. At low frequencies, the recorded wavelength approaches the length of the part of tape in contact with the head. The magnetic field takes a "short cut" through the head pole pieces which adds to the current induced when it passes through the gap in the head, resulting in some low frequency irregularities usually called "head bumps." Bass players really like this because the LF playback EQ is at a single frequency so the low frequency response is always a little irregular, and some tape machines are particularly favorable to the low end of a bass. Dan Dugan made an 8-channel parametric low frequency equallizer designed to be used to smooth out the head bump on the narrow gauge tape decks but they didn't sell very well.
Studer has always used a different head design for the record and play heads, hence needed to help out the record head with EQ when it's used for sync playback. Ampex discovered that at 15 ips, they could design a head that worked equally well for recording and playback, so the MM1000, 1100, and 1200 multitrack Ampex recorders used the same head in both positions, allowing them to use the same electronic design for both "normal" and sync playback.
TASCAM used the same design for their narrow gauge multitrack recorders, and Fostex took it a step further. They use a single head for recording and playback, saving the cost of one head, assuring that there will be no difference between sync and playback, but making adjustment difficult because you can't "see" your record adjustments in real time.
Personally, I think trying to simulate the deficiencies of a tape deck is pretty silly. There are better ways of being creative, but nostalgia has a big influence on how we think things should sound.
Originally Posted by MikeRivers
You may have missed a comment about the 224 toward the beginning of this thread:
I had reviewed the Waves Aphex emulation and made a snarky comment about why anyone would want to include noise in a plug-in. I mean, it's a no-brainer, right? Why model noise when you don't have to?
But he pointed out two things. First, for some people, that WAS a part of the sound and therefore, for the emulation to be accurate, it needed to be able to offer that sound. Besides, as with the 224, you could disable it. Second, and more intriguingly, some listeners preferred the sound with the noise and not for nostalgic reasons - they thought it added a useful sonic character that was more interesting than a sound without noise.
Prior to that conversation, I would have used this post to say "Hey UA, why do you emulate noise when you don't have to?" But now I know better.
I presume UA's motivations for being able to add system noise are similar.So I assume that from UA's standpoint, they decided "Well, if we're going to model a tape recorder, we're going to model a tape recorder." Somewhere, there's probably some customer who's using some esoteric feature of the plug-in that I'd never use. Granted, I'm sure most users will just click on auto cal, hit the "tape" hard, and that will do what they want. On the other hand, being able to re-visit the sounds you can get from over-biasing and under-biasing tape is useful. I definitely used those options with real tape (or is that "reel tape"?) to alter the sound, so there's no reason I wouldn't want to do that with a plug-in that's intended to duplicate tape.
There are now eight music videos posted on my YouTube channel, including a cover of "We Gotta Get Out of This Place," which joins "Little Pieces", "Black Market Daydreams," and "When the Grid Goes Down" (complete with disturbing video )
)
Thanks for the link, good stuff.
There are now eight music videos posted on my YouTube channel, including a cover of "We Gotta Get Out of This Place," which joins "Little Pieces", "Black Market Daydreams," and "When the Grid Goes Down" (complete with disturbing video )
)
I was wondering how I was going to get this across in an audio example, but actually, I think this one really gives a good idea of what's going on.
I used a drum loop from the Discrete Drums series, but chose one with a tom part and no cymbals or snares. The "sound of tape" is most obvious with percussive sound sources that have low harmonic content, and a tom loop works very well for that.
I used the 456 tape option at 15 IPS and set the calibraton to +7.5 because, well, I'm like that
After recording each of the four examples, I adjusted their levels for the same peak level. Therefore, if one of them appears louder, it's due to a higher average level because of the tape compression, not louder peaks.
The first four measures are the loop by itself, with no processing.
The next four measures are the loop with "neutral" tape processing - not much crunch, just a touch of "tapeness." Nonetheless, it gives a nice "lift" to the sound that will really be obvious if you listen on headphones.
The next four measures have heavy tape crunching. This is one of the tape sounds that's burned into my brain from pushing analog drum sounds hard on tape so they could cut through better on a mix.
The final four measures are at least to me, the most interesting of them all because I wanted to see if I could duplicate the sound that results from a slight amount of underbiasing, which is a technique I used quite a bit (particularly for acoustic guitar). This is another tape sound that's burned into my brain, because I always made this adjustment by ear. And wouldn't you know, I could make that same adjustment, with the same sonic results, using the A800 plug-in. I thought that was pretty effing impressive.
I used a drum loop from the Discrete Drums series, but chose one with a tom part and no cymbals or snares. The "sound of tape" is most obvious with percussive sound sources that have low harmonic content, and a tom loop works very well for that.
I used the 456 tape option at 15 IPS and set the calibraton to +7.5 because, well, I'm like that
After recording each of the four examples, I adjusted their levels for the same peak level. Therefore, if one of them appears louder, it's due to a higher average level because of the tape compression, not louder peaks.
The first four measures are the loop by itself, with no processing.
The next four measures are the loop with "neutral" tape processing - not much crunch, just a touch of "tapeness." Nonetheless, it gives a nice "lift" to the sound that will really be obvious if you listen on headphones.
The next four measures have heavy tape crunching. This is one of the tape sounds that's burned into my brain from pushing analog drum sounds hard on tape so they could cut through better on a mix.
The final four measures are at least to me, the most interesting of them all because I wanted to see if I could duplicate the sound that results from a slight amount of underbiasing, which is a technique I used quite a bit (particularly for acoustic guitar). This is another tape sound that's burned into my brain, because I always made this adjustment by ear. And wouldn't you know, I could make that same adjustment, with the same sonic results, using the A800 plug-in. I thought that was pretty effing impressive.
There are now eight music videos posted on my YouTube channel, including a cover of "We Gotta Get Out of This Place," which joins "Little Pieces", "Black Market Daydreams," and "When the Grid Goes Down" (complete with disturbing video )
) Originally Posted by MikeRivers
What I disliked the most about tape was hiss, followed by modulation noise and flutter/wow. Take those away, though, and tape becomes a signal processor with a very complex set of interactions among level, distortion, and frequency response.
I'm a "don't look back" kinda guy, and after having digital and analog recorders set up side-by-side for a while, came to the conclusion that I liked digital better. The problem was that analog had characteristics I could not remove, whereas I could add characteristics to digital if I wanted.
However, playing with the A800 plug-in is making me re-consider the value of tape as a processor. If UA had called this plug-in a Dynamic Distortion Response Processor, never mentioned tape, and included the same complement of controls but didn't include references to tape or tape machines, I'd think it was a pretty cool processor. Now, I wouldn't use it on EVERYTHING, because that was the problem I had with tape - I HAD to use its "processor" on everything. But the ability to use the processing as I did in the 2nd and 4th group of four measures in the previous post's audio example is pretty cool; it would be extremely difficult to re-create those sounds with conventional EQ or saturation.
So even though we haven't covered every permutation and combination of A800 controls, I've pretty much come to a conclusion: Forget about tape, nostalgia, people who romanticize tape because they didn't have to demagnetize/head lap/align/calibrate/lubricate/etc. etc. the $%^&* things, and forget about modeling. Does this plug-in make useful, unique sounds? The answer AFAIC is definitely yes. I can see myself using this judiciously on multiple tracks in a mix.
There are now eight music videos posted on my YouTube channel, including a cover of "We Gotta Get Out of This Place," which joins "Little Pieces", "Black Market Daydreams," and "When the Grid Goes Down" (complete with disturbing video )
)
Help Desk