Tape Compression - how worky?

[Muckelroy]

I keep hearing about how awesome TAPE COMPRESSION is, but how does it work? Every time I overload a channel my quarter-track reel to reel, I get anything BUT a pleasing "compression" sound, more like a grungy distortion that's most displeasing, and cuts out frequently.

Just acquired a Tascam M-308, now I'm looking for a tape machine. Unfortunately, I don't have meters on every channel, so I'll just have to monitor the meters on the tape machine to see if I'm overloading or not, while tracking. I don't want to worry about watching the meters the whole time while tracking, if anything, I want good, clean TAPE COMPRESSION if it overloads. How does this work? will it work? If I got a decent TSR-8, and if the level of a channel was set just a BIT too hot, would I get some good sounding tape compression if that channel just went straight to tape?

thanks alot!
-callie-

 

[Beck]

It really depends on the kind of tape and machine you're using. Tape compression is produced before extreme levels. It's always there to some degree near 0 vu. You might be overloading your machine's electronics before you reach the tape's saturation level. That creates very harsh non-musical distortion. If you're using a +6 tape like 456, try using a +3 tape like 406 or 407. That's a nice trick on the TSR-8 in particular. Trying to hit 456 hard enough with dbx only creates bad distortion and artifacts. Put a 406 tape on it and it will compress the signal before the machine itself introduces distortion.

 

[acorec]

I keep hearing about how awesome TAPE COMPRESSION is, but how does it work? Every time I overload a channel my quarter-track reel to reel, I get anything BUT a pleasing "compression" sound, more like a grungy distortion that's most displeasing, and cuts out frequently.

Just acquired a Tascam M-308, now I'm looking for a tape machine. Unfortunately, I don't have meters on every channel, so I'll just have to monitor the meters on the tape machine to see if I'm overloading or not, while tracking. I don't want to worry about watching the meters the whole time while tracking, if anything, I want good, clean TAPE COMPRESSION if it overloads. How does this work? will it work? If I got a decent TSR-8, and if the level of a channel was set just a BIT too hot, would I get some good sounding tape compression if that channel just went straight to tape?

thanks alot!
-callie-

The tape compression that is talked about is not really attainable on consumer home recording type machines. Tape compression is achieved by overbiasing the tape when aligning the electronics of a pro tape recorder. The electronics in the pro tape recorder allow a much higher headroom than home recording type recorders.

The tape compression of a non-pro deck is almost non-existant because the electronics don't truly allow such slamming of the tape.

However, the non-pro tape recorders still give the "sound" of tape and the "warmth" that tape recording gives. This is not tape compression, but sound a heck of a lot better than most non-pro digital recorders.

 

[Beck]

It all depends on the machine/tape combination -- remember that and you'll be fine. A tape that saturates at lower levels will compress the sound at lower levels. In case you're wondering, overbias does not mean "over biased" as in too much bias. Overbiasing is simply a term that refers to adjusting bias to manufacture recommendations or industry standards.

Bias settings are the same on pro and semi-pro decks for a given tape. It depends more on head gap and tape speed. If you have a machine that can't properly bias the tape you're using you need to change to a different tape.

The TSR-8 has plenty of headroom to realize compression with tapes as hot as Quantegy 499 -- without dbx engaged of course. Noise reduction is a limiting factor with consumer and semi-pro decks.

-Tim

 

[Muckelroy]

Let's say that I calibrate a TSR-8 for 456, right down to the most meticulous of standards. Now, if I use 456 tape, I just want to get a solid level TO tape when tracking, the maximum level without overloading the electronics, in order to get the best signal to noise ratio.
Let's say I'm tracking drums, and I get a good solid level. If there's occasional peaks, and whatnot, will that form "tape compression" before it forms "overloaded electronics crappy bad distorted clippy sound"?

406, eh? So if I calibrated for 456, and then stuck a 406 tape on there, how would THAT go over as far as compression/overloading, etc?


tanx a bunch
-callie-

 

[Beck]

The key behind use of the 406 tape on the TSR-8 & MSR-16/MSR-24 is that the factory levels are set to 250 nW/m, which is actually under optimized for 456. It's set that way because of the integrated dbx noise reduction. nW/m (nano-webers per meter is basically a measurement of signal level to tape. 456 tape is considered 370 nW/m tape (though often set at 320). 406 is a 250 nW/m tape -- what the TSR-8 is set for So at 250 nW/m your vu meter will be at 0 vu.

When you pop a 406 tape on the TSR-8 you don't have to change a thing. Occasional percussive transients (short peaks) will not degrade sound or disturb dbx function. With some outboard dbx units you may hear "pumping." However, the TSR-8 has an unusually nicely integrated dbx system. Your sound will be different when using 406, and technically speaking, not within published Tascam specs. But the fun in all this is the tweaking and peaking your machine into a custom rock 'n roll hot rod, which you become master.

 

[Muckelroy]

so what you're saying is...

So, what you're saying is that the TSR-8 is factory calibrated for 406 tape? And that if I use 406 tape, and fine-tune calibrate it for 406 tape, it'll sound real nice, with that driving occasional compression like I talked about?

What if I calibrated for 456 and then USED 456? would I get a similar effect?

I know, you're probably all thinking "just buy a damn tape machine and figure it out," but I just wanna have an idea of what type of tape I need to use, and generally what you more experienced guys have discovered about them.

-callie-

 

[Beck]

So, what you're saying is that the TSR-8 is factory calibrated for 406 tape? And that if I use 406 tape, and fine-tune calibrate it for 406 tape, it'll sound real nice, with that driving occasional compression like I talked about?

What if I calibrated for 456 and then USED 456? would I get a similar effect?

I know, you're probably all thinking "just buy a damn tape machine and figure it out," but I just wanna have an idea of what type of tape I need to use, and generally what you more experienced guys have discovered about them.

-callie-

You can set the TSR-8 for 456 (320 nW/m), but you can't use the noise reduction at that level, so you run without.

Other Tascam decks, such as the 32 and 34, are factory set at 250 nW/m as well. Tascam states that they can be set at 320 nW/m for 456 if not using noise reduction. A lot of 30 series users are running 456 without noise reduction at factory settings. Technically they should be running at the alternate setting of 320 nW/m to get the noise and output benefits of the tape. The BR-20 has a switch to select 250 or 320 nW/m.

If you wanted to use a hotter tape on the TSR-8, such as 499 or GP9 you would run without noise reduction and set the machine up for the tape (520 nW/m). Using 499 or GP9 on a machine like the TSR-8 at factory settings with noise reduction is a waste of time and money -- not to mention the bias setting is also different.

-Tim

 

[Muckelroy]

hmm

You must understand that I'm a big fan of calibration. I really feel awkward about using tape that the machine is not preciseley calibrated to use. So whatever kind of tape I use, I want the machine to be calibrated for that tape, unless I had a good reason not to do so.

So if I were to use 499, and calibrated exactly down to tape specs, why can't I use dbx? (I know, cuz Tascam says so, but why? anyone?)

And if I don't use dbx, but have it all set up right, will there be any major increase in noise that might be more noticable than recordings w/out dbx?

Aight.

 

[Beck]

You must understand that I'm a big fan of calibration. I really feel awkward about using tape that the machine is not preciseley calibrated to use. So whatever kind of tape I use, I want the machine to be calibrated for that tape, unless I had a good reason not to do so.

So if I were to use 499, and calibrated exactly down to tape specs, why can't I use dbx? (I know, cuz Tascam says so, but why? anyone?)

And if I don't use dbx, but have it all set up right, will there be any major increase in noise that might be more noticable than recordings w/out dbx?

Aight.

dbx cannot encode and decode at those levels. Yes it will be noisier without dbx because dbx is for all practical purposes dead silent.

 

[Beck]

There's a lot more to this than meets the eye. It's a very complex topic. Tascam also had other design considerations when using 456 at 250 nW/m. One is that the harder you push a tape into compression the lower the frequency response will be. It doesn't kill frequency response, but it lowers it a bit, which doesn't look good on paper.

Tascam also introduced the TSR series in the digital age, when a cleaner (less compressed) recording seemed like a good idea at the time. That was before the rebirth of analog. Also, while the TSR-8 isn't exactly a narrow track format it isn't huge either. So pushing levels up will increase crosstalk, but even that is part of the "analog smear", as Tom Scholz calls it. It's part of what people are trying to get when they talk of analog warmth.

The bottom line -- there really is no one right way to setup a tape deck. It depends on what you want out of it.

-Tim

 

[Muckelroy]

interesting

I just got done talking w/ a colleague of mine, and he said that if I were to use 499 tape, I should calibrate it no hotter than +6 dB, due to the narrower track width of the TSR-8. Quantegy's calibration specs are really meant for the wider bandwidth, pro-model tape machines (according to him. I believe he's right.) He told me that calibrating a TSR-8 to 499 tape specs will push distortion and clipping RIGHT off the bat due to the higher saturation of the narrower track bandwidths.

Seems about right. However I'd much rather use 456 tape, because I feel more comfortable with something that's the STANDARD all around. So it would probably be a good idea to run 456 not at +6 (370 nWb/m), but at +3 (250 nWb/m), which is the factory default setting of a TSR-8 (right?)

Should I then set the bias to 3.0 (as quantegy says), or keep it at 2.0 (as the TSR-8 is set to)?

If I did all this, could I then use dbx, no problems, and have awesome recordings with that hot, punchy drum sound to it all?

 

[Beck]

I just got done talking w/ a colleague of mine, and he said that if I were to use 499 tape, I should calibrate it no hotter than +6 dB, due to the narrower track width of the TSR-8. Quantegy's calibration specs are really meant for the wider bandwidth, pro-model tape machines (according to him. I believe he's right.) He told me that calibrating a TSR-8 to 499 tape specs will push distortion and clipping RIGHT off the bat due to the higher saturation of the narrower track bandwidths.

Seems about right. However I'd much rather use 456 tape, because I feel more comfortable with something that's the STANDARD all around. So it would probably be a good idea to run 456 not at +6 (370 nWb/m), but at +3 (250 nWb/m), which is the factory default setting of a TSR-8 (right?)

Should I then set the bias to 3.0 (as quantegy says), or keep it at 2.0 (as the TSR-8 is set to)?

If I did all this, could I then use dbx, no problems, and have awesome recordings with that hot, punchy drum sound to it all?

Well, tell your colleague the TSR-8 is more than capable of using even Quantegy GP9, as I had mine set for it. 8 tracks on 1/2" is no more narrow than 32 on 2", and the TSR-8 is an exceptional machine all around. Not all machines of the same track width are created equal. By the time the TSR-8 came along in 1989 head design had improved over older standards.

The TSR-8 outperforms many older machines with wider track width. But as I said previously, when you increase signal (flux) level you will affect the other specs, such as crosstalk and distortion. That is not bad within limits -- just different. You can't drive tape into saturation without reducing frequency response. So if the goal is compression conservative settings won't do.

+5 (320 nW/m) is recommended by Tascam as the alternate setting for 456 w/o dbx on their machines, as it is for many other manufacturers. You don't need to trim levels down to +3, unless you use dbx.

For other brands or narrower formats, such as Fostex E-16 or R8, the above recommendations will vary.

The bias setting recommended by Quantegy is dependent on record head gap length and tape speed. It is not the same for all machines across the board. The Tascam spec is a good starting point, but even bias is not set in stone.

The TSR-8 has a huge range of capabilities -- factory recommendations are just a starting point. In fact, a manufacturer is not necessarily the best authority. With nearly any product, tips and tricks develop through experimentation into field practices, which were untried at the time the product was introduced. Show me ten identical machines owned by ten experienced professionals and I'll show you ten machines set ten different ways.

As you can see there is a wide range of settings for a given machine/tape combination. If you don't like a change you just change it back. That's too much for some people to handle, especially when looking for one right answer. In that case the only real consideration is the limitations one places on oneself. So any further discussion of tape compression is really just academic because the answer will always be to leave it at factory settings.

Experience gives one a different perspective. When I get behind the wheel of a car I don't have the reservations that I did when I was learning to drive at 15, because now I know how to drive.