Skip to main content

OK, there's no way any of you would know this, but I thought I would check before I dive in myself. I have a CD master that needs to be duplicated onto cassettes with Dolby B. Thing is, the duplication system keeps it digital all the way until the signal is put on hub of tape which is then loaded into the plastic shell.

Does anyone know or have a rough idea of the EQ curve that Dolby B uses when making a master? I know it boosts the high end somehow, but I don't have the tools with me to analyze a before and after kind of thing. If I had some kind of spectrum analyzer, I was thinking I could record pink noise onto a cassette with and without Dolby B engaged and compare the two in a DAW. Then I could see basically what kind of EQ curve I needed to apply in the DAW for the CD master going to tape.
Any further ideas?

BRane hurt

Topic Tags


Reggie Mon, 07/25/2005 - 13:22

I'm open :D
Anyone know of one?

Yeah, I forgot about the dynamic part....I think Voxengo gliss EQ might be able to approximate something like that, but I still have no idea where to start for the EQ curve. Or well, I guess it starts at 1KHz, but I don't know where to go from there.

schizojames Tue, 07/26/2005 - 10:48

If one of you have the patience to explain the EQ curve extraction process...I can perhaps be of assistance. I don't have to work today, and I just picked up a really nice tape deck at ARC with Dolby B, Dolby C, Metal, CrO2 switches, and a spectrometer. I have it on a send/return with my DAW so if there is some kind of rough test I can perform to isolate the Dolby curve, I will do it.

It seems like all I would have to do is duplicate a mono file, record one into the tape player, send it back and invert it. The resulting spectrum analysis SHOULD closely approximate the difference in frequency response, right? Or would it be the opposite.

Sorry...last time I really got involved in a cassette player was when it ate my Thriller tape. But if I can be of assistance, let me know!

schizojames Tue, 07/26/2005 - 11:01

So, we are looking at something more like a side-chained EQ with a separate amplitude-dependent chain for each frequency point on the EQ? Some kind of multi-band compressor/expander? Do you think it would be possible to set up a series of plugins in Cubase that could accomplish this? I am asking because this is hurting my bain as well...

See, it made me say bain...

If someone is proficient at Max/MSP or Kyma....I know they could set that up....It's just the measuring that would be difficult.

On that note... page 4 of this PDF has an approximate chart for Dolby B.…

Ok...I'll wait to post again until I know I am not the only one interested in pursuing this.

Reggie Tue, 07/26/2005 - 16:46

Yeah......I think recreating the processing is going to be near impossible. Closest I can think is maybe getting lucky with a Gliss EQ curve, or maybe some kind of HF shelf boost starting at 1K followed by a Multiband Compressor to bring the boosted highs back down when the source material is louder. I don't think I have the patience. :? Thanks for the offer to do some tests, but now I'm not confident I could interpret the data in a way where I could set up an effect chain that would work accurately.

I'm honestly having trouble finding a place to buy a stand-alone Dolby encoder. The Model 422 looks like the ticket, but I can't find one for sale.
I'll know tomorrow if doing it the dirty way (CD, to chrome cassette master on Tascam122 with Dolby engaged, back to CD [Dolby disengaged on playback], back to bulk hub of chrome duplication tape) will work or not.

The sad thing is, I'm sure this will all be for minimal gain. I'd bet that just going straight from CD to duplication loop would turn out fine. Oi.

Michael Fossenkemper Tue, 07/26/2005 - 19:33

You could rent one possibly. I think a straight non dolby would be much better than a cassette bounce. It's been so long since i've dealt with dolby but i think is a h freq boost and then compression at the crossoverpoint, then reversing it for decode. what the points are, I have no idea.

You could record something like white noise to tape with it engaged and then play it back without it engaged to get the curve, but the compression would be left out of the equation and that's where you would run into some trouble, as well as tape compression and noise.


User login