Ok, brainiacs, let's see what you've got...

[Guest]

[quote author="bblackwood"]
Jakob, I don't know how to describe it technically. I just want to soften the high end - diffuse it,
Do you really not understand what I'm hearing vs. what I want to hear?[/quote]

Possibly. With added variable harmonic generation, something akin to the airbrush in photoshop - spreads the energy out to soften the 'image' without reducing overall energy.

Forgive me for being daft.. but reading those requirements... and trying to imagine the sound (as Steve Martin states - akin to Dancing about Architecture)
Are we talking about a sound similar to a touch of reverb or smear on the HF band with added harmonics on the reverb/smear HF portion only?

(sounds similar to the Troggs "sprinkling some f***g fairy dust" on a track)

 

[SSLtech]

I'm feeling something here...

Brad (for those who don't know him) has many great pieces of gear, has lots of experience, and a trusted set of ears. He knows what multi-band compression is and he knows what it does and how different versions do it. I'm sure he even owns several devices which can do it right out of the box.

Now, I think I might have a clue as to what he's wanting to hear sonically, but I'm by no means certain that he's for sure on the right track in describing what the technical requirement is. Only if we can quantify the effect and result correctly can we build a simulator or tool, whether the issue is digital or analog. (Following on from my recent comments on the EQ 'ping' device....)

Anyhow...

I'm not altogether certain that envelope transience is the thing. A compressor looks at the envelope and not the instantaneous waveform. A transformer deals with the instantaneous waveform and its rate of change, and is a magnetic device rather like tape. It has reluctance, remnance, permeability, and a number of other properties that tape has, though these are exploited and employed in a rather different way.

Of course, the best 'effects' devices (and I prefer to think of what I understand Brad to be searching for as an 'effect') have a way to control the amount of the effect.

I don't see that the non envelope-dynamically-related (i.e. compression based) suggestions so far have any way to govern or control the amount of the effect.

-So I don't have a way to get what I think Brad is looking for, but here are some "backward-thinking" ideas:

Brad, since the effect of tape "sound" was something that engineers always wanted to get rid of in days of yore, perhaps we should think of other things that were adjusted to proveide softening and under-performing. Something as simple as a couple of sets of Dolby 'A' units, with some sort of nonlinearity between them would induce dynamically -related softening of the signal. Dolby 361's had all-discrete class-A circuitry, with very nice transformers on them on both input and output. If you fire a signal into one, encode, decode with the other, and in between, load down the signal with a variable resistance directly across the signal lines it would provide a way to increase the load on the transformers ('bending the sound in an analog way) and also the slight level change would start to soften the HF progressively and mand-specifically, due to the way Dolby 'A' works.

Since 361's and 360's are so darned cheap on eBay at the moment, I'd seriously consider playing with this. -If it's not what you want, you can either use the 1RU cases for something else, score some serious transformers, or sell them on eBay again... they usually go for about $50 per channel if you hang around for a bit.

I just installed 72 of the darned things in a rack. I only used MH-type and SP type for multitrack 'A' type use before now, and I hated their clipping behaviour, but I assure you that the 361s are a different beast entirely. -Since the experiment involves no internal modification, you might even borrow a stereo pair and cascade them in mono to see if the effect is any use, with no internal modification at all...

Whaddayathink, Brad?

Keith

 

[bblackwood]

[quote author="buttachunk"]Brad-- can you be more specific about the diffuse HF you are looking for. is there a piece of gear which exhibits this diffusion pleasingly already ? if so, could the crossover splitting as described previously be used to process just the highs through that, or similiar unit ?[/quote]
Nah, not really. If there were, I'd have one/something just like it by now...

I've played witht he dolby stuff before and that's not exactly what I'm thinking, though with some mods, might be a viable alternative.

Compression is not what I'm after - I have colored compresors out the wazoo here for that sort of thing. I wish I knew how to describe it, but basically, I want something that will take a very 'sharp' or 'harsh' top end and smooth it out, but not make it dark. It's really about changing the character of the sound, but with a variable xover point and the ability to vary the effect (yes, this is an effect box) so that I can smooth out the top end without affecting the midrange.

This might be the kind of thing that's gonna require another hefty investment in customization from one of my buds who has designed/mod'd/built several peices for me...

 

[Learner]

Hi Brad,
Just out of curiosity, isn't that what tape simulators such as fatsoh and distressir is design to do? Do you not find them do what you want? if not, what do you think they do to your sound and what they are missing to achieve?

Just a thought, have you tried recording a sweeped signal of 20-20Khz on your tape machine and play it back to see the frequency response curve of your tape recorder? and may be duplicate the curve on a narrow band graphic EQ to see how close you can get to sound like the tape machine? May be after that you can start to investigate what is actually missing after duplicate the frequency response curve on the EQ and see if there are actually other factors that will effect the perception of the sound besides frequency response?

 

[TomWaterman]

Nobody like the variable slew rate idea????

I don't know enough about it but I imagine it to achieve an effect, like I imagine what you want to hear.

The GrooveTubes Vipre has variable slew from 0.75v/s to 6v/s a second. The pre does sound a little dark from my limited listening but if you contained that with a filter to just above say 6-8kHz then your critical midrange would be left relatively untouched. Apparently the Vipre is good at taming transient details while tracking, why couldn't a similar idea be used when mixing, mastering but more subtle than the SPL TDII or Sony plug....

Brad (I expect you have, but) have you tried splitting the audio within your daw (linear phase transparency should be easier to achieve in there - algorithmix plugs etc) and then spitting out the two streams via two DAC pairs.....

You could then process the HF portion with whatever you have available.....different compression, eq, slew rate controlled amplifier, harmonic saturation, transformer etc (although that would make more sense for the lowend in my mind....not sure about HF).

All you would need would be a 4in, stereo summing device to lump the two sections back together at the end before feeding the rest of the chain / ADC etc.

Just a thought. The LF/MF portion could be left relatively untouched and passive too, apart from the minimal damage from the digital x-over, extra DAC stage and final summing stage.

Good luck with it.

Cheers Tom

 

[Gold]

I use a Westrex HF limiter for just this application. Works pretty well. I know Clete Baker has one laying around he doesn't want.

 

[bblackwood]

Really Paul? I know you know what I mean!

That's it, I'm emailing Clete...

 

[Gold]

It uses an RC network with an LDR. I think it's in shunt but I actually don't have a schematic. It's not the fastest thing in the world. It's slower than the high frequency limiters in my lathe electronics but it works well for softening transients.

 

[Gold]

The Neumann HK66 which is in my VG66 cutting rack uses a FET with an RC network but it also gets it's control voltage from the circuit braker module. It directly converts the current through the cutterhead to a control voltage.

The later Neumann Acceleration Limiter is a super complicated beast. It is an analog multiband lookahead limiter. It splits the audio into two bands, uses inductors as delay lines and sums the bands back together. It has 18 inductors per channel!

The early tube Ortophon uses an LCR network to route audio through. Totaly passive. The only active stages are for the control voltage! It needs specific source and load impedances to work correctly.

The later Ortophon uses a Harris MM555 to switch active LPF's in and out.


That's the thumnail of approaches to HF limiting