The No.1 Website for Pro Audio
Cytomic "The Scream" Distortion Plugin
Old 23rd November 2019
  #271
Gear Maniac
 

Pretty sure Scream is a level up on them, particularly on Glue which is many years old at this point. Much earlier in this thread and the one at KVR the idea of a modern "Glue HD" update was talked about.

Personally I find the current slightly more linear, predictable and clean Glue algo to be much more useful in mixes than other supposedly more technically accurate SSL bus comp plugins.
Old 23rd November 2019
  #272
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by smellypants View Post
Excuse me Andy but would you say your other plugs "The Drop" and "The Glue" are on this extreme level of accuracy/authenticity and detail of modelling etc. as this plugin is?

Thanks
Regardless of how accurately (or not) The Drop emulates the hardware it is modelling, it is a very cool plugin! If you haven't yet, give it a try!

Alistair
Old 23rd November 2019
  #273
Cytomic
 
andy-cytomic's Avatar
 

Quote:
Originally Posted by smellypants View Post
Excuse me Andy but would you say your other plugs "The Drop" and "The Glue" are on this extreme level of accuracy/authenticity and detail of modelling etc. as this plugin is?

Thanks
The detail has been increased with each plugin I've done. The Glue is a pretty basic model featuring an ideal voltage to decibel amp, a waveshaper for the ratio, and the core envelope detector is modelled in full with the resistor and capacitor network with two diodes feeding the follower. The Drop models a bit more, for example the MS1 models two bipolar junction transistors, two resonance clipping diodes, a waveshaper for the jfet buffer, and a few resistors, capacitors, and a pretty idealised op-amp (finite open loop gain).

The Scream is a level up again, you can see from the model parameters and schematic the level of detail. The Glue will receive an update with more detail, but I think The Drop is pretty good where it's at with regards detail vs cpu tradeoffs.
Old 23rd November 2019
  #274
Cytomic
 
andy-cytomic's Avatar
 

Quote:
Originally Posted by btfnk View Post
Pretty sure Scream is a level up on them, particularly on Glue which is many years old at this point. Much earlier in this thread and the one at KVR the idea of a modern "Glue HD" update was talked about.

Personally I find the current slightly more linear, predictable and clean Glue algo to be much more useful in mixes than other supposedly more technically accurate SSL bus comp plugins.
The current model will remain as is for backwards compatibility. I'll add an HD button will switch to the new model with more detail.
Old 26th November 2019
  #275
Lives for gear
 
Mercado_Negro's Avatar
Quote:
Originally Posted by andy-cytomic View Post
The current model will remain as is for backwards compatibility. I'll add an HD button will switch to the new model with more detail.
when?
Old 28th November 2019
  #276
Gear Addict
 

Quote:
Originally Posted by andy-cytomic View Post
The detail has been increased with each plugin I've done. The Glue is a pretty basic model featuring an ideal voltage to decibel amp, a waveshaper for the ratio, and the core envelope detector is modelled in full with the resistor and capacitor network with two diodes feeding the follower. The Drop models a bit more, for example the MS1 models two bipolar junction transistors, two resonance clipping diodes, a waveshaper for the jfet buffer, and a few resistors, capacitors, and a pretty idealised op-amp (finite open loop gain).

The Scream is a level up again, you can see from the model parameters and schematic the level of detail. The Glue will receive an update with more detail, but I think The Drop is pretty good where it's at with regards detail vs cpu tradeoffs.
What about modeling gold can VCAs or some Op amps "magic"?
Old 28th November 2019
  #277
Cytomic
 
andy-cytomic's Avatar
 

Quote:
Originally Posted by linkstate View Post
What about modeling gold can VCAs or some Op amps "magic"?
In The Scream I use a Boyle op-amp macro model that includes the input transistors, and this gives extra second order harmonics, which increase with increasing drive. I’ll make sure to have a few of these in the signal path of the HD version of The Glue as well. I’ll also come up with a macro model of the DBX 220 amp using as few transistors as possible to reproduce the most important non-linearities of the circuit. It will be interesting to see (edit: and more importantly hear!) how much this adds to the sound compared to the ideal voltage to dB amp current in The Glue.
Old 12th December 2019
  #278
Gear Nut
 

Hmm i somehow missed these responses so thanks guys...

I already picked up the drop and it is really really great!

Gonna have to check out the scream next!
Old 8th January 2020
  #279
Gear Head
 
Mimieux's Avatar
 

Any update on progress Andy?
Old 8th January 2020
  #280
Quote:
Originally Posted by Mimieux View Post
Any update on progress Andy?
It feels like one of those projects where it's more about the tweaking of the product than it is about the releasing of it into the wild. Like it's being developed for the sake of the development and not for its effectiveness as a guitar pedal plugin. Maybe not, but I don't understand how it can be generally finished for ages yet it stays on the workbench while it receives minor tweaks that will only truly excite a small percentage of people. I would be using this in a slew of sessions, but I'm not putting it into commission until there's an actual release (can't risk having it one day become unavailable -- I got burned with a couple of other beta guitar pedal plugins that I can no longer use today). Just one man's opinion, of course.
Old 8th January 2020
  #281
Lives for gear
Quote:
Originally Posted by elambo View Post
It feels like one of those projects where it's more about the tweaking of the product than it is about the releasing of it into the wild. Like it's being developed for the sake of the development and not for its effectiveness as a guitar pedal plugin. Maybe not, but I don't understand how it can be generally finished for ages yet it stays on the workbench while it receives minor tweaks that will only truly excite a small percentage of people. I would be using this in a slew of sessions, but I'm not putting it into commission until there's an actual release (can't risk having it one day become unavailable -- I got burned with a couple of other beta guitar pedal plugins that I can no longer use today). Just one man's opinion, of course.
If what's holding it up is a circuit solver methodology that will dramatically improve CPU usage while allowing future products to get developed much more quickly than it should all be worth it. That certainly doesn't sound minor to me, and it's what Andy's alluded to as a major reason for the delays. I bet there's other stuff going on too that distract from this. However, yeah, it's definitely slow goings at Cytomic. I'd personally prefer more excellent products over very few perfect ones myself!
Old 9th January 2020
  #282
Cytomic
 
andy-cytomic's Avatar
 

Happy new year everyone, and all the best for 2020!

I have a good work life balance, and chip away as fast as possible on what I feel is most important at the time, but my family and health are more important to me than finishing products

The low cpu mode was definitely a big holdup for a while, but that's all done and dusted now, and I'm very happy with the results - the MD mode is around 1/3 the cpu of the HD mode, and still allows for variable battery voltage and bias, and full stereo randomisation of all component values - which is huge. Doing the R&D on this and then automating the whole process is a pretty big deal as the PDK method I came up with tackles efficiently solving highly parametric non-linear circuits in a generic way. Having this lower cpu mode is critical for a majority of people, since otherwise the plugin would be too much cpu hit for them to comfortably use.

There are a few minor bugs that have needed fixing and require doing new releases of all our products, and also support Catalina properly, so that's taken the last few weeks. The other main holdup was working on ARM NEON cpu support, which took quite a few months early on in 2019. Cytomic is already in a place to deliver all our plugins on ARM, for iOS, and for when Apple release macOS on ARM laptops - you will thank me when it comes!
Old 9th January 2020
  #283
Gear Addict
 

Happy New Year Andy!

The conundrum here is that your plugins are so good, so I, and I believe everyone in this thread, just want more.
So, I am hoping we'd get finished "The Scream" and something new as well this year.
Old 21st January 2020
  #284
Here for the gear
 

Quote:
Originally Posted by andy-cytomic View Post

the MD mode is around 1/3 the cpu of the HD mode, and still allows for variable battery voltage and bias, and full stereo randomisation of all component values - which is huge.
Nice! I thought we were going to lose battery voltage adjustment in MD mode. I've gotten some really interesting and varied tones out of starving different diodes while playing with bias. Including many similar to that gimmicky new "plasma distortion". Which is kind of funny considering they market it as "high voltage distortion".



I'm happy you took the time to sort out these great additional features and optimizations. As much as I would love more products, I would never want you to sacrifice quality for quantity. Keep being the Kubick of plugin design!



Really looking forward to the circuit randomization feature. Both for quickly rolling the dice on unexpected tones, and the subtle stereo possibilities.

Any chance we could get a randomizer for the LFO2 step values in The Drop? I love the RAND Knob's ability to have constantly shifting values for subtle changes ( and the odd step randomization feature is genius to get more drastic while still remaining somewhat controllable/musical). However, it would be great to quickly spin up a random set of values that would remain constant, for stumbling into unexpected rhythms.
Old 7th February 2020
  #285
Gear Nut
 

Quote:
Originally Posted by andy-cytomic View Post
I have a good work life balance, and chip away as fast as possible on what I feel is most important at the time, but my family and health are more important to me than finishing products

The low cpu mode was definitely a big holdup for a while, but that's all done and dusted now, and I'm very happy with the results - the MD mode is around 1/3 the cpu of the HD mode, and still allows for variable battery voltage and bias, and full stereo randomisation of all component values - which is huge. Doing the R&D on this and then automating the whole process is a pretty big deal as the PDK method I came up with tackles efficiently solving highly parametric non-linear circuits in a generic way. Having this lower cpu mode is critical for a majority of people, since otherwise the plugin would be too much cpu hit for them to comfortably use.
Stay healthy 2020.

Is there still a plot to capitalise on all this development to roll out a number of other pedals once the scream is done?
Old 8th February 2020
  #286
Cytomic
 
andy-cytomic's Avatar
 

Quote:
Originally Posted by 235711131731 View Post
Stay healthy 2020.

Is there still a plot to capitalise on all this development to roll out a number of other pedals once the scream is done?
Yes there is I want to roll out an update of The Glue first, but more pedals are on the way.
Old 5th April 2020
  #287
Gear Addict
 

I just read on Facebook that a patent from Dirk Ulrich from Plugin-Alliance on TMT technology was accepted. The patent describes the procedure for the randomization of the component values to emulate the channel-to-channel variation of the response of the analog equipment.
http://www.freepatentsonline.com/20170060527.pdf

How will it affect plans for component values randomization in The Scream and The Glue 2.0?
Old 5th April 2020
  #288
Gear Addict
 

We live in an age where Monsanto can patent seeds and plugin developers can patent value randomisation!

I'm sure there is much more to it than that, but like with the dynamic convolution patent it didn't stop Acustica developing the same technology from a different approach so there will be ways and means around it I would assume.
Old 5th April 2020
  #289
Gear Maniac
 

It's a US patent. As I understand it patents are limited to the country in which they are granted. Hence it will not apply to Australia, Europe, South America, etc.

Also, if this is tested in court I am not sure it will hold up. Not a lawyer, but imo this fails the "non-obvious" portion of the patent requirement. I'm not sure it even fully satisfies the "novelty" requirement, since differences in component tolerance are well known and so is randomisation. The only part that is novel is the application to digital emulations. And I'm not convinced that's clearly "non-obvious." It's just not been done before because as stated in the patent it's up to now been generally regarded as undesirable.
Old 5th April 2020
  #290
Cytomic
 
andy-cytomic's Avatar
 

Quote:
Originally Posted by meloco_go View Post
I just read on Facebook that a patent from Dirk Ulrich from Plugin-Alliance on TMT technology was accepted. The patent describes the procedure for the randomization of the component values to emulate the channel-to-channel variation of the response of the analog equipment.
http://www.freepatentsonline.com/20170060527.pdf

How will it affect plans for component values randomization in The Scream and The Glue 2.0?
Is this another patent that should never have been awarded? I'll contact Dirk and double check but I doubt very much such a thing is even enforceable. Randomising component values for circuit simulation has a very long history. It's called Monte Carlo simulation / analysis. Things like LTSpice can do it, and also process audio files, so is he going to go ahead and take on Linear Technologies for having this feature? Good luck with that.

http://electronicsbeliever.com/monte...tep-tutorials/
Old 6th April 2020
  #291
Lives for gear
 
Bouroki's Avatar
 

The latest TDR SlickEQ GE seems to have this better implemented as the LR differences are frequency & gain dependent, while Brainworx TMT has fixed static differences per channel. Sweep the knobs around in Plugin Doctor to see what I mean. TDR measured actual channels from an SSL console instead of randomizing values based on tolerances (which BX seem to have missed taking freq/gain dependency into account).
Old 5 days ago
  #292
Here for the gear
 

Hi Andy, Hope you and yours are safe and sane Any updates on the release date?
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump