Cytomic 'The Glue' Compressor

[djsubject]

The examples sound awesome. How about some vocals? Just to help keep me from pulling the trigger on the UAD-1 promo right now...

this is what i want to hear

i can supply some singing & rapping vocal samples if you need some?

Subz

Yes please! Just post a link here, or email me on info atsign cytomic dotsign com. And also please let me know if you want any particular style of compression - heavy, light, limit etc, otherwise I'll just compress them to whatever sounds good. Also if you email please let me know any credits you want possibly including a url.

Andrew Simper

i will email you later when i get home

Subz

[fjolle]

The examples sound awesome. How about some vocals? Just to help keep me from pulling the trigger on the UAD-1 promo right now...

this is what i want to hear

i can supply some singing & rapping vocal samples if you need some?

Subz

Yes please! Just post a link here, or email me on info atsign cytomic dotsign com. And also please let me know if you want any particular style of compression - heavy, light, limit etc, otherwise I'll just compress them to whatever sounds good. Also if you email please let me know any credits you want possibly including a url.

Andrew Simper

i will email you later when i get home

Subz

I usually wouldn't use SSL busscomps on vocals, as they're on the edgier side of compressors. It all depends on what sound you're looking for, of course. Could be interesting

[living sounds]

I usually wouldn't use SSL busscomps on vocals, as they're on the edgier side of compressors. It all depends on what sound you're looking for, of course. Could be interesting

Depends. I've used mine at least once on a background vocal track, where it worked very well.

[djsubject]

email sent

[Nokenoku]

@fjolle
What should all that useless Anti-Hacking-Stuff be for?
Basically he has 2 choices:
1. He uses a Syncrosoft Dongle.
2. His app will be cracked sooner or later no matter what other copy-protections he uses.

The protection should first be comfortable for the real customer. That's the most important in my opinion.

[bmanic]

Exactly. It's either syncrosoft or hacked. It really seems to be that simple.

So, the protection needs to make it impossible for ordinary users to "share" their plugin or abuse the demo. It doesn't make any sense trying to defeat the hackers.

Cheers!
bManic

[fjolle]

@fjolle
What should all that useless Anti-Hacking-Stuff be for?
Basically he has 2 choices:
1. He uses a Syncrosoft Dongle.
2. His app will be cracked sooner or later no matter what other copy-protections he uses.

The protection should first be comfortable for the real customer. That's the most important in my opinion.

Useless?

While it will be cracked after time, its still not a bad idea to make some sort of protection to ensure that it cannot simply be copied. If you eliminate the usual flaws that makes beginners able to crack it easily, it will usually take some serious popularity before the better crackers look into it. And then again, if you take just a little time to change the checks and algorithms each time, it will break their methods with each update. What i worked with was self modifying code that would decrypt the code sections needed at runtime. Everytime the protector was applied, the encryption was different, and the keys where stored randomly for then to be combined to the major key. Since self-modifying code and newer processors running with 64-bit windows wasn't really compatible anymore, we had to do it with code injection , which was a little messy and far less secure because it was easier to catch and work with. My partner and i ran into some problems later, also because we both started at university (studying software development, you guessed it). Nevertheless, its interesting to work with and you learn a lot, both from the dark and white sides of it.

You protect stuff to lengthen the time it stays on the market.

I do not want to discuss such matters with individuals who calls it "anti hacking" and "useless". I have worked more in this business than you have and ever will.

Depends. I've used mine at least once on a background vocal track, where it worked very well.

Hi!

Of course, sometimes it might just hit the spot
But i'm interested in the range knob's effect on this thing.


hmm..
This reminds me of another idea.. Don't tell if you discover that its a cracked version (check consistency of the file, or keep some of the serial checks silent), but wait a week or so and then just stop working throwing a "this is cracked" message. If its discovered that its cracked, be sure to set a flag in the file so if the cracker don't start with a clean file, he'll have even more problems. Also check your checks, its easy to read your own process and by doing that you can see if some instructions have been changed. Internal consistency checking is always a bugging thing

It's either syncrosoft or hacked. It really seems to be that simple.

So, the protection needs to make it impossible for ordinary users to "share" their plugin or abuse the demo. It doesn't make any sense trying to defeat the hackers.

Yes, you get the point. But I also talk about the thousands of newbie crackers out there. With just a little knowledge and time, you can make it impossible for them to crack it. Only the better will be able to do it, and then you already bought some valuable time on the market.

[poonna]

hmm..
This reminds me of another idea.. Don't tell if you discover that its a cracked version (check consistency of the file, or keep some of the serial checks silent), but wait a week or so and then just stop working throwing a "this is cracked" message. If its discovered that its cracked, be sure to set a flag in the file so if the cracker don't start with a clean file, he'll have even more problems. Also check your checks, its easy to read your own process and by doing that you can see if some instructions have been changed. Internal consistency checking is always a bugging thing

I think this is basically how Zebra does it -- delayed notification.

[fjolle]

hmm..
This reminds me of another idea.. Don't tell if you discover that its a cracked version (check consistency of the file, or keep some of the serial checks silent), but wait a week or so and then just stop working throwing a "this is cracked" message. If its discovered that its cracked, be sure to set a flag in the file so if the cracker don't start with a clean file, he'll have even more problems. Also check your checks, its easy to read your own process and by doing that you can see if some instructions have been changed. Internal consistency checking is always a bugging thing

I think this is basically how Zebra does it -- delayed notification.

Yeah i recall Zebra has this implemented

And its really annoying. You know how it is to feel "hah i got it!" when you have to realize later that you didn't get it. I've tried it with various crackme's i was working on (though your time delay should be longer). Simply the thing that you get the "correct serial" thing and then restart to still have the demo version without any notification is a bit "plah"

and before people start wondering, no i'm not a cracker. For several months i spent a large amount of time on anti-cracking algorithms and such, and of course i've been practicing a little. I've also helped a few smaller shareware authors with planning and implementing license systems, so i've spent my time being good

[Dean Aka Nekro]

im an iLok user fwiw, but i will use whatever happily if i like the software enough to know i will use it and of course buy a license (im pretty simple in that regard)

Nekro/Dean

[aud.io]

I do not want to discuss such matters with individuals who calls it "anti hacking" and "useless". I have worked more in this business than you have and ever will.

Fjolle, I'm sure you're not the only one with a little security knowledge in here. (BTDTGTTAWIO ... Rob Northen copylock anyone?) None of your "ideas" sound new or uncommon at all though. I am sure that a developers energy is always better spent on good support and aiming at a fair pricing policy than on implementing some quirky cracking countermeasures.

[fjolle]

I do not want to discuss such matters with individuals who calls it "anti hacking" and "useless". I have worked more in this business than you have and ever will.

Fjolle, I'm sure you're not the only one with a little security knowledge in here. (BTDTGTTAWIO ... Rob Northen copylock anyone?) None of your "ideas" sound new or uncommon at all though. I am sure that a developers energy is always better spent on good support and aiming at a fair pricing policy than on implementing some quirky cracking countermeasures.

These certainly aint my "ideas", nor are they new, but they're simple to implement and rather effective. Just the fact that you perceived it like i thought this was "new" stuff, shows your amount of intelligence.

I took the time to read through your posts here at KVR at least, and you don't seem like.. eh.. nevermind

I never wanted to discuss these issues here, just give my input to why he should at least spend a little time on it. Why on earth would i sit here writing of my ideas here? Was that a joke? Besides, I don't think the companies (who by the way, is the reason i can live like i do and have a decent studio while being a student) would like me to tell any ideas of what i'm working with. Not that it would have any interest to the majority, its still something they don't like. Security against crackers is of their least concern, and true enough that was just a hobby/side project i did. But reveal anything of it, i did not.

Have a nice day

[andy-cytomic]

Everyone, thanks for you input on security matters. Let's just leave it there please and keep on topic.

Andrew Simper

[l0calh05t]

Why is there no 0.1 release setting like on the hardware unit? I also thought I was hearing even faster release than that on your examples.

Trust your ears. I'm not sure why they labelled it as 0.1 ms. Analog gear sometimes isn't all that precise, but the following numbers are based on a 2.2 multiplier of the time constants Tor = (R*C):

= release ssl =
label {0.1, 0.3, 0.6, 1.2}
actual {0.18612, 0.27918, 0.57904, 1.2408}

= release cytomic =
label {0.2, 0.3, 0.4, 0.6, 0.8, 1.2}
actual {0.18612, 0.27918, 0.39292, 0.57904, 0.8272, 1.2408}

So as you can now see, but already heard, it's the same on the fastest setting.

Andrew Simper

Isn't the fastest release in the SSL bus comp defined by a 100kOhm resistor and a 0.47uF (the gyraf bus comp, and same for the 4000E schematic)?

10e5 * 47 * 10e-8 * 2.2 = 47 * 2.2 * 10e-3 = 0.1034 seems to be pretty exactly 100ms...

EDIT: Eh sorry, was just the crappy font/low resolution of the schematics... the resistor is a 180k resistor, but I'd like to know where the 2.2 multiplier comes from

[andy-cytomic]

... but I'd like to know where the 2.2 multiplier comes from

If you go through all the other times, it appears to be the closest multiplier to match them. The multiplier of the time constant is a subjective thing to give the attack and release time, different manufacturers use different scalings. There are quite a few other factors that effect the time, including other filtering in the sidechain, and on the main amp, as well as the fact that for attacks the resistor charging up the cap through the diode is quite a different thing to a linear RC filter.

Andrew Simper