The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Any ideas how to deal with inter sample peaks with Slate FXG? Dynamics Plugins
Old 28th May 2012
  #1
Lives for gear
Any ideas how to deal with inter sample peaks with Slate FXG?

Hi there i just couldn't control myself with the current 48 hr special at audiodeluxe, 99 bux for a $250 plugin, often raved about for it's ability to do LOUD especially in modern club genres...

so i grabbed it...the demo was an older version that had a high fixed ceiling and it always had isps but that was known of the earlier build and i thought it had been improved :(


I have no issue with the sound whatsoever.. like most current crop limiters.. doesn't work on everything, but, sounds great on alot of stuff! the problem is, even with a ceiling of -0.3 db and just a couple of db gain reduction, i am getting inter sample peaks all over the place. the true peak meter (3rd party) is showing it at + 2db!! Bouncing to wav or aiff is fine but bouncing to mp3 had about 147 clips using wavelab analysis. This was only a 4 bar section!

I just wonder if anyone has gotten around this in any way?

I even turned the "oomph" sliders to zero... i.e the "extra special" stuff was not going on.. ok this reduced the peaks to only a half a db over but it was of course no where *near* as loud.

This is where it gets weird(er)..

i lowered threshold to -1 yes as in -1.0 which is obviously way lower (or higher lol) than any of us would use (i have always been of the school of -0.3 for my music, it's just what i was taught i guess), and even at a ceiling of -1.0....

.....YEP.. isp's !!

I haven't seen this many isp's on any of the crop of current limiters or even the much older sonnox. Don't know what to do.. I have NEVER prepared my music in the past for the "compressed junkies media" ie itunes or mp3, this is new territory for me.. and have never been so concerned about isp's before..But i am now conforming to the masses.. hehe... But... i can't bounce to mp3 without clipping.. as in real clipping not just intersample..

could really use some help.

I was going to buy plug in alliance vertigo and the slate has probably saved me $200 because the compressor section itself is great and just as good as the vertigo.... so it is WAY worth it for the compressor alone, and that's how i will look at it..still a very worthwhile purchase.. but, if there are any tips with the loudness maximizing section,

I am ALL ears, and thank you all in advance!

cheers!

p.s if this is just the way it is and i am using it right, and basically, if you want loud with FGX, there is no way to do it without isp's, then i'd appreciate that info also!! So far it measured the highest isp's of any ITB limiter i have ever used, ever.
Old 28th May 2012
  #2
Lives for gear
 
norbury brook's Avatar
 

As far as I'm aware when you're bouncing to mp3 there is no way of dealing with inter sample peaks,it's a problem with the mp3 algorithm and conversion.

I remember Justin frankel the reaper Developer explaining this to reaper users a few years ago.


MC
Old 28th May 2012
  #3
Lives for gear
 
valis's Avatar
Fwiw one of the main features of FGX seems to be soft clipping specifically so you can raise the overall program level, so I would warrant that reconstruction meters aren't going to tell the whole truth.

The MP3 format uses 32bit FLT values to recall not pcm data, but 'frames' from a lookup table that are 1. of a given size (specified by the encoder for CBR and via the preset values for VBR/Lame) and 2. vary in length. So what happens upon converstion from an input format is not 100% predictable, and I have no idea how saturated peaks are represented if they are hitting full scale.

I would bounce to 16bit INT with the usual dither that you'd use, and bring that into an mp3 encoder externally. I've had more reliable results this way, as you can check the 16bit (or 24bit) waveform before and after mp3 conversion. If nothing else this will at least tell you what is encoded into the PCM output stream and what the mp3 encoder is reacting to.
Old 28th May 2012
  #4
Slate Pro Audio / Slate Digital
 
Steven Slate's Avatar
 

Hi there. What is your concern about intersample peaks? My only concern about them would be if I were to use a CD Player made prior to about 1995.

Other than that, it's not an issue in modern music, especially now with the MP3 and AAC format being the primary music listening platform.

In the FG-X 2.0 we'll have an option to defeat intersample peaks, but if you don't remember to turn it on, life will go on.

I'm not intending to disrespect or diminish your questions, but I just wanted to point out that isps are not something that are destructive in modern music.

Cheers,
Steven
Old 29th May 2012
  #5
Lives for gear
Hi thanks Steven and no disrespect to you either,

I am very happy with my purchase overall, the compressor is brilliantly transparent, exactly what i needed!

My equipment is pretty old school and i am sure many others have old school cd players etc?

I will not pretend i am some expert on intersample peaks..

but you said it doesn't matter with mp3 but with mp3 is when i get hard clipping when i bounce.

is the best solution in this matter to simply lower the master fader to say -1db when i bounce to mp3 to avoid any clipping? Does this solve the problem?

Thanks for the other answers i will definitely look into them regarding dither and mp3 conversion.

So far cubase seems to have the best mp3 bounce i have seen, so much that i started to import my finished track from logic into cubase just to use the mp3 bounce!

Whatever the meter tells me is what Cubase bounces too.. 100% accurate.

In any case

the slate DOES allow alot of isp through so i look forward to version 2,

just to have the option, can't hurt!

Now remember even if i don't have HARD clipping and i bounce my track to aiff or wav, if i throw the meter on i have tons of ISP clippin

I use various meters to suggest this, ssl, k, and now the steinberg one

also studio one has an isp meter built in that seems to be even more sensitive than the others!

What IS interesting,

if the studio one ISP meter shows ZERO isps, then even an MP3 bounce has ZERO clips...

if mp3 is impossible to avoid clipping, then i don't understand the above, i have tested it time and time again and the meter is rock solid accurate
Old 29th May 2012
  #6
Lives for gear
 
Beatworld's Avatar
Quote:
Originally Posted by Steven Slate View Post
the MP3 and AAC format being the primary music listening platform.
I get mp3 being a great format when we had 28.8k modems and MB hard drives.
I don't get it now we have hi-speed adsl2+ or faster and TB hard drives.

...back on topic....I feel more comfortable using FF Pro-L with the ISP meter on and knowing I'm not getting any ISPs but still hitting a good level.
Old 29th May 2012
  #7
Lives for gear
 
m_gant's Avatar
 

I always bounce a processed squashed (to whatever degree necessary) 16bit render first. Then I open that file in sound forge & render my mp3.
Old 21st June 2012
  #8
Gear Head
 

Just came across these issues with FGX as well. And the odd thing is, it was very hard for me to hear them coming out of my Cranesong HEDD, but the clients could clearly hear the ISP's through his 003 and on a laptop. I guess that's what superior D/A converters do, mask any errors or distortion.

I got that free SSL ISP meter, which told me they were there. Using FF Pro L with 4x oversampling and ISP on has eliminated them, as far as I can tell. Odd that this has never come up, but FGX is clearly adding something bad here...can't wait for an update to eliminate this.

FWIW, I can hear the same things happening with DSM, so it's just some byproduct of the soft-clipping process.
Old 21st June 2012
  #9
Lives for gear
 
mirrorboy's Avatar
 

I've noticed this for a while and have a work-around....it's nothing great and I'd like to try something else.

Anyways, I just put the Massey L2007 limiter last in the chain with the threshold off and the output max set to -.3dB.

It works (you will get NO inter-sample peaks) but the problem is I still do get about .5dB to 1.5dB of GR from the L2007 which does change the sound of the mix slightly (but noticeably).

Obviously, I'd rather not have to put it there.

Scott
Old 21st June 2012
  #10
Slate Pro Audio / Slate Digital
 
Steven Slate's Avatar
 

Quote:
Originally Posted by Jimmy B View Post
Just came across these issues with FGX as well. And the odd thing is, it was very hard for me to hear them coming out of my Cranesong HEDD, but the clients could clearly hear the ISP's through his 003 and on a laptop. I guess that's what superior D/A converters do, mask any errors or distortion.

I got that free SSL ISP meter, which told me they were there. Using FF Pro L with 4x oversampling and ISP on has eliminated them, as far as I can tell. Odd that this has never come up, but FGX is clearly adding something bad here...can't wait for an update to eliminate this.

FWIW, I can hear the same things happening with DSM, so it's just some byproduct of the soft-clipping process.
Jimmy can you send me a snippet of this file? An 003 or laptop or any D/A made past the mid 90's should not have any audible ISP's and I've never heard one with my own ears produced by the FG-X, so I suspect what your client is hearing is something else. Please send to [email protected]

My suggestion to those who are using the FG-X are to use your ears and do not worry about an ISP meter. You'd be shocked to realize how much of your music collection has ISPs and you never even realized it.

While the next update of FG-X will have an ISP blocker, it's not something to worry about as D/A's have handled them fine for over a decade now. I can say for sure that the MacBooks have no issues at all with them, as I listen to FG-X mastered tracks out of it all day long.

Steven
Old 21st June 2012
  #11
Gear Head
 

I even removed all of my analog hardware ,and simply used a BX digital EQ and FGX - the client STILL heard it and complained of distortion.

His levels were low enough, so I have no clue what's going on. The second I removed FGX and put FF Pro L on, he said it sounded great.

When pressed about it, he said both of those sections used Massey Tape Head on the bass guitar for some grit - each about 30-50 seconds long. I don't see how this could really make a difference, but when I analyzed my entire session in Samplitude with the SSL ISP Meter, it showed peaks all around - I've never been able to hear them through my Cranesong HEDD since buying FGX last year, but I think these have been on all the masters since starting to use this...which kinda worries me. But I suppose if it wasn't severe and no one else heard it, is it really there...?

I'm gonna analyze a few other sessions with the SSL ISP plug and see what's up. But for the time being (no massey on the Windows VST world) Pro L seems to clear it all up for me.
Old 22nd June 2012
  #12
Slate Pro Audio / Slate Digital
 
Steven Slate's Avatar
 

Hi Jimmy, did you hear any distortion on the FG-X master? Are you certain it isn't just settings issue with the FG-X? Where was your ITP slider? Can you send me a snippet to examine? [email protected]

steven
Old 22nd June 2012
  #13
Slate Pro Audio / Slate Digital
 
Steven Slate's Avatar
 

Here you can check out a song before and after FG-X mastering. This track has played everywhere including radio, tons of laptops, CD player, car... without issues, as have many major hits mastered with the FG-X by Howie Weinberg, Ted Jensen, etc.. Again, ISP's are not an issue with D/A's of the last decade, probably even more.

If anyone has problems with the FG-X, I'd be more than happy to personally assist. It can be very sensitive to settings, which is why the new update has my presets which give you a great starting point for effective use.

------------------------------------------------------------------------


This 16bit 44.1khz mix has a lot of headroom left for mastering. Pay careful attention to the punch, dynamics, and frequency response of the mix, which we'll want to preserve in the mastering stage.

MIX LINK:
www.slatedigital.com/fgx/WastedMix.wav

Here is the same mix mastered with the Slate Digital FG-X using the Steven Slate "Mastering Hard" preset. The compressor threshold was set so that about 1db of compression was being applied on peaks, and the GAIN knob was increased 9db!! Despite this huge amount of added gain, the punch and dynamics have been preserved and the master still has the same depth and life of the original mix.

MASTER LINK (TURN YOUR MONITORS DOWN):
www.slatedigital.com/fgx/WastedMaster.wav
Old 22nd June 2012
  #14
Lives for gear
 
mirrorboy's Avatar
 

Quote:
Originally Posted by Steven Slate View Post
Here you can check out a song before and after FG-X mastering. This track has played everywhere including radio, tons of laptops, CD player, car... without issues, as have many major hits mastered with the FG-X by Howie Weinberg, Ted Jensen, etc.. Again, ISP's are not an issue with D/A's of the last decade, probably even more.

If anyone has problems with the FG-X, I'd be more than happy to personally assist. It can be very sensitive to settings, which is why the new update has my presets which give you a great starting point for effective use.

------------------------------------------------------------------------


This 16bit 44.1khz mix has a lot of headroom left for mastering. Pay careful attention to the punch, dynamics, and frequency response of the mix, which we'll want to preserve in the mastering stage.

MIX LINK:
www.slatedigital.com/fgx/WastedMix.wav

Here is the same mix mastered with the Slate Digital FG-X using the Steven Slate "Mastering Hard" preset. The compressor threshold was set so that about 1db of compression was being applied on peaks, and the GAIN knob was increased 9db!! Despite this huge amount of added gain, the punch and dynamics have been preserved and the master still has the same depth and life of the original mix.

MASTER LINK (TURN YOUR MONITORS DOWN):
www.slatedigital.com/fgx/WastedMaster.wav
Awesome!!

The master sounded great and the mix was translated very well!! Listened to both a couple times.

Would you mind posting the settings you used on the FG-X to master that (if u have it saved)? And anything else in that chain (order too (....

Again, sounds really, really great!!!


Also, after reading your post, I remember a good engineer/producer friend of mine (Chris Manning/SF) telling me that the 192's had ~ 6dB of headroom.

So this would make sense and I think I can safely remove the L2007. I'm using an Aurora 16.

Thanks again Steve & looking fwd to V.2!

Thx.

Scott
Old 22nd June 2012
  #15
Lives for gear
 
Alexey Lukin's Avatar
 

Quote:
Originally Posted by Steven Slate View Post
This 16bit 44.1khz mix has a lot of headroom left for mastering. Pay careful attention to the punch, dynamics, and frequency response of the mix, which we'll want to preserve in the mastering stage.

MIX LINK:
www.slatedigital.com/fgx/WastedMix.wav
Looks like it has some digital transfer errors: a minor crackle above 10k that is not present in the mastered variant.


Quote:
Originally Posted by Steven Slate View Post
MASTER LINK (TURN YOUR MONITORS DOWN):
www.slatedigital.com/fgx/WastedMaster.wav
ISP are not an issue here because the peaks only reach -2 dBFS
But I tend to agree with Steven: intersample peaks are usually not as bad as they are sometimes presented. However modern loudness standards (like BS / EBU) require keeping these "true peaks" in control.
Old 22nd June 2012
  #16
Slate Pro Audio / Slate Digital
 
Steven Slate's Avatar
 

Quote:
Originally Posted by mirrorboy View Post
Awesome!!

The master sounded great and the mix was translated very well!! Listened to both a couple times.

Would you mind posting the settings you used on the FG-X to master that (if u have it saved)? And anything else in that chain (order too (....

Again, sounds really, really great!!!


Also, after reading your post, I remember a good engineer/producer friend of mine (Chris Manning/SF) telling me that the 192's had ~ 6dB of headroom.

So this would make sense and I think I can safely remove the L2007. I'm using an Aurora 16.

Thanks again Steve & looking fwd to V.2!

Thx.

Scott
Hi Scott thanks! I used the Steven Slate Preset "Mastering Hard" and I think I gave it a tad more lo punch in the transient section to keep those kicks sound big. I reduced the threshold on the compressor so that the GR was about 1db. I've been using the FG-X for longer than anyone on earth but it still amazes me how loud I can get my masters without ruining the impact.

Cheers,
Steven
Old 23rd June 2012
  #17
Lives for gear
 
Beatworld's Avatar
Quote:
Originally Posted by Alexey Lukin View Post
ISP are not an issue here because the peaks only reach -2 dBFS
The devil is in the detail.

For the last week or more I've been doing a serious limiter shoot out.
I've got a couple of good EBU compliant loudness meters at the end of the chain.
These measure "true peak" readings.

It is very interesting watching which limiters exceed their output threshold with ISPs.

Oversampling certainly goes a long way to stopping ISPs.
Old 23rd June 2012
  #18
Lives for gear
 
popmann's Avatar
Very interesting...so...these are things you see but don't hear? If you hear them, whats the manifestation? What's the tell?

I hear "distortion" with every limiter...the more "effective", the more distortion. How do you differentiate the audible ISP distortion from plain old clipper distortion? Or plain old "now everything's jammed up and the low level distortion already there is more audible" distortion?

Or is this just a geek meter thing?
Old 24th June 2012
  #19
Lives for gear
 
Beatworld's Avatar
Quote:
Originally Posted by popmann View Post
Or is this just a geek meter thing?
Bottom line: I imagine many would say it is a geek meter thing.

But: Depends on what device you are listening to the playback on.
Most modern CD players can handle peaks greater than 0db apparently but some older CD players can't and you would hear clipper distortion.
But: Depending on the extent the peaks exceed 0db threshold you may experience problems down stream if converting to mp3 or some other codec for digital distribution. "Mastered for iTunes" Tips suggests thresholds at -1.0db.
But: Geek peace of mind knowing your output level is as you set it.
Old 24th June 2012
  #20
Lives for gear
 
mirrorboy's Avatar
 

Quote:
Originally Posted by Beatworld View Post
Bottom line: I imagine many would say it is a geek meter thing.

But: Depends on what device you are listening to the playback on.
Most modern CD players can handle peaks greater than 0db apparently but some older CD players can't and you would hear clipper distortion.
But: Depending on the extent the peaks exceed 0db threshold you may experience problems down stream if converting to mp3 or some other codec for digital distribution. "Mastered for iTunes" Tips suggests thresholds at -1.0db.
But: Geek peace of mind knowing your output level is as you set it.
Haha ya that's totally what it is for me:

"Geek peace of mind!" haha love that

Scott
Old 7th September 2016
  #21
Here for the gear
 

I think the main concern, at least for me, is whether or not my mastered music will be rejected from iTunes. In order for a track to be Mastered for iTunes, there needs to not be any inter-sample clipping. I've been using the FG-X with great results to the ear, but digitally, it's clipping at the isp level. A good way to see this is by using AU Lab's RoundTripAAC plugin. It's free download from Apple. Anyway, I'm about to give it a go by lowering the ceiling on the FG-X to -1dBFS, but if that doesn't work, I heard that Ozone's Maximizer does the trick. Fingers crossed.
Old 8th September 2016
  #22
Slate Pro Audio / Slate Digital
 
Steven Slate's Avatar
 

Hi!

Mastered for iTunes won't reject any files made by FG-X nor will it even reject files that have ISP's... However, to act in good form, for mastered for iTunes, use FG-X with the ceiling set to -1.

Cheers,
Steven
Old 8th September 2016
  #23
Lives for gear
Quote:
Originally Posted by Steven Slate View Post
Hi!

Mastered for iTunes won't reject any files made by FG-X nor will it even reject files that have ISP's... However, to act in good form, for mastered for iTunes, use FG-X with the ceiling set to -1.

Cheers,
Steven
Does that guarantee no ISPs Steven?
Old 8th September 2016
  #24
Lives for gear
oh, necro bump!

I use ozone 7 mostly these days, as i find it far superior to fg-x (as is ik stealth, dmg limitless, and a few others).

however i will definitely check out the (fabled?) version 2 when it comes... is the delay because so many strong competitors have come out since the announcement, that you now have to take it up a notch to even compete?
Old 8th September 2016
  #25
Slate Pro Audio / Slate Digital
 
Steven Slate's Avatar
 

Quote:
Originally Posted by dontsimon View Post
Does that guarantee no ISPs Steven?
Probably. But ISP's don't affect any playback systems so it doesn't really matter anyway. ISP's are the boogie men of digital audio.

Quote:
Originally Posted by TNM View Post
oh, necro bump!

I use ozone 7 mostly these days, as i find it far superior to fg-x (as is ik stealth, dmg limitless, and a few others).

however i will definitely check out the (fabled?) version 2 when it comes... is the delay because so many strong competitors have come out since the announcement, that you now have to take it up a notch to even compete?
Hi! If you need help with FG-X settings I'd be glad to assist. I'm confident it can hold it's own to the competitors you mentioned and keep mixes more dynamic sounding!

Thanks for the discussion!

Cheers,
Steven
Old 18th September 2016
  #26
Lives for gear
Quote:
Originally Posted by Steven Slate View Post
Probably. But ISP's don't affect any playback systems so it doesn't really matter anyway. ISP's are the boogie men of digital audio.



Hi! If you need help with FG-X settings I'd be glad to assist. I'm confident it can hold it's own to the competitors you mentioned and keep mixes more dynamic sounding!

Thanks for the discussion!

Cheers,
Steven

i don't know why you keep saying this.. Personally I just don't' feel FGX can compete with the likes of limitless and O7.. not only that, there are hundreds of posts around various forums saying the same..

So, I think the better test would be to send in some content that is pre master, we can set a pre determined target, say -1 ceiling max, and 4db of GR max? Then I post ozone 7 versions, someone post limitless versions, NO OTHER treatment, just the limiter and how it preserves the source, then YOU post fg-x versions? That way, you can't say that we aren't getting the best out of fg-x.

Now, if you match or even exceed the others in sound quality, then you have been right all along. It also means, that since so many of us would therefore be wrong, that you obviously would need to make your product more user friendly, cause so many people seem to be unable to get the best out of it to match the current limiters. Surely that would point to a flaw in the product design. PS the compressor is not allowed in these tests, again, just the limiter. Would you be up for this Steven?
Old 19th September 2016
  #27
Gear Nut
 

Quote:
Originally Posted by TNM View Post
i don't know why you keep saying this.. Personally I just don't' feel FGX can compete with the likes of limitless and O7.. not only that, there are hundreds of posts around various forums saying the same..

So, I think the better test would be to send in some content that is pre master, we can set a pre determined target, say -1 ceiling max, and 4db of GR max? Then I post ozone 7 versions, someone post limitless versions, NO OTHER treatment, just the limiter and how it preserves the source, then YOU post fg-x versions? That way, you can't say that we aren't getting the best out of fg-x.

Now, if you match or even exceed the others in sound quality, then you have been right all along. It also means, that since so many of us would therefore be wrong, that you obviously would need to make your product more user friendly, cause so many people seem to be unable to get the best out of it to match the current limiters. Surely that would point to a flaw in the product design. PS the compressor is not allowed in these tests, again, just the limiter. Would you be up for this Steven?

Well if you want to make this a fair test, then listening should be done blind so that people don't fall victim to nasty old expectation bias, ie: simply picking the limiter they've already decided they like best.

That would make it a fair test; and if FG-X really can't compete as you say, then it should be easy to pick out in a blind listening test, right?


And personally when I'm sending ref mixes out to clients, often I'll use FG-X, often it'll be Pro-L - both are awesome tools, I pick whichever works/I like at the time and go about my day.
Old 19th September 2016
  #28
Lives for gear
Quote:
Originally Posted by SpicyPonyHead View Post
Well if you want to make this a fair test, then listening should be done blind so that people don't fall victim to nasty old expectation bias, ie: simply picking the limiter they've already decided they like best.

That would make it a fair test; and if FG-X really can't compete as you say, then it should be easy to pick out in a blind listening test, right?


And personally when I'm sending ref mixes out to clients, often I'll use FG-X, often it'll be Pro-L - both are awesome tools, I pick whichever works/I like at the time and go about my day.
ok fair enough post mate.. but it does go to show how different we are.. i don't consider pro L even 'good' these days. I even prefer the old sonnox to it TBH..

I guess i am all about preservation of source.. pro L in all modes changes the bottom end, even with half a db reduction. But i digress.. back to your point sure, blind test, as long as Steven agrees to the parameters of course (or suggests others we can agree to). The main point here is that HE is the one using the FGX as no one could use it better.. the aim here is to hear FGX in it's best possible light.

I've been searching through my stuff, i'd rather Steven submit a track that he is legally allowed to do so, if that's ok.
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…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
Jonny5 / Music Computers
14
Jake 2.0 / Drums
24
elarreal / Electronic Music Instruments and Electronic Music Production
10
AlexLakis / So much gear, so little time
24

Forum Jump
Forum Jump