ValhallaVintageVerb 1.7.1. Two new reverb modes (Chaotic Hall, Chaotic Chamber)

[Sampleconstruct]

I only upload wavs, gives me the best result. Converting an MP3 to an MP3 just doesn't cut it, and there is no way around the SC encoder.

Btw - just encountered the ominous noise burst again, thank god there is now a limiter built in that cuts the signal at +12 dB, was still loud as hell though....

[t3toooo]

i heard the best results come from uploading wav files, can anybody confirm?

or is it better to upload ur own crappy 128kps copy?(which would be presumably better than their even crappier encoding)

not to turn it into a soundcloud topic,although it is rather a "concern".

when uploading a "to hot" wav file even without clipping the soundcloud limiter definitely takes place in the conversion process.

a 16 bit wav file was just horrible for some reason.
the 32 bit was ok,maybe to loud,i should lower them -3db but i already made a test with a -3db 16 wav file.

the best thing imo is to convert the file to 128kb mp3 without clipping.
that's maybe a problem,since lame tend to spike more and then the music is lower.
otherwise i did not test it properly.

soundcloud seems to leave a non clipping 128kb file untouched.

read my post and listen if you get not too much bored,i will delete the other two tests in the next days.

[antithesist]

Yes, it's been said that 128 MP3s don't get re-encoded for soundclown streaming. I've also heard that's not the case. It's pretty common knowledge that most lossy encoders produce better results from 24 bit files. Also, keep your files to -1 dB FS. Others add to have a half second of silence at the beginning and end. I've found these three tips to be better than the totality of "Mastering for itunes."

Brawndo: "It's got Electrolytes."

Speaking of dB FS and back on-topic, what is +12 dB?

[egbert]

Speaking of dB FS and back on-topic, what is +12 dB?

That limiter is operating on a 32 bit float signal. At that level, if you bounced it to a wav file at unity gain it would have the top 12dB clipped off it.

[t3toooo]

with untouched i mean a non clipping 128kb is probably encoded again but the process will not produce any clipping which is exactly activating the "nice limiter".

[antithesist]

Speaking of dB FS and back on-topic, what is +12 dB?

That limiter is operating on a 32 bit float signal. At that level, if you bounced it to a wav file at unity gain it would have the top 12dB clipped off it.

Ah, I get it now, thanks. Still mighty loud I'd imagine and as Sampleconstruct noted, but I guess it might save a tweeter or two without affecting normal operation. Of course, +11 would be more better.

[Sampleconstruct]

Speaking of dB FS and back on-topic, what is +12 dB?

That limiter is operating on a 32 bit float signal. At that level, if you bounced it to a wav file at unity gain it would have the top 12dB clipped off it.

Ah, I get it now, thanks. Still mighty loud I'd imagine and as Sampleconstruct noted, but I guess it might save a tweeter or two without affecting normal operation. Of course, +11 would be more better.

The Master Outputs still showed +6dB but the burst this time wasn't as loud as the ones in the past, so I assumed it had to do with the internal limiting, but maybe I'm wrong and the Limiter didn't do what it was supposed to do in a situation like this. My ADAMs survived it and the Sub had it's usual noise-burst-orgasm.

[valhallasound]

Yes, it's been said that 128 MP3s don't get re-encoded for soundclown streaming. I've also heard that's not the case. It's pretty common knowledge that most lossy encoders produce better results from 24 bit files. Also, keep your files to -1 dB FS. Others add to have a half second of silence at the beginning and end. I've found these three tips to be better than the totality of "Mastering for itunes."

No matter what input file you provide, a 128 kbps MP3 is going to destroy the spatial information for the higher frequencies. This would work OK for some reverb algorithms like the Quantec QRS, where the stereo field is created by mixing together different inversions of the output taps (i.e. no delay differences between left and right, just sign differences). For algorithms that have different delays going left and right, 128 kbps MP3 will make a hash of things. I wish I had known that before I ponied up the cash for the SoundCloud account.

Sean Costello

[valhallasound]

Here's a comparison that serves two functions:

- Comparison of Soundcloud with a 192 kbps MP3
- Hawking my own wares (you can find me down in the market, in the booth with the "Reverbmonger" sign. Try the Sanctuary - it's fresh!)

Here is one of my VintageVerb demos on Soundcloud:

http://soundcloud.com/seancostello/moog ... orus-space

The same file, converted from 16 bit / 44.1 kHz .wav into a 192 kbps MP3 using iTunes:

http://www.valhalladsp.com/sounds/MoogR ... 92kbps.mp3

Sean Costello

[justin3am]

Hey Sean,

When I upload a .wav file to my Soundcloud account and make it available for download, the streaming file gets compressed but the download file does not. That is to say, you can download tracks as .wav when they are uploaded in that format. I'm not sure if this feature is unique to the pricier soundcloud accounts but it's very useful.

Forgive me if you were aware of this already.

[valhallasound]

Hey Sean,

When I upload a .wav file to my Soundcloud account and make it available for download, the streaming file gets compressed but the download file does not. That is to say, you can download tracks as .wav when they are uploaded in that format. I'm not sure if this feature is unique to the pricier soundcloud accounts but it's very useful.

Forgive me if you were aware of this already.

I had heard of this, but haven't used it before - thanks for pointing this out.

It looks like you can download the original .wav from that Soundcloud link, if anyone wants to make a comparison between the source, 192 kbps MP3, and the streaming 128 kbps MP3.

Sean Costello

[Beatworld]

The same file, converted from 16 bit / 44.1 kHz .wav into a 192 kbps MP3 using iTunes:

http://www.valhalladsp.com/sounds/MoogR ... 92kbps.mp3

Superior in every way.....more open, wider, smoother decay, better example of the Reverbmonger's wares.

Why stop at 192kbps ?
Go the full distance with 320kbps

[valhallasound]

The same file, converted from 16 bit / 44.1 kHz .wav into a 192 kbps MP3 using iTunes:

http://www.valhalladsp.com/sounds/MoogR ... 92kbps.mp3

Superior in every way.....more open, wider, smoother decay, better example of the Reverbmonger's wares.

Why stop at 192kbps ?
Go the full distance with 320kbps

Meanwhile, I put 10/12 bits of floating point quantization into the Sanctuary mode, and it sounds SUPER wide and deep. It is weird to think about how shaving off a few bits in one context can sound good, while in the context of psychoacoustic compression it results in noticeable degradation. My guess is that the floating point bit reduction in Sanctuary (which is probably the best name for the process, although it isn't exactly how it was implemented) emphasizes the differences between the left and right output channels, while the 128 kpbs "joint" stereo will reduce left/right differences.

Sean Costello

[Beatworld]

It looks like you can download the original .wav from that Soundcloud link, if anyone wants to make a comparison between the source, 192 kbps MP3, and the streaming 128 kbps MP3.

Sean Costello

Downloaded the source......(of course) it is the winner
Beautiful.

I thought that you had uploaded a 192kbps file to SoundCloud which was then streaming at 128kbps.
More coffee perhaps....

[Beatworld]

My guess is that the floating point bit reduction in Sanctuary (which is probably the best name for the process, although it isn't exactly how it was implemented) emphasizes the differences between the left and right output channels, while the 128 kpbs "joint" stereo will reduce left/right differences.

Sean Costello

If you haven't already done so Sean, I highly recommend at least trying the demo of the Sonnox Pro-Codec.
You can audition every quality mp3 and AAC in real time and hear exactly how your source is going to sound when converted to lossy codecs.

http://www.sonnoxplugins.com/pub/plugin ... -codec.htm