Digital and Analogue outs conversion

[Music Manic]

Ok so you are mixing in STM, with your tracks coming from Cubase.

Cubase works in 32bit floating point as does the STM mixer.Is that correct?

ok now our Master out is sent to the analogue out does it get convertef into 16bit Integer and is it dithered automatically?

Thanks for your help.

[garyb]

no, the Scope mixer is(as i remember) 48bit integer. the asio driver is 24 or 32bit float. going out the DA, the bit depth is whatever the DA is, 16, 18, 20 or 24bit.

no dither is required, but you can dither all you like if you think it helps.

[Music Manic]

no, the Scope mixer is(as i remember) 48bit integer. the asio driver is 24 or 32bit float. going out the DA, the bit depth is whatever the DA is, 16, 18, 20 or 24bit.

no dither is required, but you can dither all you like if you think it helps.

Thanks Gary.

Well I have just read that 48bit Integer is superior than 32 float so that may explain the difference in quality compared with Vst plugins.

Could you expand to how the signal is "transfered" to 48bit?
Also how is the signal processed when it outputs into the RCA phono plugs? If it isn't dithered won't that affect low level noise.

One more question: Is the Vdat floating point or Integer.


Just interested in the Software domain at the moment.

Thanks

[garyb]

VDAT is integer.

i don't know the details of the bitdepth conversion in the mixer, as i understand it, it's a matter of resolution, like the dots on a newspaper picture. higher bitdepth = smaller dots. this matters most to high frequencies. there may be dither, there may not be. S/C might answer that. i don't think it matter so much for data that remains in the digital realm. i think it's more important in the conversion back to analog, though i'm likely to be corrected.

afaik, the analog outs are 24bit, though i believe they were 18bit on the pulsar 1.

[Shroomz~>]

Internally, SFP operates at a 186dB dynamic range (32 bit). That includes the mixers (STM series etc) which also operate at 186dB dynamic range.

[Music Manic]

Internally, SFP operates at a 186dB dynamic range (32 bit). That includes the mixers (STM series etc) which also operate at 186dB dynamic range.

Now I'm confused.
Who is correct?

Where do you guys get your info from?

Thanks

[garyb]

i got mine from Frank Hund(and other sources). i can always be wrong. so can he. cantact S/C for further info. tell us what you discover.

[Music Manic]

Hi Gary,

Just got the answer here:

http://www.sonic-core.net/en/products/platform.html

"It always processes the audio with a 32-bit resolution, and algorithms and parameters are computed with a floating-point resolution of even 40 bits"

[Shroomz~>]

It's mentioned in various documentation including the STM 24/48 mixer manual (look in the 'signal flow' section).

Audio in Scope is internally processed in 32-bit integer & there's no 48-bit mixers. The STM mixers as with the rest of the audio processing have a dynamic range of 186 dB.

[Music Manic]

It's mentioned in various documentation including the STM 24/48 mixer manual (look in the 'signal flow' section).

Audio in Scope is internally processed in 32-bit integer & there's no 48-bit mixers. The STM mixers as with the rest of the audio processing have a dynamic range of 186 dB.

Yes I'm just reading it Shroomz. I couldn't see anything about 32bit integer.

Thanks

[Shroomz~>]

No, the STM manual only refers to it as 32-bit, but it is indeed integer. No doubt about it.

[Shroomz~>]

31 bits @ 6 dB per bit.

[Music Manic]

Ah! It sounds like the mixer converts the Audio signal to Integer so that it has a fixed number the processes with 32bit Float when a plugin is inserted.

I would it would have been floating point all the way then Integer at the end.

Hard game to understand completely.

Thanks for help

[garyb]

there are devices that are higher resolution internally, or so i've been told, and one was the stm mixers. 32bit integer resolution is correct, otherwise. i'm happy to be corrected if there is official word....it certainly wouldn't be the first time i was incorrect.

[Shroomz~>]

There's 32-bit & 40-bit floating point math at play, but mostly (and I mean MOSTLY), audio in Scope is actually processed in 32 bit integer. Maybe the 48-bit rumor originates from the fact that the small amount of on-chip Sharc memory can handle 48-bit operations.

[garyb]

yes, the Scope environment is 32bit, i know. that's always been in the lit.

the 48bit thing isn't a rumor, except that it comes from me. as i said, when i got the info from several horses mouths, i could have misunderstood or the horses could have been just wrong, but i didn't just hop up and speak based upon my guess. i don't have enough info about the internals to do that. if the matter is seriously important to the users, however, an official word isn't that hard to get. i'll stop saying anything other than 32bit in the mean time.

[Shroomz~>]

yes, the Scope environment is 32bit, i know.

So are the mixers Gary. There might be something different employed in the likes of the EQs or other elements, but that doesn't change the fact that the mixers are basically processing audio @ 32-bit INT.

[garyb]

so you said.
i know little to nothing of the nuts and bolt internals of the mixers, Mark, only what i've been told. if you like, you can tell me again. i'll use a smiley face instead of a roll eyes so that we stay friends.

[astroman]

well, noone said that the mixer is a single stage device...
other than that I'd ask: can you hear it ?
no ? chopp it off

cheers, Tom

[Music Manic]

well, noone said that the mixer is a single stage device...
other than that I'd ask: can you hear it ?
no ? chopp it off

cheers, Tom

It doesn't matter if you can hear it or not but when you deal with processing it makes a big difference.
For instance if your audio signal is converted from 32 bitfloat from your DAW and becomes
distorted, then further processing will distort the signal. Further to that, distorted low level signals will become very evident when they are Normalised or have gain added to them.
If you finalised file has become distorted then it is not good.

So my point is,is that I want to know what is happening to the audio signal from A to B that's why I said that VDat has a special sound and it may be due to the audio stream and processing.

Thanks