Can Fantom G or SonicCell Do sampling?

[Septimo]

Soon they'll be combined as one....You'll see...they are already giving you a window into the future...can you see through it?

[V-CeeOh]

mp3 sounds nice on my phone but I would neeeeeeever use it as file type for music work :-/

[The Audacity Works]

Soon they'll be combined as one....You'll see...

Actually, they won't. Probably ever, because by the time mass-voice playback and pitch/tempo manipulation of MP3 files without amplifying the already egregious compression artifacts might affordably exist, storage sizes will be such that compression's unnecessary.

Besides, even if it were feasible (is isn't), it'd be pointless to go down that road, as pitching (realtime sample rate adjustment) MP3 files up and down sounds like complete ASS. Roland would never support something that makes their stuff sound worse, just to optimize their RAM and storage usage.

Unless you mean Fantom-style RAM sampling alongside single MP3 file playback ala the RD-700GX... which isn't MP3 sampling at all.

[Septimo]

What? please...the wavs used on these things are so compresed I'm surprised that they aren't MP3's...Personly I don't thinK you can tell the difference..especially in single sounds... in a stereo mix with a whole spectrum of instruments, maybe slightly, but in single one instument sound almost nothing if anything...It depends on what you're doing...people still are sampling vinyl...It will work, and you will be seeing this in the future...

[SRUT, BUT...]

Ahem,...

...isn't wave form compression and audio file compression two different shoes?


Just my 2cents...

[The Audacity Works]

Septimo:

What? please...the wavs used on these things are so compresed I'm surprised that they aren't MP3's...

You're confusing the Fantom's ROM sounds with samples loaded into RAM. The ROM is compressed (which is a proprietary format), any samples you record/load aren't.

It will work, and you will be seeing this in the future...

Hmm... Interesting. My DSP engineer says it won't. In fact, I think his words, verbatim, were "That's a completely pointless exercise."

[The Audacity Works]

...isn't wave form compression and audio file compression two different shoes?

We've only been discussing data compression.

[Septimo]

"My DSP engineer says it won't"

Uh-Oh, audacitys engineer said? then you gotta be right...

We'll see....

[Septimo]

You still end up loosing sound quality...how much more over Mp3 is unknown...

[The Audacity Works]

Septimo:

Uh-Oh, audacitys engineer said? then you gotta be right...

Well, Septimo, what makes you think it's easy for Roland to just ADD MP3 sampling? Explain the process and implementation. After all, you must have some sort of electronics or MI product design background.

This is getting retarded. You might as well visit a coding forum and say "They should add drag and drop jpegs to C++ lines!" When coders show up and say "Uh, yeah, Septimo, it doesn't work that way", you can ignore them too.

[The Audacity Works]

You still end up loosing sound quality...how much more over Mp3 is unknown...

What you SAMPLE into a Fantom and play back is UNCOMPRESSED. .WAV is the default, but you can switch this to AIFF if desired. In fact, you can convert .WAVs to AIFFs and back a million times and the audio portion of the end result will be bit-for-bit identical with the original.

The built-in ROM sounds ARE indeed compressed, as it is with nearly every other ROMpler on the planet. They have nothing to do with the Fantom's sampling engine.

[Septimo]

That's what I'm talkin about...the built in rom sounds...not the wav you sample from...

How much does that compression differ over an MP3 comression?

You seriously think this is impossible? or are you just trying to be right? Or the one who only knows?

They have Ipod DJ systems that can manipulate tempo and pitch of mp3's and ogg vorbis and they sound awesome...what makes you think Roland is retarded? Your thing is that it can't be pitched without sounding like ass, well let me tell you, i've seen these things and they sound pretty damn good while pitching them...Anything can be done...

again,,,, we will see this sometime...

[The Audacity Works]

Septimo, this whole discussion has been about SAMPLING. The guy gave Roland crap because they didn't support compressed SAMPLING. Even if Fantom-G 2.0 supports the single streaming of MP3s (Audio Key) from USB, this is NOT SAMPLING.

I am intimately familiar with streaming MP3 files from flash/USB. At this point, I really wish I wasn't.

Your iPod DJ system isn't triggering 128 voices (64 simultaneous stereo songs) through LFOs and envelopes that can get instantly pitchshifted 48 semitones (400%) up and down instantly with no additional artifacts. The Fantom (and indeed, most samplers since the Akai S5000 way back in '99) can do this because the samples are UNCOMPRESSED.



Okay, we are only now switching the discussion to ROM playback:

Neither you nor I know to what capacity Roland's compression scheme affects the end result, much less what ratio. As I understand it, the ratio will not only change from patch to patch, but from raw sample to raw sample. 128kbps MP3 compresses at approximately 10:1 across the board, and I doubt Roland's is that extreme, but I do know one thing: Their custom compression scheme is designed specifically for their ROMpler engine. MP3 opens a huge can of worms, and honestly, since we don't have access to edit the ROMpler sounds anyway, why would anyone care whether or not they use MP3? It's like caring whether or not your favorite movie was edited in AVID or FCP.

And I'm only going to peripherally mention the cost of licensing Fraunhöfer algorithms.

[Septimo]

Why you bring that up is beyond me....I Know damn well that the sampler records uncomressed. I never said different. when I replied to V-C post I was talkin about the ROM wavs not sampled wavs..in other words if ROM sounds are compressed how much more over MP3? and if we have MP3 playback how is that different from using wav?.And I never said actually sampling in MP3 either although I know that will be possible too...But to at least load one in and do with it as you wish isn't rocket science. And not being able to edit ROM wavs is irrelivent...cus you still manipulate them in the patch edit..pitch them, LFO them, etc...all except reverse them...how would MP3 differ here? other than being able to reverse it?

Is it possible? Yes, I think it is...and You know so as well...You're just trying to will an argument instead of seeing it;s possiblity(although I think you see it too)...And if that's the case, then okay.....You win....It's impossible...end of discussion....You're right....

[The Audacity Works]

1. MP3s generally sound like crap in a studio environment (especially high-transient material like hihats and snares). They're barely listenable on a decent quality home stereo. On cheap earbuds or in the car, they're fine.

2. All DSP functions apply to uncompressed waveforms. For the most part, any time you want to alter an MP3 file in hardware, you have to convert it to .WAV or AIFF, and then back to MP3. You're losing sound quality in every conversion.

3. Loading a singluar MP3 file into sample RAM is a far, far cry from loading 128. A simple file player (such as a flash-based iPod, the RD-700GX's Audio Key, your iPod DJ thingy, or the box I'm designing) is a piece of cake in comparison.

4. Applying pitchbend adjusts the playback sample rate of the sample. Trying to do this to MP3 files sounds like complete ass and amplifies compression artifacts.

5. Realtime timestretch is many times worse-sounding on MP3 files than uncompressed files and requires completely rewritten algorithms that have to crunch numbers faster. Therefore, it takes more DSP power.

6. Licensing costs.

7. Extra steps in implementation, slowing down your workflow.

8. It's a huge step backward, concept-wise.

I'm not a DSP engineer myself, but I spend WAY too much time conferring with those who are on this exact subject to admit I know less than your average Fantomized member.