Why can’t Virus use “Samples” for oscillators?

[synthman1]

Something I’ve been requesting since the Virus A appeared is being able to use samples for the oscillator section either internally or externally. (Similar to what the Nord Wave achieves for instance.)

I’m not sure why we don’t have this ability yet as the virus already has 64 spectral waves and wave tables. I would love to at very least have access to 64 different samples – unusual vocals and choirs, saw waveforms from various vintage synths, maybe a simple piano sample, etc.

In coming back to this idea in how to achieve it, if the sample storage can’t be obtained internally on the Virus:

Option 1:

Implement this ability via an external usb flash drive for the multi-sample storage connected to the usb port of the Virus and have the samples accessible from the oscillator section. The Alesis QS line uses this exact method. The multi-samples are burned to a flash card, the card is inserted into the card slot on the synthesizer and the custom samples are now accessible. Piece of cake. This was limited to 8MB of storage per card for the QS line.

The Roland XV-5080 has this ability as well by using a standard SmartMedia card. Today’s thumb drives and memory cards hold GB’s worth of data. You would be able to have access to a massive customizable sample library with relative ease. This would be a very powerful and virtually limitless feature for extending the Virus’s capability tonally.

Option 2:

By using a simple external hardware device developed by Access or a 3rd party developer for exclusive sample storage and usage with the Virus. This small, afforable hardware box would have a pair of ¼” audio outs where the samples are accessed via audio inputs on the Virus. The writable samples would be transferred via a usb interface from a computer to the hardware device.

A 3rd option, though nowhere as direct or straightforward as option 1 or 2 is:

Perhaps developing a simple software sample manager that operates as a stand alone application. This would operate similar to a softsynth except the samples are routed out through your audio interface and then is accessable via the audio inputs just like any other audio source where it’s routed through the Virus signal chain. This already works to some degree via the audio inputs routed to the filters/effects sections.

In each of these 3 options, the Virus OS would need to be modified to accommodate this implementation of course.

[Timo]

Hi Synthman, I agree 100%.

I've long been campaigning for user wavetables, but larger user-PCMs would be phenomenal. The Korg Radias has onboard preset synth PCMs, but I have to say in this case I only tend to use three or four of them as the majority of them don't suit my music tastes. (However it also has one-shot drum PCMs on there, and I use these all the time.)
I would kill for the ability to actually overwrite the synth PCMs with my own, though, as it would be far more useful to me.

8MB flash ROM or a discreet insertable card (SD, MMC, or Compact Flash, etc.) for user-PCMs would be absolutely killer on a Virus. In addition to extensive personal creative use, it would also open up the possibility of creating/swapping/selling sample data sets for the Virus.

Sound-wise, I dare say it could make use of multi-sampling (to avoid aliasing or the chipmonk effect) for longer samples, but this would open up a whole new exciting and creative area for the Virus. Or you could use the samples as wavetables (with a wavetable designer inside Virus Control), or longer looped multi-samples, or one-shots (drums, etc.), the list goes on...

I'm fully with you.

In fact, the Korg Trinity's "PlayBack Sampler" option used 8MB of Flash ROM, and you could use it however you wished. - Be it for single-cycle waveforms (effectively wavetables), longer looped multisamples, or one-shots. It was very well implemented. Ironically I've multi-sampled various Virus patches and have imported the waveforms into my Trinity.

The official Korg soundsets also used proprietary 2:1 lossless compression ratios, effectively shoehorning 16MB of waveform data into the 8MB available space. Korg never publicly released details of their proprietary compression implementation, so the rest of us were limited to using 1:1 compression for user-created samples.

Welcome to the forum, btw. ;)

[psy604]

2:1 lossless compression is ordinary called FLAC.
think it's like wave audio with variable bitrate.
(you could grab such encoder and try if it can read it)

16 user defined waveforms in the virus would be basically enough
and one more tab in Virus Control to manage them ...

always wondered how some artists do this voiced synth sounds, now i know how :)

[synthman1]

If Access figures out a way to implement this, why limit it to 8MB? That's nothing by todays standard. Most usb flash drives are well over a GB.

But I'll be more than happy to take 16 MB 2:1 in 8MB or 16 user defined multi-samples. Thats better than none. You could still put a few of your favorite multi-sample in there. I would love to sample my DSS-1 and push those great vintage 12 bit samples through a Virus.

I'm advocating writeable samples. That way you can replace the stocks one's with what ever you like. Mind you, I'm not suggesting the Virus should be a sampler or contain a 4 layer, velocity switching 88 note sampled Steinway. 1 or 2 single layer samples per octives (single or dual osc) would be great though. Straight up 44.1 16 bit. Down and dirty, no sample editing, load and go. Though a 12 and 8 bit reducer feature in the ocs section would be awesome! (Think vinatage DSS-1, Emulator or Mirage quality.)

Look how many people contributed patches to the Virus. Imagine the sample database Virus users could create. If Access could do it via a flash drive, then you could put what ever you want on it.

How ironic. I just read the new Blofeld Keyboard has guess what - user sample implementation. Come on Access, catch up! You guys are the leaders.

[IamEvil]

If Access figures out a way to implement this, why limit it to 8MB? That's nothing by todays standard. Most usb flash drives are well over a GB.

Wrong end of the USB cable goes into the virus. The virus is the USB slave and not the USB master controller.

But I'll be more than happy to take 16 MB 2:1 in 8MB or 16 user defined multi-samples. Thats better than none. You could still put a few of your favorite multi-sample in there. I would love to sample my DSS-1 and push those great vintage 12 bit samples through a Virus.

Ha ha , you would be happy with 16MB ram ? How much ram do you think is actually available inside the Ti ?

Each full soundbank is about 128k so you would have to lose 10 whole banks for 1MB of sample ram. 1MB of sample ram @ the virus internal frequency 96khz 24bit isn't going to give you much, 2-3 seconds or so.

It's a lovely idea but I really doubt the virus hardware is made for it. Maybe some lofi user wavetables which take up about 50k each or something but PCM samples is probably going too far.

[synthman1]

>Wrong end of the USB cable goes into the virus. The virus is the USB slave and not the USB master controller.

Wait a min. If you store the samples on flash ram drive externally and connect it to the Virus, in theory, how would accessibility be any different than if you connected the flash drive to a computer to play mp3's or samples from it? The USB interface is 2 way communication, right? Maybe Access can clarify.

>Ha ha , you would be happy with 16MB ram ? How much ram do you think is actually available inside the Ti ?

That’s why all 3 of my suggested options used other methods than internal storage include 2 option for using the audio inputs.

>It's a lovely idea but I really doubt the virus hardware is made for it. Maybe some lofi user wavetables which take up about 50k each or something but PCM samples is probably going too far.

Not to over simplify this, unless Access can clarify, the virus is really a computer with modifiable software, essentially "a softsynth in a hardware host". Theoretically Access could rewrite the entire OS to make the Virus be something completely different if they choose to only limited to the amount of physical storage size for the OS.

This is how Access was able to add the Phaser effect in the B and the Moog filter in the C, right? It's just software, 1's and 0s.

[IamEvil]

Wait a min. If you store the samples on flash ram drive externally and connect it to the Virus, in theory, how would accessibility be any different than if you connected the flash drive to a computer to play mp3's or samples from it? The USB interface is 2 way communication, right? Maybe Access can clarify.

No, one is a master and one is a slave. The master gives power to the slave. When you plug your USB stick into the PC, the pc powers it.
I'm not saying that the virus cannot power the USB stick. I'm saying that the virus has the "slave" plug and because USB is a global standard for data communication between devices, I can't see access breaking this standard or releasing a new hardware revision with an extra usb port on.

Simply put.. USB memory sticks have the wrong socket and making a cable so it would fit could damage all other usb devices if accidently used on them.




That’s why all 3 of my suggested options used other methods than internal storage include 2 option for using the audio inputs.



The virus would still need enough memory internally to buffer the audio because the USB memory sticks couldn't handle it.
Try copying lots of small files too and from a USB memory stick and watch it grind to a halt while you see a little bar update and copy files. This is what would happen if you hit a chord or played too fast.



Not to over simplify this, unless Access can clarify, the virus is really a computer with modifiable software, essentially "a softsynth in a hardware host". Theoretically Access could rewrite the entire OS to make the Virus be something completely different if they choose to only limited to the amount of physical storage size for the OS.

This is how Access was able to add the Phaser effect in the B and the Moog filter in the C, right? It's just software, 1's and 0s.

Yes. In theory the virus is a hardware dongle for a softsynth. There shouldn't be any reason at all why you can't turn it into a DX7, a Supernova 2 , a Vari-os etc.

It boils down to the available spare memory inside the virus. The atomizer makes me believe that there isn't much left because it doesn't let you use really low BPMs or full bar looping.

[psy604]

PC/MAC -> USB Cable -> Flash RAM -> USB Cable -> Virus TI

... so that the flash just sits on/in the cable.
but it's the question if this would cause sync/timing problems.

[IamEvil]

PC/MAC -> USB Cable -> Flash RAM -> USB Cable -> Virus TI

... so that the flash just sits on/in the cable.
but it's the question if this would case sync/timing problems.

Flash ram with USB connection :-
http://www.dansdata.com/images/flashthings/usbdrive440.jpg

Virus Ti USB connection :-
http://www.accessmusic.de/events/11-2004/virus_ti_details_usb300.jpg

Difference in ports:-
http://telefonkabel.flatrate-discount.de/images/usb-kabel.jpg

They won't physically fit together :)

[synthman1]

What about using a female to female USB adaptor or USB cable that would accommodate the USB plug end on the Virus to Flash drive? Is there such a thing?

I'm not disputing this still couldn't be achieved, but if it’s just a matter cable configuration holding this functionality back, it seems this could easily be overcome.

>Try copying lots of small files too and from a USB memory stick and watch it grind to a halt...

Why would you have to copy the samples to the Virus? A flash drive or flash based MP3 player such as an ipod doesnt copy the files anywhere when you listen to music from it? The QS doesn't copy the samples from its flash card to the synth, yet the sample are still available to create patches with. The QS flash card, USB flash drive or a Flash drive Ipod is RAM that is both used for storage/RAM, right?

What about options 2 and 3 then?

[IamEvil]

ha ha , I didn't know those dual female end adapters existed

ok you win

i give up :)

[Timo]

Addressing RAM or ROM is much, much faster than pulling data off a removable storage device. Flash ROM essentially allows data to be burnt (in bulk) onto a ROM chip. Furthermore, the flashing process takes time.
It's not a "flash drive", as such. Completely different kettle of fish.

If you were to allow sample uploading via a USB pen-drive or the like, it would have to be burnt to memory (either volatile RAM [meaning that sample data hosted on the Virus will be lost when powered down, and would need to be re-uploaded again next time] or flash ROM) for the Virus to use it. Therefore if you really wanted to use a 1GB of sample data (lol), you'd likely also need 1GB of flash ROM or RAM in there too to copy it to, so it could access the data quickly when needed.

For large-scale sample manipulation, you might as well buy a ROMpler like the Triton/Oasys/M3/Motif/Fantom.

Single-cycle waveforms for wavetables take up a few tens or hundreds of kb.
PCM samples would take up far larger. I think 8mb of flash ROM is a good size for dirty creative sample manipulation, and 8mb wouldn't be too expensive (given that the Virus is now £1700, I'd damn well hope the TI mk2 would do that for that price).

I'm also interested in how Access currently implement their wavetables. I've always thought that, ideally, you'd need to have multisampled waveforms (say, at least one per octave) so they don't chipmonk or alias when pitched up or down, so I imagine the current wavetable implementation may use mathematic vectors or somesuch to possibly create the waveforms?

The standard digital waves in the Virus B have a tendency to sound a little toy-like, so I think they might even be just one single-cycle sample stretched up and down the keyboard.

I hope when they bring out user single-cycle waveforms (for the TI mk2) and you can upload or create your own type of 'saw' wave (say, sampled from a different synth), that it wont sound toy-like, like the other digital waves do.

Longer user-PCMs (say 8mb) would be ace, though.

Here's what the Korg Trinity's 8mb flash ROM playback sampler allowed:-

Multisamples: 100 (max)
Drum samples: 200 (max)
Individual samples: 500 (max) (total includes normal Samples and Drum-samples).

A "sample" merely refers to a single waveform. A "multisample" refers to a collection of samples, key-mapped across the keyboard by the user.

So you could have 500 samples and 100 multisamples and no drum samples. Or you could have 200 drum samples, 300 samples and 100 multisamples. etc.

Samples can be looped or one-shot as desired. (Drum samples are forced one-shot).

Of course you'll never reach using those amounts of numbers unless you're sampling single-cycle waveforms (which you can do on the Trinity) as the 8MB otherwise becomes the limit.