Author |
Message |
creamware
Joined: May 25, 2004 Posts: 3 Location: uk
|
Posted: Tue May 25, 2004 12:10 pm Post subject:
Whats the aliasing like on the g2 |
|
|
I would like to buy a g2 soon but can some one tell me
Whats the aliasing on the full-range oscillators and filters.
thanks |
|
Back to top
|
|
|
Tim Kleinert
Joined: Mar 12, 2004 Posts: 1148 Location: Zürich, Switzerland
Audio files: 7
G2 patch files: 236
|
Posted: Tue May 25, 2004 2:27 pm Post subject:
|
|
|
The G2 has, IMO, the best DACs I've personally ever heard on a VA. (Can compare it to NL2, NL3, Virus B, Korg Z1, Waldorf Q)
The fullrange oscs (as well as any audiobandwidth modulation) runs strictly on 24/96. I've yet to detect any aliasing -except of course when you do very crazy stuff.
I'm very satisfied.
...my 2cc. |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
Posted: Tue May 25, 2004 2:43 pm Post subject:
|
|
|
Welcome, Creamware, to electro-music.com
I agree, aliasing isn't much of a consideration. Could you suggest a way to test for aliasing, other than subjective listening? |
|
Back to top
|
|
|
creamware
Joined: May 25, 2004 Posts: 3 Location: uk
|
Posted: Tue May 25, 2004 3:16 pm Post subject:
|
|
|
Thanks for the tips guys
But can it deliver perfect sound over the entire frequency spectrum
with out aliasing say from -c2 to c8 with a sin wave eg.
1. The nord 3 can not as i have one here.
The creamware Minimax can thanks to the high quality of the algorithms employed.
2. Please give it a test and let me know
3. I need a good synth for the film work that i do
and the g2 sound good but on the nord 3 i have
there is aliasing in the top and this tend to put me off the g2 |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
|
Back to top
|
|
|
elektro80
Site Admin
Joined: Mar 25, 2003 Posts: 21959 Location: Norway
Audio files: 14
|
Posted: Tue May 25, 2004 4:23 pm Post subject:
|
|
|
Ehh.. a sweep from sub to hyper.. and no audible aliasing? What if you add a very very subtle LFO to the oscillator pitch and redo the sweep? _________________ A Charity Pantomime in aid of Paranoid Schizophrenics descended into chaos yesterday when someone shouted, "He's behind you!"
MySpace
SoundCloud
Flickr |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
Posted: Tue May 25, 2004 4:25 pm Post subject:
|
|
|
Opps, the upload didn't work because wav was not enabled. I just did some admin duties and got wav one the list.
I did use an lfo to sweep it. |
|
Back to top
|
|
|
elektro80
Site Admin
Joined: Mar 25, 2003 Posts: 21959 Location: Norway
Audio files: 14
|
Posted: Tue May 25, 2004 4:27 pm Post subject:
|
|
|
Well.. I heard no aliasing there.. _________________ A Charity Pantomime in aid of Paranoid Schizophrenics descended into chaos yesterday when someone shouted, "He's behind you!"
MySpace
SoundCloud
Flickr |
|
Back to top
|
|
|
elektro80
Site Admin
Joined: Mar 25, 2003 Posts: 21959 Location: Norway
Audio files: 14
|
Posted: Tue May 25, 2004 4:29 pm Post subject:
|
|
|
Uh... yeah.. but I meant like having an LFO wobble it while the general sweep was going on. That tends to break the illusion on some lesser gear. _________________ A Charity Pantomime in aid of Paranoid Schizophrenics descended into chaos yesterday when someone shouted, "He's behind you!"
MySpace
SoundCloud
Flickr |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
|
Back to top
|
|
|
elektro80
Site Admin
Joined: Mar 25, 2003 Posts: 21959 Location: Norway
Audio files: 14
|
Posted: Tue May 25, 2004 4:52 pm Post subject:
|
|
|
yes... exactly... ! Hey this is cool! And still now sign of aliasing _________________ A Charity Pantomime in aid of Paranoid Schizophrenics descended into chaos yesterday when someone shouted, "He's behind you!"
MySpace
SoundCloud
Flickr |
|
Back to top
|
|
|
ian-s
Joined: Apr 01, 2004 Posts: 2669 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
Posted: Tue May 25, 2004 7:52 pm Post subject:
|
|
|
ummm, sine is not the best choice for testing aliasing. It is easy to get an 'alias free' sine with no fancy dsp tricks what so ever. Try repeating your tests with waves that have sharp edges, the worst offenders are the 10% pulse and the sawtooth from the shape oscillator at 99%. The latter is the worst that I tested in terms of aliasing. The saw and square on the other oscillators are pretty good.
Not wishing to be hard on my beloved G2, aliasing is probably the biggest problem with VA and there is no such thing as superior algorithms. It is simply a trade off between inaudible aliasing in the top few octaves, and waves which still have enough high harmonics to produce great sounds. If you are aware of the problem, its not too much trouble to 'design' around it. I think Clavia have got it pretty much on the nail. |
|
Back to top
|
|
|
Tim Kleinert
Joined: Mar 12, 2004 Posts: 1148 Location: Zürich, Switzerland
Audio files: 7
G2 patch files: 236
|
Posted: Wed May 26, 2004 12:54 am Post subject:
|
|
|
Yes -VA ain't perfect. It's a tradeoff between pros and cons. But imagine carrying around the equivalent of possible G2 patch modules in real hardware... ...not to mention having to pay for them . Oh, and saveability too.
The only time I got intrusive G2 aliasing was with my custom filter design -the resonance was very pronounced and when sweeping it up into the HF... voilà the artefacts. But implementing Robs wonderful "'warmth mixer' integrator" into the FB loop fixed it splendidly.
As Ian said -you have to work around it. All VAs have to. Hardwired applications like the Minimax have such workarounds already implemented into their algorithm. On the G2, you have to do this yourself -albeit not very often. And the unparalleled flexibility of the G2 makes it possible to find an elegant solution. A challenge in itself.
And as a long time NL3 user, I can say that the G2 is in another league in terms of sonic quality. The DACs sound very similar -pristine punchy but very sober- the typical Clavia sound. But it's the code that's different, I suppose. And because of the modularity, you can bend the overall sound in so many different ways that it is fair to say that the G2 in many ways does not have a "sound" of its own. The sound is what you make it. |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
Posted: Wed May 26, 2004 7:14 am Post subject:
|
|
|
g2ian wrote: | Try repeating your tests with waves that have sharp edges, the worst offenders are the 10% pulse and the sawtooth from the shape oscillator at 99%. The latter is the worst that I tested in terms of aliasing. The saw and square on the other oscillators are pretty good.
|
Hmmm. Here are two more tests. I don't hear any problems with the 90% pulse, but the 99% saw from shape osc does sound pretty nasty.
Description: |
test using a 10% pulse wave |
|
Download |
Filename: |
g2-sweep-10pct.wav |
Filesize: |
5.87 MB |
Downloaded: |
2408 Time(s) |
Description: |
test using a 99% saw from the shape osc |
|
Download |
Filename: |
g2-sweep-99saw.wav |
Filesize: |
5.92 MB |
Downloaded: |
2407 Time(s) |
|
|
Back to top
|
|
|
Rob
Joined: Mar 29, 2004 Posts: 580 Location: The Hague/Netherlands/EC
G2 patch files: 109
|
Posted: Tue Sep 21, 2004 3:15 pm Post subject:
|
|
|
creamware wrote: |
But can it deliver perfect sound over the entire frequency spectrum
with out aliasing say from -c2 to c8 with a sin wave eg.
1. The nord 3 can not as i have one here.
|
The Nordlead 3 has a relatively low sample rate, the G2 sample rate is more than twice as high. Also. I've understood that many of the oscillator anti-aliasing algorithms on the G2 have been improved upon.
The only thing that cannot be anti-aliased is hardsync, it will always sync to the 'sample boudary', which is more like frequency jitter. Anti-aliasing of hardsync would be much to dsp-intensive to implement. Still, hardsync is also improved by letting the synced waveform restart at different positions for different waveforms, flanked waveform sync at the anti-aliased flank while flankless waveforms sync at the zero crossing.
All in all the G2 sounds pretty aliasing free. |
|
Back to top
|
|
|
7om
Joined: Sep 23, 2004 Posts: 29 Location: SF Bay Area, Calif.
G2 patch files: 1
|
Posted: Thu Sep 23, 2004 3:06 am Post subject:
Filter aliasing or resolution? |
|
|
I just received my G2 today and am very impressed by how deep this instrument is. My only concern while shopping was a comment in Craig Anderton's review about audible "stepping" with high resonance settings. As I was exploring that tonight, I believe that the problem is due to MIDI controller limits of values 0 to 127. As best as I can tell, the rotary knobs only send values 0 to 127. Even when the sweep only covers an octave, one can hear the steps, and it is not uncommon to have a sweep cover more than an octave.
I'm trying to determine whether this is a serious problem for me or not. Shoshin mentions "Robs wonderful 'warmth mixer' integrator." Can anyone elaborate on this? Where do I find it and how best to use it? Any other comments on audible stepping would be appreciated. As brilliant as this synth appears, that could be a show stopper for me.
Thanks,
Tom |
|
Back to top
|
|
|
ian-s
Joined: Apr 01, 2004 Posts: 2669 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
Posted: Thu Sep 23, 2004 3:43 am Post subject:
|
|
|
If you set the filter to sweep on a morph group knob, the steps are not obvious up to about a four octave range. I was going to suggest the Pitch Stick would be good for big smooth sweeps but tests show that Clavia have not implemented the LSB on the device module which makes it as bad as the knobs for stepping. Maybe the CTRL.PEDAL input is not quantized.
Should point out that I personally don't have a big problem with this, my filters tend to get swept by an army of other modules in full 24bit glory.
I understand that 'playing the panel' is a long standing synthesis tradition however, so I would be surprised if Clavia were not thinking hard about this. |
|
Back to top
|
|
|
Rob
Joined: Mar 29, 2004 Posts: 580 Location: The Hague/Netherlands/EC
G2 patch files: 109
|
Posted: Thu Sep 23, 2004 4:15 am Post subject:
Re: Filter aliasing or resolution? |
|
|
tom wrote: |
I just received my G2 today...
<snip>
Tom
|
Stepping is indeed due to the MIDI specs that only allow 128 values for a midi cc#.
The way to get smoother control is to use a ConstantSwitch module, which is in fact a module with a knob that provides you with a control signal (Control Voltage one would have named it in the past). This module not only has a knob to tweak but also a pushbutton that can instantly switch the control signal on or off and act like sort of a Mute button. Additionally this pushbutton can be named and the name shows up in the G2 displays.
By using a Smooth module after the ConstantSwitch and then routing the smoothed signal to the Resonance modulation input of the FltNord you can very smoothly crack up the resonance.
You can also use it to control the Cutoff frequency of a filter. Additional advantage is that you can limit the range by presetting the filter to e.g. a few hundred Hz and setting the filter control input attenuation knob to e.g. about a quarter open. This way you can get a very smooth resonance sweep over just the right range.
What you need to realize is that the G2 is a modular synthesizer, which means that not all is preprogrammed for you, you will need to patch something yourself now and then. But there is an enormous amount of tricks and techniques to get exactly what you want. Much, much more than on a 'preprogrammed' synth. Just look at it this way, where the frustration starts on a 'preprogrammed' synth the fun starts on a G2.
The warmth trick is not really a trick, it is an attempt to psychoacoustically correct the very high frequency response of digital instruments. Digital instruments in general have a flat frequency spectrum. But contrary to amplifiers and speakers that are preferrably 'flat', a synth with a flat frequency spectrum sounds overly bright, which generally kills the high in a mix. Analog synths have a smoother high frequency roll-off and hence they are said to sound warmer and perform better in a mix. How to correct any digital synth is quite a deep subject, involving how the mind perceives spatiousness and depth in sound and how the balance in different frequency bands in the spectrum should be to draw the mind's attention to certain instruments in a mix.
This sort of subjects is where the G2 makes a difference with almost anything else, as the G2 allows to create this sort of corrections yourself to your own taste and needs.
A 'preprogrammed' synth is something like a magnetron meal, push the button and after a few seconds the stomach will most certainly get filled to the brim. The G2 is more like cooking a 'haute cuisine' meal, at first one looks puzzled at the ingredients before you, but after a while you can get much better results and create one's own endless variety in flavours that one won't find in magnetron meals. Aah well, that's how I like to look at it, others may strongly disagree.
/Rob |
|
Back to top
|
|
|
modular
Joined: Jul 26, 2004 Posts: 185 Location: Rome, Italy
Audio files: 5
G2 patch files: 72
|
Posted: Thu Sep 23, 2004 8:15 am Post subject:
Re: Filter aliasing or resolution? |
|
|
Rob wrote: | Just look at it this way, where the frustration starts on a 'preprogrammed' synth the fun starts on a G2.
/Rob |
I'll frame these words then put them on the wall up to my sitting room!!
|
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
Posted: Thu Sep 23, 2004 8:19 am Post subject:
|
|
|
Congrats, Tom, on your new synth, and welcome to electro-music.com.
I'm sure that after a while, you'll really love your G2. I feel similarly about the 128 steps and would like to see at least one controller on the G2 with higher resolution. Still, Rob is quite right, this isn't a show stopper. Internal CV resolution is 24 or 32 bits if I remember correctly.
There are many ways to solve patching problems. When you hear something that's bothering you, just take some time and figure out exactly what is the cause and often you discover a solution. In the process you'll discover some useful techniques that will help you in subsequent patches.
If you have a truly "unsolvable" problem, try posting it here on the forum. Maybe someone will take it as a challenge and come up with a solution.
Back to the stepping, I find that most of the time I don't notice it, or if I do, I'm not bothered by it. When there are instances where it is bothersome most of the time just patching in a smoother does the trick. I like to sweep oscillators over the full 10 8va range with a knob. _________________ --Howard
my music and other stuff |
|
Back to top
|
|
|
modular
Joined: Jul 26, 2004 Posts: 185 Location: Rome, Italy
Audio files: 5
G2 patch files: 72
|
Posted: Thu Sep 23, 2004 8:20 am Post subject:
Re: Filter aliasing or resolution? |
|
|
Rob wrote: | Aah well, that's how I like to look at it, others may strongly disagree.
/Rob |
Me completely agree. |
|
Back to top
|
|
|
7om
Joined: Sep 23, 2004 Posts: 29 Location: SF Bay Area, Calif.
G2 patch files: 1
|
Posted: Thu Sep 23, 2004 12:30 pm Post subject:
|
|
|
Thanks to all for the replies. I know I have a lot to discover in terms of creating and tweaking patches, but it is definitely the reason that I chose the G2.
Rob, you mention a "smoothing module," though I can't seem to find it. In which group would I find that module? This seems like the right idea, kind of like "glide" for MIDI controllers.
Tom |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Thu Sep 23, 2004 12:38 pm Post subject:
|
|
|
tom wrote: | Rob, you mention a "smoothing module," though I can't seem to find it. In which group would I find that module? |
You'll find it as 'glide' on the 'note' tab, 'smooth' was the word used for the NM Classic.
Jan. |
|
Back to top
|
|
|
Rob
Joined: Mar 29, 2004 Posts: 580 Location: The Hague/Netherlands/EC
G2 patch files: 109
|
Posted: Thu Sep 23, 2004 2:17 pm Post subject:
|
|
|
tom wrote: |
Rob, you mention a "smoothing module," though I can't seem to find it. In which group would I find that module? This seems like the right idea, kind of like "glide" for MIDI controllers.
Tom |
Sorry, its the Glide module from the Note tab. |
|
Back to top
|
|
|
7om
Joined: Sep 23, 2004 Posts: 29 Location: SF Bay Area, Calif.
G2 patch files: 1
|
Posted: Thu Sep 23, 2004 2:18 pm Post subject:
|
|
|
YES!!! Just tested it with the Glide module. Not perfect, but it makes a HUGE difference. This example also proved the unbelievable flexibility and power of this machine. It's reminding me so much of my early days on a Buchla 200, only more productive.
Thanks again for the help!
Tom |
|
Back to top
|
|
|
|