Zebra's biggest flaw

Official support for: u-he.com
Post Reply New Topic

What's Zebra's biggest flaw?

Too high Resonance in XMF in high frequencies
15
11%
Inconsistent volume of filter algos
11
8%
XMF Overload too loud
8
6%
Clicks in FMOs with filtered FM
3
2%
Mixers are linear gain, dB scale would be preferrable
8
6%
Other (have your say!)
22
16%
Chorus sounds too flat
4
3%
Waveshapers lack guts
67
49%
 
Total votes: 138

RELATED
PRODUCTS

Post

mabian wrote:
Urs wrote:
Note: As these changes would make things sound different, I've come up with a spiffy idea: Existing presets load with the currently existing algorithms. A little icon in the main display will indicate "2.3" for Zebra V2.0 to V2.3. That means, everything will sound exactly the same as before. But you can also just switch to the improved stuff and work from there.
:clap: :clap: :clap:

Urs, standing ovation for the above: I'm seeing other plugin devs "improving" their current products completely disregarding the impact on existing patches. An "enhanced" reverb or filter, to make a couple examples, could screw up resonance and character obtained tweaking to death the "legacy" behaviour...

Talk about the appeal of Lo-Fi... ;)

Again, a strong praise for this thought.

- Mario
Yes, but would existing sequencer projects load Z2.5 with the correct presets that they were saved with in 2.3? I wouldn't want to upgrade to 2.5 unless my presets were correctly loaded (including modifications) with a project.

Post

Sorry but I can't think of anything.
I'm just satisfied with Zebra in everyway.

Some have mentioned, Zebra distortion as not being good?
I would disagree with that. It's great distortion, but might be hard to find for many.

Post

Urs wrote:
Note: As these changes would make things sound different, I've come up with a spiffy idea: Existing presets load with the currently existing algorithms. A little icon in the main display will indicate "2.3" for Zebra V2.0 to V2.3. That means, everything will sound exactly the same as before. But you can also just switch to the improved stuff and work from there.
ganymead wrote: Yes, but would existing sequencer projects load Z2.5 with the correct presets that they were saved with in 2.3?

Post

I'll just second that the promise of backwards compatible patches, default adapting the algorithms to the authoring version, is GREAT.

:tu:

Post

MitchK1989 wrote:I got it to work reasonably well with keytrack on 64
64 is it. This is it. I mean it. 64 is the final keytrack value to go for. Anywhere. Except KeyFollow in filters of course, where it's 100%.

With a keytrack of 64 you usually have to dial in a constant (untracked) frequency of the middle E. Which is also note number 64 (coincidence?!?)

Post

MitchK1989 wrote:Also, is there no random function for the arp?
Good call! Always been on the list.

Post

ganymede wrote:Yes, but would existing sequencer projects load Z2.5 with the correct presets that they were saved with in 2.3? I wouldn't want to upgrade to 2.5 unless my presets were correctly loaded (including modifications) with a project.
But of course. Backwards compatibility has always been a major concern here. That's the beauty of the new solution, an old preset/project will just open with 2.3 in mind, while new presets will utilize the newer, let's call them "calibrated" algorithms.

It's one of these simple things that come up in early morning half life, which are usually the most stupid things when you think about them. But yesterday I kind of kept hold of that and actually put it in code already. So that now all existing patches will just use the original algorithms (which I wisely preserved in extra files) while I'm free to go into all those algorithms and correct things where it's itching me.

Like slowriot said, most things are fine like they are. I do not think that there is need for a complete overhaul. Most presets will even probably sound the same. Howie and I have a saying (or a running gag, if you want so) where we'd say "it's in the small details". And this is what we're looking at. A few 3 dB more here, a slightly different scaled knob there. After all, Zebra has been bred over quite some years, and sometimes it turned out that some things might have been better done a little differently. In 2002 it wasn't imaginable that 2^x could be calculated a dozen times per sample. Nowadays SSE and AltiVec are given, so we can. Which e.g. make the necessity for linear gain obsolete in many areas.

Anyhow, we'll see how it pans out in a couple of weeks. I'm optimistic, but if it doesn't work out then it won't happen. Otherwise we'll have a more consistent and more accessible/flexible synth.

Post

I would love to see tweaks to the filters...

However, I voted for waveshapers because it is the area where I feel Zebra is weaker. Whether that is new or improved modes for the existing waveshapers, or a new distortion module, or even both.

I would also like to see the MSEG's get a revamp so they have a mode like the stepper in Massive but still editable if desired. Oh, might be nice to have a few more noise types... again, waveshapers/distortion are a clear first choice for me

Post

OTHER: Chord memory so you can play Zebra like a sampled chord and random arp (meaning an arp plays random notes from the chord you hold)

Bi-directional values for LFO's.

GUI = pure gold with diamond knobs ;)

/Michael
www.xsynth.com - Sound Synthesis with Vintage flavour

Post

Urs wrote: It's one of these simple things that come up in early morning half life, which are usually the most stupid things when you think about them.
Interesting.

If by "half life" you mean half conscious/half dream state after waking up, then you're in good company.

Apparently Thomas Edison trained himself to slip in and out of that state. He managed to be quite creative also...
Urs wrote: Like slowriot said, most things are fine like they are. I do not think that there is need for a complete overhaul.
1. Yes they are :tu:

2. Indeed there's no need for a complete overhaul. Zebra2 is great as it is.
Urs wrote: "it's in the small details". And this is what we're looking at. A few 3 dB more here, a slightly different scaled knob there.
Well, while we are talking infinitessimal details, would it be possible to move that LPvintage2 algo from the bottom of the filter list up so it can live with the rest of the Low Pass filters near the top? 8)

Cheers,
bagginz

Post

Urs wrote:
ganymede wrote:Yes, but would existing sequencer projects load Z2.5 with the correct presets that they were saved with in 2.3? I wouldn't want to upgrade to 2.5 unless my presets were correctly loaded (including modifications) with a project.
But of course. Backwards compatibility has always been a major concern here. That's the beauty of the new solution, an old preset/project will just open with 2.3 in mind, while new presets will utilize the newer, let's call them "calibrated" algorithms.

It's one of these simple things that come up in early morning half life, which are usually the most stupid things when you think about them. But yesterday I kind of kept hold of that and actually put it in code already. So that now all existing patches will just use the original algorithms (which I wisely preserved in extra files) while I'm free to go into all those algorithms and correct things where it's itching me.

Like slowriot said, most things are fine like they are. I do not think that there is need for a complete overhaul. Most presets will even probably sound the same. Howie and I have a saying (or a running gag, if you want so) where we'd say "it's in the small details". And this is what we're looking at. A few 3 dB more here, a slightly different scaled knob there. After all, Zebra has been bred over quite some years, and sometimes it turned out that some things might have been better done a little differently. In 2002 it wasn't imaginable that 2^x could be calculated a dozen times per sample. Nowadays SSE and AltiVec are given, so we can. Which e.g. make the necessity for linear gain obsolete in many areas.

Anyhow, we'll see how it pans out in a couple of weeks. I'm optimistic, but if it doesn't work out then it won't happen. Otherwise we'll have a more consistent and more accessible/flexible synth.
That's great. I'm glad you care about these things more than most other developers! :)

Post

Here's a tweak - in preset view clicking on a module in the grid should automatically switch to the synthesis view.

I've been using Zebra for long enough now that I should know that this doesn't work, yet I still find myself clicking in the grid in preset view and being momentarily confused when nothing happens. I am easily confused though.

Post

hakey wrote:Here's a tweak - in preset view clicking on a module in the grid should automatically switch to the synthesis view.
.
At last, someone else :tu:

So it's not just me who thinks this :D

cheers,
bagginz
Last edited by bagginz on Sat Jul 04, 2009 12:24 pm, edited 1 time in total.

Post

That would be a nice feature! I occasionally find myself annoyed with preset view for the same reason.

My 2 cents about potential tweaks to Zebra, with the respect to Zebra being one of my favourite soft synths :)

New wave-shaping algo's combined with more a aggressive distortion would be superb.

The area of filter resonance in general could certainly do with a few tweaks, XMF or otherwise. It just doesn't seem very "natural" when compared to other synths (hardware or software), it distorts the output too easily with the "wrong" settings. In general, the resonance is what I find hardest to get right. Taming of the XMF ear-piercing would also be welcome, but I don't have a problem with Overload being too loud, there are other ways to turn it down.

But, whatever happens it looks like 2.5 is going to be excellent.

Oh, BTW. Not sure if this has been reported elsewhere but the Zebra interface is absorbing key presses in Logic, so it's not possible to use the space bar to start the transport. Presumably this will be fixed at some point? It's the beta 2.3.3 version.

Peace,
Andy.
... space is the place ...

Post

I voted "Too high Resonance in XMF in high frequencies" because it almost makes the XMF filters unusable for my taste. But your list contains some others than I completely agree with too:

Inconsistent volume of filter algos: This is especially noticable with the bandpass filters on the VCF modules, it's hard to drive a saturator when the level of the signal is so low at the filter.

XMF Overload too loud: This is true, which is why I often use it to compensate for the inconsistent volumes of the filters ;)

Waveshapers lack guts: Well, I have been moaning about this for a while now (as you know), and it is indeed true. With the exception of the Vintage2 filter drive, the saturators and drive algorithms are too mellow. I reach for the Virus TI for aggressive sounds and tend to use Zebra2 for smooth & silky sounds (which it does extremely well). You own a Virus TI don't you Urs? Check out OS3 and the new distortions, they are to die for!

I love Zebra2 heaps though, regardless of these flaws, all synths have their weaknesses.

Keep up the great work Urs!!

Post Reply

Return to “u-he”