Logic 9.1.4 fix ultrabeat sample attack cut-off? [SOLVED]

I took a break from using ultrabeat awhile ago due to having my own loaded samples having their attacks cut off by about a millisecond, I understand this was a glitch or how ultrabeat was deisgned...just wondering if this has been fixed with the current ultrabeat? as it is really a great drum machine for editing sequences inside it, and I want to switch back to it but need to make sure I don't get my samples cut off again...

Are you sure you can't remove this creep-in behaviour of the envelope by just moving the attack line somewhat right and hard left again? There was a long discussion on this, and although most are convinced Ultrabeat always cuts off the attack, I am not, after I experimented a lot.

Hey!

Attached is a project done in Logic 9.0.2 at 96k sample rate. Sample is a 20kHz sine. Shown are the results of bouncing EXS vs UB. If you want to see if anything's been fixed, mess with the attack of UB's Envelope 4 on the C1 voice to get it faster, then solo UB and bounce using the current locator setting. If you can't get it any better, then it hasn't been fixed.

TestUB_Attack.zip

Yes, the issue seems to have been fixed in 9.1.4.

 

although most are convinced Ultrabeat always cuts off the attack, I am not, after I experimented a lot.

 

Experimenting a lot can be worth nothing if you don't do the ONE experiment that will give you your answer: bounce a sample that has a very strong attack with Ultrabeat and compare the bounced file with the original.

 

There's no need to be "convinced" or "not convinced", just do the test for yourself and compare the results.

 

This was in 9.0.2:

 

Original sample:

 

http://www.logicprohelp.com/files/original_463.png

 

Bounced through Ultrabeat:

 

http://www.logicprohelp.com/files/ultrabeat_104.png

 

Bounced through EXS24:

 

http://www.logicprohelp.com/files/exs24_101.png

 

Are you sure you can't remove this creep-in behaviour of the envelope by just moving the attack line somewhat right and hard left again?

Yes. Again, it's an easy test.

Edited August 2, 2011 by David Nahmani

Yes, 9.1.4 has fixed the bug!!

 

See below, where I ran fader8's test through both 9.1.0 and 9.1.4. The first picture is 9.1.0, second picture is 9.1.4:

Edited August 2, 2011 by David Nahmani

At last!!!

That's good to know. Thanks.

Cool! I was hoping they would fix this. Finally

Good news indeed!

9.1.3 obviously fixed it as well. I wonder why nobody believed me when I wrote my observations. I just tried Fader 8's example project: 9.1.3 does it well with no slow attack. So I recommend 9.1.3 rather than 9.1.4 because of the ES 2 bug (if you are not on Lion).

I wonder why nobody believed me when I wrote my observations.

Which observations? From all the stuff I've read from you earlier in this thread or in the thread you linked, you said you never had that problem with Ultrabeat, or you offer workarounds like adding silence in front of the sample attack, or you say you think samples should always start on a zero crossing anyway.

Edited August 2, 2011 by David Nahmani

I wonder why nobody believed me when I wrote my observations.

What observations?

 

The experiments and observations I described in this thread (also linked to above).

The experiments and observations I described in this thread (also linked to above).

 

As I said in the linked thread, I never really understood your pictures or comments. It's easier to clearly demonstrate something if you use Logic screenshots, or at least screenshots where you can clearly tell where the attack of the sample is, and if you compare an original sample with the same sample bounced of Ultrabeat.

 

Good to know the bug was fixed in 9.1.3 already (I never used 9.1.3 myself).

Edited August 2, 2011 by David Nahmani

I don't see any mention of 9.1.3 fixing the bug, and I don't understand your complex experiments for a simple problem.

 

The point is, there was no problem on my system, the experiments proved this. It was not so complex at all. I Just had generated a waveform that had a sort of marker at its very first period, and there was no slow attack after bouncing. It means the bug had already been fixed then.

there was no problem on my system, the experiments proved this.

When you want to prove that a piece of equipment generates distortion (or does not for that matter), it's common procedure to share screenshots BOTH of the original waveform (before processing) and the resulting one (after processing). Since we're talking about attack distortion, it's also a good idea to make it clear where the attack is.

 

In the linked thread, you mention a square wave, but the screenshots show a square wave where the first period is distorted. You don't show the original sample. And we can't really see exactly where the attack of the sample is.... that was confusing.

there was no problem on my system, the experiments proved this.

When you want to prove that a piece of equipment generates distortion (or does not for that matter), it's common procedure to share screenshots BOTH of the original waveform (before processing) and the resulting one (after processing). Since we're talking about attack distortion, it's also a good idea to make it clear where the attack is.

 

In the linked thread, you mention a square wave, but the screenshots show a square wave where the first period is distorted. You don't show the original sample. And we can't really see exactly where the attack of the sample is.... that was confusing.

 

I apologize for my posts being somewhat inconvenient. The rest was in the text. I thought if I say UB doesn't change the waveform's attack and show you by a picture that the waveform's start is marked by a unique gap (that you call distortion), that would be sufficient.

Ah, something isn't really said if it can't be understood. Sorry again.

Good news thanks. So to be clear (and because I am a beginner with ultrabeat and drum machines), to use my own samples, I simply right click, init sample, drag sample into UB, and by default the entire attack will be played? basically, I don't need to adjust any setting to get the whole attack, it is the default setting?

Good news thanks. So to be clear (and because I am a beginner with ultrabeat and drum machines), to use my own samples, I simply right click, init sample, drag sample into UB, and by default the entire attack will be played? basically, I don't need to adjust any setting to get the whole attack, it is the default setting?

 

I don't exactly understand your right-click procedure, but the important thing is, formerly you simply couldn't make the attack true zero and now you can. Come on, it's not so much work to pull the attack line to the left at all.

One important thing is, when you drag the attack point at very high zoom levels, the numeric value might say "<0.001ms" while the real attack phase is longer - don't trust the display, just drag the line hard left and the slow attack will go away.

Good news thanks. So to be clear (and because I am a beginner with ultrabeat and drum machines), to use my own samples, I simply right click, init sample, drag sample into UB, and by default the entire attack will be played? basically, I don't need to adjust any setting to get the whole attack, it is the default setting?

 

The Init > Sample voice has a default attack of 1.000 ms so you'd have to lower that to hear your full attack.

 

Try the preset "Drag and Drop Samples", which has the lowest attack by default (nothing to do there to get your whole attack to play).

Edited August 4, 2011 by David Nahmani

Come on, it's not so much work to pull the attack line to the left at all.

 

Ok that's fine, but I am a beginner, so where is this attack line I am suppose to be dragging? Which section of ultrabeat? the envelopes? if so, there are 4?

Ok that's fine, but I am a beginner, so where is this attack line I am suppose to be dragging? Which section of ultrabeat? the envelopes? if so, there are 4?

You will notice that "envelope 4 (Amp)" is a hard wired mod source for amplitude. You can still use it as a mod source for other parameters, but it will always control the amp envelope. Strongly suggest you spend some time with the manual. The Ultrabeat section is nicely comprehensive.

where is this attack line I am suppose to be dragging? Which section of ultrabeat? the envelopes? if so, there are 4?

 

Env4, click on the 'A' Button next to 'ZOOM' to zoom in on the attack. You'll see a vertical blue line to the right of the attack ramp up, drag that all the way to the left.

Hi,

 

I think I got it, I attached and circled the verticle blue line I think you're talking about...correct? It appears by default when you init and drag in a sample, it is all the way to the left/<0.001ms.

OK... when choosing Init > Sample, I don't get the same envelope as you... here's what the Attack looks like on mine:

EXCELLENT The EXS has been great but now it looks like time to upgrade! Thank you all so much for all the posts on this topic, you are all epic THANK YOU DAVID!