ORIGINAL: thx1200

I do own Rapture and did test it after hearing Dimension puke. My Rapture was patched to 1.2. It happily pitch bended all over the place with no problems. In fact, Rapture 1.2 in general is incredibly stable when compared to Dimension.

If Rapture 1.0 (and maybe 1.1) DID do it and Rapture 1.2 DOESN'T do it, then it's encouraging that whatever engine fixes they did in Rap may also make it's way into the next Dim patch. Can't wait until Dim "1.2" comes out because Dimension as it is right now is pissing me off. heh.

Yep, me three
Good info
I think DimPro IS up to 1.2, no ?

Been busy and haven't had time to log on! I apologize for the delay. First, I was wrong in my original post. I had my versions backwards. Yes, Dim Pro is at 1.2 and Rapture is actually at 1.1, which is what I have and it works fine as previously stated, so apparently the version numbers do not necessarily indicate which version of the engine (if they even share the same engine codeline - i thought they would actually, but maybe not) is in use. I'm very disappointed to hear that Dim Pro 1.2 still has this problem. It appears there is more wrong with 96/24 than it first seems. Now that Sonar 7.0 is out, maybe Cake can work on outstanding problems in Dim Pro 1.2. It would be great if more people could test this pitch bend problem out. I really am shocked more people do not work in the 96/24 domain. But browsing the forums it seems QUITE a few record at 44/16 and are happy as a clam. Really when I A/B 96 to 44, I can tell quite a bit of difference in definition, especially when chained through multiple layers of processing. But to each their own. It's a shame more people don't work in 96 or this Dim Pro problem would have been taken care eons ago. I mean barfing on pitch bend seems to be something beta testers would catch. It's not exactly rare to hit the pitch wheel, even if don't really use it a whole lot in the final recordings.

For the record, I'm not buying Dim Pro until this problem is fixed. In fact, I'm finding it difficult to even use the Dim LE or just regular Dim from P5 because it's so frustrating to have the damn thing crash.

I may try the ol' phone support, but I'm not sure how far I will get since I'm just an "LE" user and not a pro user.

Very strange because Rapture works flawlessly, esp since I found a workaround for the UAC problem and they are SO similar under the hood (from Cake's literature anyway).

Out of curiosity (doubt this is related) but what sound card hardware is everybody using here? I'm running two Echo Layla 3Gs using the latest driver and i'm not using ASIO in cakewalk, i'm using the native windows driver (not MME) but the name/acronym slips my mind at the moment.

I think I may start a new thread to pique interest from lurkers who might be avoiding this "old" thread for the pitch bend problem. It sure is a lot easier to reproduce and test for than the original distortion problem and who knows, they might be related somehow!

thx1200,

Hello once again!

I found myself nodding in agreement with everything you've said about 96/24, processing. I've been using it for years, plus double-precision processing.

--------------------

I was also quite disappointed in the beta testing of DPro, so much so that I wonder if it ever was beta'd, especially at release 1.2 which acts every bit like a beta, as far as I'm concerned. It does sound very good when it works, though, and the overall concept is just what I was looking for in a sampler/synth combo thingy for playing my generated samples. Now if it would only work at 96/24 as I had foolishly assumed it would (silly me!).

--------------------

On Dimension Pro 1.1 versus 1.2: Although I cannot speak to the pitchbending problem, I have tested 1.1 versus 1.2 for the sfz distortion problem. This problem either appeared at 1.2, or it became significantly worse at 1.2 compared to 1.1.

On sound card: Rendering in software then examining the results (i.e. spectra) without ever running through the drivers at all clearly shows that something is seriously wrong with sfz loading independent of the sound card.

Regards,
Dave Clark

The new pitch thread. Sorry to hijack the SFZ thread! :-) I hope that will be taken care of too!! Even better, I hope they are related.

http://forum.cakewalk.com/tm.asp?m=1159575

Not yet. We're still all patiently waiting at the pitch problem thread for a bug fix for that too. We suspect these two problems are related since they both only occur at 96khz in Dimension. Cakewalk has acknowledged there is a bug, but has of yet not released any sort of workaround or patch for it.

However, as I said at the pitch thread, contacting Cakewalk by phone or email support is the best way to escalate priority. The more people who tell them they want this fixed, the better of a chance this bug fix will occur sooner rather than later.

Cakewalk knows about both of these threads, so include links back to these threads so that they know you are talking about the same issue.

And, of course, be polite! :-)

DaveClark - I apologize for oversimplifying the relationship! Thank you for clarifying.

Also, thank you for the heads up on a sample player that works at 96khz! As much of a Cakewalk apologist and fan that I am, I too am growing frustrated at the pace these MAJOR bugs are being fixed.

The tests were created initially to prove a point that was audible on some systems, but the tests are unfortunately confusing and require many steps and extra software. There is no word from Cakewalk yet why it is reproducible on some systems and not others nor how it will be addressed. Cakewalk has said they are aware of the issue and are actively researching it.

A far simpler way to determine whether or not it is happening on a particular system is to have Dimension Pro or Rapture "internally" generate its own sine waves. Position each sine wave at the same pitch on multiple notes while listening to the timbre while the host is running at 48,000 or 88,200 or 96,000 samples per second. If the described distortion is audible then go further to view what is being heard by additionally recording the output to a wave file and viewing it in a host (or Sound Forge or Audacity).

Here is an example SFZ file that will produce the anomaly on certain systems:

 //-------------------------------------------------------------------------------------------------
 // This SFZ file sets a A 440Hz sine wave reference tone on each A key on the keyboard.
 //
 // When the 440Hz reference sine wave is in a 44100 sample-rate .wav file and when the host controlling
 // the synth is using 44100 sample-rate each region sounds identical and looks identical in a wave editor.
 //
 // On some systems when the 440Hz reference sine wave is in a 44100 sample-rate .wav file but the host 
 // controlling the synth is using a 96000 sample-rate each region sounds slightly different.  The regions
 // triggered on the higher MIDI notes grow progressively more distorted on some systems because
 // new frequencies are being added above the fundamental.  Applying an offline Spectrum Analyzer 
 // (like in Sound Forge) reveals a set of frequencies above the sine-wave fundamental tone and also the
 // "attack" of the note is pitched up initially and quickly slides down to the fundamental tone.
 //-------------------------------------------------------------------------------------------------
 <region>sample=*sine key=21
 <region>sample=*sine key=33
 <region>sample=*sine key=45
 <region>sample=*sine key=57
 <region>sample=*sine key=69
 <region>sample=*sine key=81
 <region>sample=*sine key=93
 <region>sample=*sine key=105
 <region>sample=*sine key=117
 

Launch a softsynth host configured for 96,000 samples per second and load Rapture or Dimension Pro into it. Turn off all elements but the first and reset it so there are no effects or processors applied to the multisamples. Load the above .sfz file into the element in the instrument, play the octave A notes up and down the keyboard. Listen intently at a comfortable volume to the timbre. The fundamental pitch will be the same on every A note on the keyboard, but if the problem is there then a slight sliding into the fundamental pitch will be heard on the attack--a sound best described as a chirp on the attack. Record the output to a .wav file via the host application while playing single monophonic notes on each key then bring that recorded .wav file up in a program like Sound Forge so the waveform can be seen. If the problem exists on the test system then the waveform will show a frequency shift on the attack. Open Sound Forge's spectrum analyzer and if the problem exists there will be additional frequency spikes on the frequency plot above the fundamental frequency.

The problem is only exhibited when the key= parameter is used to move the waveform playback to a different MIDI note number without changing the waveform's pitch. If a multisample file is played without the use of the key= logic then the problem will not be heard.

I'm terribly sorry but you're test is not valid.


If I have to download files, that did not come with the product I paid for to do a bug test, then i expect to find a bug.

wavelab is a fully professional audio software and I could not find any clipping.

I happen to know a few major professional sound designers who use cakewalk products, and all consider rapture to be quite fantastic.

Does it have bugs? Yes. All software does. Can I reproduce the bug you are talking about NO.

Will rapture cause distortion on a wave table, or an SFZ file!! YES OFCOURSE. If you file sample is too big. Its not designed for that. RTFM. I am an intermediate designer. so I can't explain, but if you understand certain things about walsh you know that you're sample rates must be small.

They need to be cut. or something. Rene should be in here explaining this to you. Its simple mathematics. but you can not put so big a sample and do unison detune, you will get a distortion. but that is how the synth is designed.

This is how you intelligently prove your point?

By saying I just misunderstand, and I look foolish.

Somebody disagrees with you, they have to misunderstand, and look foolish!!

Meanwhile, where are all the pro sound designers I know using rapture. Not anywhere near this thread.

Cakewalk has acknowledged there is a bug, but has of yet not released any sort of workaround or patch for it.

Hey Tim (and Dave). Fascinating thread. Just received Sonar7 SE with Dim Pro upgrade, and have been boning up on synth programming in general, and SFZ in particular. Sorry I wasn't around with the tools in hand to help work this with you guys back when it came up. Looks like it was quite a party.

Anyway, I'm waiting for a new, faster PC and E-mu 1820m interface to show up before I install all this garb... uh, I mean software, at which point I may be wanting to move up to 96kHz, so this could eventually be an issue for me. So I was just curious what form this "acknowlegement" from Cakewalk took, and how I should refer to it in any inquiries I might make. I might have missed it, but I didn't see a CWBRN mentioned anywhere.

Greetings all,

Unfortunately, René's description of what happens when you contact Tech Support about this problem, complete with links to a description of the problem, is simply NOT TRUE. Been there, done that.

Here are a couple of exchanges about this particular problem to show how Cakewalk actually handled this problem which one may compare to René's imagined scenario:

--------------------------------

Hello David,

Thank you for contacting Cakewalk Technical Support. I apologize for
the delay. The sound you're hearing is the tone going through the ADSR
envelope. Try adjusting the attack curve under the AMP section of the
Modulators in Dimension Pro.


I hope that helps.

***Please do not reply to this email. All replies must be submitted
here http://www.cakewalk.com/Support/EMAIL.ASP

Thank you.

Steve Cox
Cakewalk Technical Support
www.cakewalk.com

------------------------------------

This particular sound has no ADSR envelope activated. This should have been obvious from the files I made available.



Second try:

Hello David,

Thank you for contacting Cakewalk Technical Support. We are working on
an official patch release for Dimension Pro that may or may not
address this issue. In the mean time, please try this Beta fix. You
can download the updated file here:
http:// (snip)

Please replace or overwrite the existing .dll or .vst file in your
main VST folder.

Let me know if that works for you


***Please do not reply to this email. All replies must be submitted
here http://www.cakewalk.com/Support/EMAIL.ASP

Thank you.

Steve Cox
Cakewalk Technical Support
www.cakewalk.com

------------------------------

In other words: Here, try this and see if it works. Of course it didn't because this particular release was aimed at a completely different problem.


Based on my experience in working with computational physics software from a small company, I would hazard a guess that there is basically a wall between Tech Support and the developers who are unfortunately for everyone treated like some sort of godlike being. Such a wall prevents real communication, so you can't believe anything anyone says about reproducibility of problems, status of problems, or anything like that. It's like dealing with two different companies that don't talk with each other.

There is hope, I suppose, given that Cakewalk finally reproduced the problem in Dimension LE. I think it's funny that René seems to believe that OK, it actually did appear, but only just recently --- ????


Regards,
Dave Clark


P.S. Sadly:

René has also accused someone or more than one person of having reported back on this problem after disguising their identity. I can state categorically that I myself DID NOT DO THIS. Given that I started this thread which automatically associates my name with this accusation, René should not leave this potentially libelous situation in its current state. Either retract the accusation or provide more details. Assuring everyone that it wasn't me should be sufficient because no other name is so prominently connected with this problem and this thread. Thank you.

Hi Bob,

Thanks for your recent post here.

Let me see if I understand this correctly. Here is what René posted:

"I would ask everyone really experiencing this issue, and wanting to help us getting a fix, to contact Technical Support and write a detailed report. It is probable that, if you're willing to do it, you'll be asked to sign an NDA form and then probably they will deliver you a testing build."

So by steps:

1) Submit problem report.

2) Receive NDA.

3) Sign and send back NDA.

4) Receive test build of program or subcomponent.


Is this what you are saying happens to everyone but me when they contact Tech Support?

I could understand that this may be true for beta testers, but surely it would be suicide to do this in general. Frankly, based on other questions/problems I've submitted, I believe that the experience I quoted is very typical.

-----------------------

Again, I do thank you very much for your participation in this thread. As described above, I've found a workaround for my particular problem; in addition I'm about to acquire a retail version of Wusikstation 4 which does work at 96/24, so I really don't care about Dimension Pro any more. I would like to help those folks who are having problems, but I can't afford to waste a bunch more time with Cakewalk by starting all over again at Square One as you suggested.

Although it may not show here, I actually have a lot of experience troubleshooting very complex systems, so I know what you are talking about with regards to email. The problem is that they have just arrived at the duplication step after six months or so. That's terrible and doesn't have anything to do with email. It's a process problem (probably more than one) that is undoubtedly compounded by their internal structure.

Well, I certainly did try....


Regards,
Dave Clark

I started following this thread because it sounded (at the beginning) like there might be a real (if minor) issue - it's obvious now that there isn't (at least not with the software). The only reason I'm waisting any time to respond is for the benefit of those who may not be familiar with Cakewalk or their software.

I've been doing computer/music since the early eighties and have had more than my share of experience with bad software and companies. The guys and gals at Cakewalk are certainly not the devious unsupportive jerks you've tried to make them out to be, and their software isn't flawed. If it wasn't the best, I wouldn't still be using it – and if there's ever a (real) problem, it's always been addressed in a timely manner. I think Rene's response was right on (and more than you deserved).

"..I'm about to acquire a retail version of Wusikstation 4 which does work at 96/24, so I really don't care about Dimension Pro any more."

Great, good luck and don't let the door hit you in the head.

P.S. In case you aren't really leaving, I'll graciously allow you to have the last word:

ORIGINAL: DaveClark

Is this what you are saying happens to everyone but me when they contact Tech Support?

I could understand that this may be true for beta testers, but surely it would be suicide to do this in general. Frankly, based on other questions/problems I've submitted, I believe that the experience I quoted is very typical.

Sorry Dave, I'm not singling you out. I'm just pointing out that Cakewalk has a standard procedure for handling support issues and isolating bugs, Cakewalk support handles a large volume of support issues on a large number of products, and sometimes individual customer's don't feel their issue is handled properly, and sometimes Cakewake may actually mess up in their support. However, I don't believe the screwups or mishandling of a customer or his issue is the norm. That is why I suggested you carry on via telephone rather than email to help them help you better. I can't speak for their internal structuring and I just won't comment on something like that--I do believe them when they say they haven't reproduced what you've described, and I do believe you when you have shown it consistently to be an issue on systems you've meticulously tested. I personally don't think its worth giving up on although I do understand the frustration of waiting.

Hi all. I just wanted to clarify something which seems to have been misinterpreted in my previous post (sorry to quote myself):

I would ask everyone really experiencing this issue, and wanting to help us getting a fix, to contact Technical Support and write a detailed report. It is probable that, if you're willing to do it, you'll be asked to sign an NDA form and then probably they will deliver you a testing build.

I'd like to highlight the "...it is probable that...", and "...probably they will..." snippets. I intended to express that this is *not* what will happen to everyone, but to folks who Tech Support thinks might get any benefit of the procedure. I apologize for my English, which is well known to be suboptimal.


Once again, we're looking at this issue closely. Unfortunately (or fortunately?) the issue affects a very small fraction of customers, which makes replication harder.