STATUS: Not a bug
ASSIGNED TO: Ben
VERSION: 1.04, Windows Vista Home Premium SP1, OSX 10.5.6, Logic 8.0.2, Intel Core 2 Duo 3gb Ram, Macbook Pro Coreduo
REPRODUCE: A source file linked below recreates the problem
EXPECTED BEHAVIOUR: Load source, filters behave as normal
OBSERVED BEHAVIOUR: Various filters appear to have inverse behaviour, or generally don't sound right!

ADDITIONAL INFO: Very strange bug.
While messing around with Alchemy, trying to create a pad sound, I managed to "break" the K20 filter with a plain SAW source. I didn't create this source myself, I copied it from a preset called "Come Down" (Pads folder). This source, a simple saw which sounds detuned even though it has only 1 oscillator running (perhaps this is the bug?). When using this source combined with the LP2 K20 filter or LP2 K20 HQ filter, you get very strange behaviour. The filter stops working as a low pass and instead seems to be more of a high pass!

You can download the buggy source, that breaks the filter, here:

Buggy Source

In short, to reproduce: Load the provided file into source A, select LP2 K20 filter for main filter. Boom.. broken.

[EDIT - ZenPunkHippy] Can reproduce in Logic Studio. Notice that most LP/HP filter types behave strangely when this Source file is loaded. Clearing the patch resets the filters to regular behaviour, re-loading and setting filter repeats the problem. More testing in other hosts required.

[REPLY- PaulSC] I don't think there's a bug here. The source you're copying and pasting has FilterMix set to 50%, so part of the signal is going to the second Main Filter (which is probably Off). I suspect several of the analog-modelled filters cause phase shifts, which leads to cancellations when the filtered portion (passing through Main Filter 1) and the unfiltered portion (passing through the inactive Main Filter 2) are mixed together.

Does this explain your results? (Also, should the phase shift be considered a bug, or do you think it's acceptable?)

[REPLY - ZenPunkHippy] A quick surf turned up the following link about the Access Virus filters:

http://www.keyboardmag.com/article/creating-phase- shift/jan-06/17606

So it seems that some filters do exhibit phase shift. I've never noticed it before* with any of the analogue or digital filters that I use, but I can see how it might be useful. Let's see what the devs have to say - feature or bug!

*edit - actually I have noticed it before - in the Fabfilter Volcano plug where you can dial up a (very) small delay offset for each filter producing a similar effect and its a very useful feature.

Ben: PaulSCs explanation is correct here, I believe. The problem is caused because the filter is introducing a phase shift, and you are setting filter mix to 50%.

STATUS: Confirmed
ASSIGNED TO: Ben

VERSION: 1.04, Win / Cubase SL3.1.1, MAC G5 / OS X

REPRODUCE: Check the presets in the zip below...assign a performance knob to source tune, performance knob at zero (no modulation), then compare the sound.

EXPECTED BEHAVIOUR: With the knob at zero it should be as if no modulation is applied.

OBSERVED BEHAVIOUR: Tuning is different, indicating that some modulation is present.

ADDITIONAL INFO:

Zip of presets: Hard-Bell-bug-test.zip

They are identical except that one has performance knobs assigned to the pitch of each source. The performance knobs are set to zero and yet there is a pitch/tone difference between these two presets. I would have assumed that if the knobs were at zero then no modulation would occur, and therefore the presets would sound the same. If the modulation is cleared for the tune knobs, the two presets sound identical again, so there is definitely a change in pitch/tone even though the performance knobs indicate they are at zero. (Note: the tonal difference from source pitch is because these presets are using a keytracked comb filter.)

ADDITIONAL INFO: PaulSC

I can confirm on Mac G5. This is a strange one.

I've cleared and disabled Source B, turned off the Main comb filter, and disabled all the FX, in order to isolate what's going on in Source A.

Performance knob 1 modulates Source A coarse tuning. When comparing (a) the result when the knob value is zero to (b) the result when the modulation assignment is deleted, no difference between (a) and (b) is expected, but I observe that (a) is noticeably LOUDER than (b).
Last edited by ugo on Fri Jan 09, 2009 7:36 pm; edited 2 times in total

STATUS: Confirmed
ASSIGNED TO: Rob

VERSION: OS X (OS: 10.4.11) VST
PROCESSOR: G5, 2 X 2.66 GHz Dual Core, 4 GB 667 Mhz

REPRODUCE: Install Alchemy on a G5 Mac running Tiger

EXPECTED BEHAVIOUR: Both the AU and VST versions are installed

OBSERVED BEHAVIOUR: The VST version is not installed

ADDITIONAL INFO: Followed installation instructions for Alchemy on a Mac G5. After the installation appears to run successfully, only the AU version is available, not the VST. The user has searched all possible locations using Finder but the VST is present.

The user solved the problem by running the installation procedure again and selecting for a Custom install. The check box for "VST" was not ticked, setting this option installed the VST version correctly.

This indicates there must be a difference in the installation scripts for G5 processors, or Tiger when compared to Intel or Leopard installation scripts.

STATUS: Confirmed
ASSIGNED TO: John

VERSION: 1.04, AU, OSX 10.5.6, Logic 8.0.2 (+ Novation Automap Universal 2.1.1234.6)

REPRODUCE: see Additional Info

EXPECTED BEHAVIOUR: Alchemy loads correctly between Logic sessions
OBSERVED BEHAVIOUR: Error message is displayed, Alchemy does not function

ERROR INFO: OS X Error: -10110 telBadProcID: invalid procID

Failed to load Audio Unit, please contact the manufacturer for an updated version or further assistance.
(-10110)

Crash logs from Logic and AlchemyLog.txt sent to support under the title "Logic: Invalid Proc ID Bug". The user may need to be contacted by a developer for more information and has been informed of this. E-mail address is in support e-mail.

ADDITIONAL INFO:
This happened two times now. I can't recall any irregularities in my behaviour. I saved the project under a new name as I quit Logic, shut down the computer. Next time I open logic (autoload last project enabled) I get the error message. I am not 100% shure, but I think I left the Alchemy GUI open.

All other AUs loading properly - it happens to Alchemy only. No Alchemy problems occur under Reaper (VST/AU) or Ableton live lite (AU), didn't try to save and reopen though.

Note In all steps that mention "error" the error ID is the same as above i.e. -10110

STEPS TO ERROR:
- Restart Logic, error
- Reboot machine, starting Logic, error
- Remove from Automap, starting Logic, error
- Quit Logic, reinstall Alchemy, Staring Logic, error
- Add to automap Wrap, starting Logic, error
- Re-scan Audio Unit, click done in AU manager, error

- Tried loading a new empty project template from scratch, save it (no plugins loaded), quit Logic, re-open Logic, load Alchemy Audio Unit (note: automap version only availiable) - now works

- Re-scanned the non Automap version (validation successful, AU not availiable in channel, done, load it in my channel - now works

USER STEPS TO REPRODUCE:
- If I load the non automap version, save the project, reopen it, the error is displayed
- Leaving open the GUI does not seem to make a difference

EDIT: Not so easy to reproduce the error as I thought. Rescanned the non automap version and it becomes available again. Saving, switching projects etc no problem.

- creating a user patch, save the project without saving the patch. reopen project, same error

Note: patches in my other project not available - maybe due to unloading the plugin, but there was no prompt to save the patch

EDIT: definitively reproducible:

- Create a project
- Load alchemy
- Modify the default preset
- Save and quit the project without saving the modifyed default preset
- Re-start logic, error is displayed

NOTE: applying the solution described earlier, switching the a project that is in error via recent files works as long as Logic has not been closed

REPRODUCE:
- Saving the project
- Quit Logic
- Re-open it (autoloading last project with unsaved preset)
- Error is displayed

WORKAROUND:
- Change your default song to an empty project
- Restart Logic
- Load Alchemy in to the empty project (no need to save the default project)>
- Now it is possible to load the project broken by alchemy. The settings should still be there if alchemy has not been unloaded before saving.

EDIT: Saving the preset in the User folder before closing Logic does not help, the error will re-occur

EDIT: The Alchemy GUI has to stay open before switching the project, else Logic will crash. Reproduced this twice. If you interrested in the crash dump let me know.

Hardware Overview:
Model Name:   MacBook
  Model Identifier:    MacBook3,1
  Processor Name:   Intel Core 2 Duo
  Processor Speed:   2.2 GHz
  Number Of Processors:   1
  Total Number Of Cores:   2
  L2 Cache:   4 MB
  Memory:   4 GB
  Bus Speed:   800 MHz
  Boot ROM Version:   MB31.008E.B02
  SMC Version:   1.24f2
  Serial Number:   
  Sudden Motion Sensor:
  State:   Enabled
Audio Interface:
EDIROL UA-25:
  Product ID:   0x0074
  Vendor ID:   0x0582  (Roland Corporation)
  Version:   1.06
  Speed:   Up to 12 Mb/sec
  Manufacturer:   Roland
  Location ID:   0x1d100000
  Current Available (mA):   500
  Current Required (mA):   480
Audio unit Manager scan dump:

NOTE: The following AU validation log is identical to that produced via a working Logic configuration. Some minor differences exist like time to open the plug-in

validating Audio Unit Alchemy by Camel Audio:

  * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
         AU Validation Tool
         Version: 1.2.1b3
          Copyright 2003-2007, Apple, Inc. All Rights Reserved.

         Specify -h (-help) for command options
  * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

--------------------------------------------------
VALIDATING AUDIO UNIT: 'aumu' - 'CaC2' - 'CamA'
--------------------------------------------------
Manufacturer String: Camel Audio
AudioUnit name: Alchemy
Component Info: Alchemy Synth
Component Version: 1.0.4 (0x10004)
Component's Bundle Version: 0.0.0

* * PASS
--------------------------------------------------
TESTING OPEN TIMES:
COLD:
Time to open AudioUnit:         763.780 ms
WARM:
Time to open AudioUnit:         429.374  ms
FIRST TIME:
Time for initialization:        19.384 ms

* * PASS
--------------------------------------------------
VERIFYING DEFAULT SCOPE FORMATS:
Input Scope Bus Configuration:
 Default Bus Count:0

Output Scope Bus Configuration:
 Default Bus Count:1
    Default Format: AudioStreamBasicDescription:  2 ch,  44100 Hz, 'lpcm' (0x00000029) 32-bit little-endian float, deinterleaved

* * PASS
--------------------------------------------------
VERIFYING REQUIRED PROPERTIES:
  VERIFYING PROPERTY: Sample Rate
    PASS
  VERIFYING PROPERTY: Stream Format
    PASS
  VERIFYING PROPERTY: Maximum Frames Per Slice
    PASS
  VERIFYING PROPERTY: Last Render Error
    PASS
  VERIFYING PROPERTY: Instrument Count
    PASS

* * PASS
--------------------------------------------------
VERIFYING RECOMMENDED PROPERTIES:
  VERIFYING PROPERTY: Latency
    PASS

* * PASS
--------------------------------------------------
VERIFYING OPTIONAL PROPERTIES:
  VERIFYING PROPERTY Supported Number of Channels
    PASS
  VERIFYING PROPERTY Bypass Effect
WARNING: GetPropertyInfo (Bypass Effect) returning valid information for scope/element for [Output/0], which should be invalid.
WARNING: GetPropertyInfo (Bypass Effect) returning valid information for scope/element for [Part/0], which should be invalid.
    PASS
  VERIFYING PROPERTY Host Callbacks
    PASS

* * PASS
--------------------------------------------------
VERIFYING SPECIAL PROPERTIES:

VERIFYING CUSTOM UI
Carbon View Components Available: 1
  auvw 00000000 CamA  -  Camel Audio view
    PASS

Cocoa Views Available: 0

HAS FACTORY PRESETS

VERIFYING CLASS INFO
WARNING: Preset name is not retained in retrieved class data
    PASS

TESTING HOST CALLBACKS
    PASS

* * PASS
--------------------------------------------------
PUBLISHED PARAMETER INFO:

# # # 287 Global Scope Parameters:
WARNING: PARAMETER SHOULD PUBLISH NAME AS CFString
.
.
.
WARNING: PARAMETER SHOULD PUBLISH NAME AS CFString
WARNING: PARAMETER SHOULD PUBLISH NAME AS CFString

* * PASS
--------------------------------------------------
FORMAT TESTS:

Reported Channel Capabilities (explicit):
      [0, 2] 

No Input, Output Chans:
0-1   0-2   0-4   0-5   0-6   0-7   0-8
      X                                   

* * PASS
--------------------------------------------------
RENDER TESTS:
Output Format: AudioStreamBasicDescription:  2 ch,  44100 Hz, 'lpcm' (0x00000029) 32-bit little-endian float, deinterleaved
Render Test at 512 frames
Slicing Render Test at 64 frames
  PASS

Render Test at 64 frames, sample rate: 22050 Hz
Render Test at 137 frames, sample rate: 96000 Hz
Render Test at 4096 frames, sample rate: 44100 Hz
Render Test at 4096 frames, sample rate: 192000 Hz
Render Test at 4096 frames, sample rate: 11025 Hz
Render Test at 512 frames, sample rate: 48000 Hz
  PASS

Checking parameter setting
Using AudioUnitSetParameter
Using AudioUnitScheduleParameter
  PASS

* * PASS
--------------------------------------------------
AU VALIDATION SUCCEEDED.
--------------------------------------------------


validation result: successfully validated

updating properties of AU Alchemy by Camel Audio...done.

The following are reported bugs with copy and paste of modulation assignments

======================================
STATUS: Confirmed
ASSIGNED TO: Rob
VERSION 1.04 Mac Live 6.0.10
HOW TO REPRODUCE: Create VA-based preset using only source A. Copy from source A. Paste into source B. Solo source B.
EXPECTED BEHAVIOUR: Should hear exactly the same as before copying source A.
OBSERVED BEHAVIOUR: Source B is silent.

John: I have emailed the preset that I created to Ben and Rob.

Paul: Cannot reproduce v1.04 in Logic 7.1, copy/paste between sources seems fine from both Source All page and individual Source ABCD pages.

=======================================
VERSION 1.04 Mac

STATUS: Confirmed
ASSIGNED TO: Rob
HOW TO REPRODUCE: Apply one or more Arp Sequencers to Arp Tune Coarse knob, then Copy Modulation from that knob and Paste Modulation to a non-Arp knob (such as Source Coarse Tune)
EXPECTED BEHAVIOUR: not sure, probably nothing should get pasted (since Arp Sequencers are Arp-specific)
OBSERVED BEHAVIOUR: other seemingly arbitrary mod sources such as Channel AT and Poly AT get pasted
Last edited by ZenPunkHippy on Wed Jan 21, 2009 4:12 pm; edited 2 times in total

STATUS: Confirmed
ASSIGNED TO: John

VERSION: 1.04, All

HOW TO REPRODUCE: Use the additive editor e.g. follow the example in the manual:

http://www.camelaudio.com/alchemymanual/source/additive-edit or/#Ex_add_from_scratch

The problem usually starts occurring when editing data while swapping back and forth between the amplitude, panning and pitch editor for individual break points and overall data. Using the breakpoint copy and paste function from the menu may also trigger this but, but that is not confirmed. Using the zoom control to first decrease, then increase the level of zoom can also trigger this behaviour.

General user comments about this issue say it is difficult to reproduce reliably but it is definitely happening.

EXPECTED BEHAVIOUR: No additive breakpoints are created without request.
OBSERVED BEHAVIOUR: After a short time (e.g. 10 minutes usage) breakpoints will appear that were not created by the user.


A related issue is the overall performance of the additive editor. It can take up to several seconds to move a parameter. Several users have reported this problem.

ADDITIONAL INFO:

The following are some general observations from user bmanic:

The editor is living a life of it's own. As soon as I try to edit, delete or move a point, it creates 5 - 10 new ones, seemingly randomly. New points are sometimes created after changing pages and modes, with no actual clicking or adding of new nodes take place.

I've been trying to reproduce precisely what is going on but it really feels random, without any logic.

STATUS: Fixed In Update (1.05)
ASSIGNED TO: Rob

VERSION: 1.04, Windows Vista, XP / Live or Reaper, OS X 10.5.6 / Logic 8.0.2 AU

REPRODUCE: Copy a single folder containing more than 256 samples or SFZ files to <Installation Folder>/Alchemy/Samples, open the Alchemy GUI

EXPECTED BEHAVIOUR: The GUI loads and when the user clicks on a Source menu and navigates to the new folder, the contents of the folder are displayed correctly

OBSERVED BEHAVIOUR:
1) Only the first 256 samples in the folder are available from the menu. This must be because an unsigned 8 bit number is used to enumerate the files.
2) If a folder contains more than 127 characters in a path specification the folder will not display any files
3) This is possibly related to this is another bug where the file list is not always loaded correctly. This is based on observation and not repeatable behaviour, but it is definitely occurring - it has happened too many times now. When the bug is triggered, the list of files will only contain a small number of files with no folders, or a single random Factory or User folder and associated files, and this remains in effect until the GUI is closed and reopened.


IMO there seems to be a problem with the file / folder enumeration e.g. pointers or objects are not being maintained correctly. Sometimes this leads to a full crash of the host. Crash reports from Logic and associated Alchemy log files related to this problem have been forwarded to the support e-mail address under the title "Alchemy Many Files Bug".

In my experience populating menus with dynamic data queries e.g. from database, or folder as is the case here, seems like a good idea, but in practice can cause problems. As we are seeing, users will put multi-gigabyte sample collections in this folder and expect it to work, as you can see below.

USER INFORMATION FOLLOWS:

The following link is a screen shot of the Alchemy GUI that is the result of the conditions described by the user, below:

http://i545.photobucket.com/albums/hh392/auricle_photos/Empt yAlchemy.jpg

ADDITIONAL INFO: Toshiba Satellite A200 laptop - Intel Centrino Core2Duo 2GHz, 3Gb Ram, 2 x 250Gb HDD.



Follow up PM:



[ZPH] Fixed in 1.05, OS X 10.5.6, Logic 8.0.2 AU

STATUS: Confirmed
ASSIGNED TO: Rob

VERSION: 1.04 OS X Only / 10.5.6 / 10.4.x

HOW TO REPRODUCE: Close all instances of Alchemy. Use System Preferences -> Security -> File Vault to turn on the file vault. After the files are secured and then supplying the user password for the account - ensuring that the users home folders are visible in the Finder, open an Instance of Alchemy and use the Import Audio browser to navigate to the Users folder.

EXPECTED BEHAVIOUR: Users home folder is visible and accessible. Access to these files and folders should be transparent.

OBSERVED BEHAVIOUR: File Vault encrypted folders in the Users folder are not visible or accessible.

STATUS: Confirmed
ASSIGNED TO: John

VERSION: 1.04 / Windows XP SP 2, Savihost / OS X 10.5.6, Logic 8.0.2 AU, Numerology 2
HOW TO REPRODUCE: In any host load the linked patch.

Before The Gnomes (right click to save)

EXPECTED BEHAVIOUR: The patch plays correctly. In Logic the Copy & Paste functions work as expected.
OBSERVED BEHAVIOUR: Several problems with the linked patches exist, possibly all related:

1) Open a new instance of Alchemy. Use the File -> Load menu to open the patch "Before the Gnomes". Play a few notes, there will be no sound. Enter the editor for Source A and click "Additive". Play some notes, the sound will start up. Now load the first patch in the title bar browser "A Forest Gnome Dancing". Host crashes. Reproduced in Logic and Savihost.

Logic crash logs have been sent to support in an e-mail titled "Crashing Gnomes Bug"

The sound that is heard after entering the Additive editor is not the sound that was saved. The Additive data is being reset to a default sound when the user loads the patch and enters the Additive editor.

This is clearly affecting the Logic AU wrapper Copy & Paste functionality.

The crash report named Logic Pro_2009-01-13-021316_ZenPunkMac.log was produced by causing the host to crash in the above manner.

Further testing shows that the data in the additive editor is not being saved and loaded correctly, or at all in some cases.
Last edited by ZenPunkHippy on Wed Jan 21, 2009 4:17 pm; edited 1 time in total

STATUS: Confirmed
ASSIGNED TO: John

VERSION: 1.04 / XP Pro SP3

HOW TO REPRODUCE: Import a sample via the Source menu, "Load Audio"
Tim: then select ALL for loop mode.

EXPECTED BEHAVIOUR: Loop marker start and end points are located at reasonable places in the loop
OBSERVED BEHAVIOUR: When loading samples via "Load Audio", the start and end loop markers are placed at the end of the sample.

ADDITIONAL INFO: Have requested specific sample file from the user to re-create the problem.

The user has replied with a sample file that illustrates the problem, it is available at the link below. If you import this file using granular or additive modes, the loop points will be set at inappropriate points in the editor i.e. right near the end, rather than encompassing the whole file as other samples do.

The linked zip contains a sample that illustrates this problem when imported in to a source as described.

http://www.zenpunkhippy.mindhacker.co.uk/alchemy/audio/artie dream1.wav.zip

Tim: I also get this, but this only happens if a sample does not have loops already designated. So Alchemy assigns the loop points at the end rather then the whole file, and it is up to the user to designate loop points where they actually want them. Having said this, when selecting 'all' as the loop mode, it should loop the whole file. So this part is a bug.

STATUS: Confirmed
ASSIGNED TO: John

VERSION: v.104 / Mac 10.5.6, Intel CPU, Logic 8.0.2

REPRODUCE: Additive-Import Vocals > Jonni-Hello.wav. Make note of the sound. In Additive Editor, select the second breakpoint. Partial 40 has amplitude -∞. Raise it to -15 dB or so, and notice a high-pitch transient is now present. Then return it to -∞.

EXPECTED BEHAVIOUR: Originally imported sound should be restored.
OBSERVED BEHAVIOUR: Noticeable traces of the high-pitch transient remain.

[ZPH 22-Jan] Following the instructions above reproduced the problem first go. The additional step of using the right-click functionality to return transient 40 to -∞ was not required.

John@Paul: I cannot discern any change in the sound when using Windows or Mac. I think I must be missing part of the process to reproduce it or something. It would be really helpful if you could produce before and after presets that show the problem.

Paul: here's some presets and audio, hope they help.

presets
hello-raw = default additive import
hello-plus40 = amp of partial 40 boosted at breakpoint 2
hello-minus40 = amp of partial 40 restored to zero at breakpoint 2 (should sound same as hello-raw)

AUDIO (in order: hello-raw, hello-plus40, hello-minus40)

John: Thanks for the demos Paul. I can hear what you mean and I can see that it seems to be happening because breakpoint 3 is having its value for partial 40 boosted when it it boosted at breakpoint 2. However I still cannot reproduce the problem myself by following the steps that you did. I think it must be another problem that only happens on PPC.

John@Rob: Could you try to reproduce this on PPC please Rob?

Rob: I tried this on PPC, but the high pitched transient goes away when I bring the partial slider back down to 0.

Paul: Ah, here's an interesting discovery (now using v0.93). The final REPRODUCE step is, "Return it [level of partial 40] to -∞."
- If I do this by dragging the partial level all the way down, then there is no unwanted transient. But if I do it by right-clicking the partial, then the unwanted transient remains. I expect right-clicking a bar should be identical to dragging it down to zero, but it seems there's a difference in this case.

STATUS: Investigating
ASSIGNED TO: John

This bug needs confirmation in v1.04

VERSION: V 0.92 Mac (suspected AU-specific)

REPRODUCE: From a freshly instantiated Alchemy, click Source A Edit, Click ADDITIVE, and then choose Import Audio from the source content field at the top left of the edit window.

EXPECTED BEHAVIOUR: clean sample browser window
OBSERVED BEHAVIOUR: breakpoint and warp marker data are redrawn on top of browser window

John: I could not repeat this on Windows
John@Rob: Could you try this on PPC please? I cannot repeat it on Windows or Mac.

Rob: I wasn't able to reproduce this on PPC.

Paul: More details. I can reproduce this consistently using the AU in Logic 7 and in DSP Quattro. It does not occur using the VST in DSP Quattro. It only occurs if the Additive Editor page is open the very first time you open the sample browser. If you cancel the Import dialog and then re-open, it's drawn cleanly. (For this reason, I don't consider it a very serious bug, although it's clearly undesirable.)

Tim v93 Mac: could not reproduce.
Tim: sent followup 10-dec-08

Pauls report:
Yes, I can repeat this consistently in v0.93. It only occurs if the Additive Editor page is in view when the import browser is opened for the very first time in a session.

STATUS: probably fixed 1.05 (see note 26-Jan)
ASSIGNED TO: Rob

This bug needs confirmation of fix in v1.05 or later, observation during regular testing is the only way this can happen.

VERSION: 1.00 Mac

REPRODUCE: open Alchemy, select sine from the VA menu in source A, then click on right arrow to get to square, then try to select a wav file from Load Audio->Factory meny
EXPECTED BEHAVIOUR: shows subfolder system for samples
OBSERVED BEHAVIOUR: only shows samples from one particular subfolder

Rob: This is a long time bug that I think I'm finally able to reproduce, whereby trying to select a wav only displays sounds from one subfolder. SHould be able to finally fix it now!

Rob: After making some improvements to the file menu system, this bug no longer seems to be occurring.

Col: 1.05 Mac - seems to be fixed
Tim: 1.05 Mac - seems to be fixed, although it did happen once, then could not reproduce it. Clicking into the dialog again fixed it. Win: OK.

[Col] if it happened once, it can't be fixed entirely.

[ZPH 26-Jan] I've seen this bug once while running v1.05. Closing the Source file menu and re-opening it showed the correct details. No crash.
Last edited by tconrardy on Fri Jan 23, 2009 12:50 pm; edited 1 time in total

STATUS: Confirmed
ASSIGNED TO: John

VERSION: 1.01 Win Live7

REPRODUCE: In Add and spectral editor, select from bottom file/import image and look at default path.
EXPECTED BEHAVIOUR: Default path 'should' go to Alchemy library folder
OBSERVED BEHAVIOUR: Path is the Live7 project files. I don't know if this is a bug, but perhaps default paths for file should be specific to Alchemy library.
Last edited by tconrardy on Thu Jan 15, 2009 2:12 pm; edited 1 time in total

STATUS: Confirmed
ASSIGNED TO: Rob

VERSION: 1.01 Win Plogue Bidule / Orion 5

REPRODUCE: Create projects on these hosts and reload them.
EXPECTED BEHAVIOUR: Should see correct preset displayed.
OBSERVED BEHAVIOUR: the preset name shows as 'default' after loading project again, but other than that its correct - it plays the correct preset.