Gain plugin bug? Writes distortion to start of first processed clip

[iain_m]

Hi all

Please check out this PT10 session created with 10.3.5 on Windows 7. It's 44.1kHz/32bit:
https://dl.dropboxusercontent.com/u/...ain%20Test.zip

The session contains two clips (they reference the same audio file, but that's not important to this bug, AFAIK). When you open it, the audiosuite Gain plugin will appear, with both clips selected for processing.

Render the Gain plugin with the visible settings (Entire Selection, Create Individual Files). Now turn down your speakers and play the first audio file. On my system, high-level distortion is written to the first 0.3seconds of the first clip. This is shown in the attached screenshot.

Only the first clip is affected. It happens to whichever is first on the timeline. It does not happen when selecting the clips individually for processing.

I'd be surprised if this is system-specific, as it looks like a clear plugin bug, but am willing to troubleshoot my system thoroughly if nobody else can reproduce the issue. For the moment I'd just be grateful if you could check out the session and report the results.

Thanks.

[iain_m]

Ok, update:

I completely uninstalled PT 10.3.5 and then installed 10.3.2.

I got the same results with the 32-bit test session (above) when following the same steps in 10.3.2 - if anything, the distorted part of the rendered audio was slightly longer.

I then tried a few other plugins (in the same session, in 'create individual files' and 'entire selection' mode). EQ3, BF-76, Dyn3 De-Ess, and Maxim work without issues. But the Channel Strip, Dyn3 Comp/Limiter, Dyn3 Expander/Gate, and Ping-Pong Delay all write similar distortion to the start of the first file on the timeline. WTF?

And the attachment shows what happens to the start of the file if I resave the session and audio as 24-bit, then try the Gain plugin.

[iain_m]

So I tried this again in a fresh installation of Windows 7 SP1 64-bit, with PT 10.3.5 (Sandra report below - NB that for testing, the system partition was Hard Disk (C:) and the session was on Data (G:). Aero was disabled and processor scheduling set to background services).

Nothing was installed except the iLok driver and client helper, Pro Tools, and RME Babyface driver (1.30).

I used the Gain test session mentioned above and got EXACTLY the same distortion at the start of the processed audio.

This seems clearly to be a bug. Is there any way to report it to Avid without having to pay for a support code?

Computer
Model : Samsung 700G7C SAMSUNG SENS
Chassis : Samsung Laptop
Mainboard : Samsung SAMSUNG_NP1234567890
BIOS : AMI (OEM) P03AAB.082.120330.DK 03/30/2012
Intel vPro : 8.00.04.1441
Total Memory : 16GB DDR3 SO-DIMM

Processors
Processor : Intel(R) Core(TM) i7-3610QM CPU @ 2.30GHz (4C 8T 3.3GHz, 3.3GHz IMC, 4x 256kB L2, 6MB L3)
Socket/Slot : FC PGA988

Chipset
Memory Controller : Samsung Core (Ivy Bridge) Mobile DRAM Controller 100MHz, 2x 8GB DDR3 SO-DIMM 1.6GHz 128-bit

Memory Module(s)
Memory Module : Samsung M471B5273DH0-CK0 4GB DDR3 SO-DIMM PC3-12800S DDR3-1600 (11-11-11-29 5-40-13-6)
Memory Module : Samsung M471B5273DH0-CK0 4GB DDR3 SO-DIMM PC3-12800S DDR3-1600 (11-11-11-29 5-40-13-6)
Memory Module : Samsung M471B5273DH0-CK0 4GB DDR3 SO-DIMM PC3-12800S DDR3-1600 (11-11-11-29 5-40-13-6)
Memory Module : Samsung M471B5273DH0-CK0 4GB DDR3 SO-DIMM PC3-12800S DDR3-1600 (11-11-11-29 5-40-13-6)

Video System
Monitor/Panel : Samsung 173HT02-C01
(1920x1080, 23.9")
Video Adapter : NVIDIA GeForce GTX 675M (8CU 384SP SM5.0 1.24GHz, 2GB 3GHz, PCIe 2.00 x16)

Graphics Processor
OpenCL GP Processor : NVIDIA GeForce GTX 675M (384SP 8C 1.24GHz, 2GB 3GHz)
Compute Shader Processor : NVIDIA GeForce GTX 675M (384SP 8C 1.24GHz, 2GB 3GHz)

Storage Devices
Hitachi HTS727575A9E364 (750.2GB, SATA300, 2.5", 7200rpm, 16MB Cache) : 699GB (F:) (G:)
SanDisk iSSD P4 8GB (8GB, SATA300, 1.8", SSD) : 7GB
Hitachi HTS727575A9E364 (750.2GB, SATA300, 2.5", 7200rpm, 16MB Cache) : 699GB (C:) (D:)
TSSTcorpBDDVDW SN-506BB (SATA, BD-RE, DVD+-RW, CD-RW) : N/A (E:)

Logical Storage Devices
SYSTEM (F:) : 100MB (NTFS) @ Hitachi HTS727575A9E364 (750.2GB, SATA300, 2.5", 7200rpm, 16MB Cache)
System Reserved : 100MB (NTFS)
Data (G:) : 675GB (NTFS) @ Hitachi HTS727575A9E364 (750.2GB, SATA300, 2.5", 7200rpm, 16MB Cache)
Windows 7 (D:) : 650GB (NTFS) @ Hitachi HTS727575A9E364 (750.2GB, SATA300, 2.5", 7200rpm, 16MB Cache)
Hard Disk (C:) : 49GB (NTFS) @ Hitachi HTS727575A9E364 (750.2GB, SATA300, 2.5", 7200rpm, 16MB Cache)
SAMSUNG_REC : 24GB (NTFS)
Optical Drive (E:) : N/A @ TSSTcorpBDDVDW SN-506BB (SATA, BD-RE, DVD+-RW, CD-RW)

Peripherals
LPC Hub Controller 1 : Samsung Panther Point LPC Controller
Audio Device : Samsung Panther Point High Definition Audio Controller
Audio Codec : Realtek Semi High Definition Audio
Audio Device : Samsung GF110 High Definition Audio Controller
Audio Codec : nVidia 0016
Audio Codec : nVidia 0016
Audio Codec : nVidia 0016
Disk Controller : Samsung Panther Point Mobile 6 port SATA AHCI Controller
Disk Controller : Samsung PCIE CardReader
USB Controller 1 : Samsung Panther Point USB xHCI Host Controller
USB Controller 2 : Samsung Panther Point USB Enhanced Host Controller #2
USB Controller 3 : Samsung Panther Point USB Enhanced Host Controller #1
SMBus/i2c Controller 1 : Intel ICH SMBus

Printers and Faxes
Printer : Microsoft XPS Document Writer (600x600, Colour)
Fax : Microsoft Shared Fax Driver (200x200)

Network Services
Network Adapter : Bluetooth Device (Personal Area Network) (Ethernet)
Network Adapter : Realtek PCIe GBE Family Controller (Ethernet)
Wireless Adapter : Intel(R) Centrino(R) Advanced-N 6235

Power Management
Battery No 1 : SAMSUNG Electronics 89.09Wh/5.19Ah

Operating System
Windows System : Microsoft Windows 7 Home 6.01.7601 (Service Pack 1)
Platform Compliance : x64

[iain_m]

Has anyone had a chance to try the test session yet? It will take 30 seconds to check out.

I'm really struggling with this bug: it seems incredible that such a commonplace operation could create such an obviously bad/broken result -- which would make me think it's my system -- yet it's surprising that I'm seeing the bug on a fresh install of Windows and Pro Tools.

Would be grateful if any of you folks could try the session and let me know what happens!

[jasonthurley]

I will test this when I get a moment but will you let me know if clip gain does this?? If you just use the new clip gin slider on the clip itself instead of processing with gain plugin?

[jasonthurley]

We just tested gain plugin and no distortion on any clips we tried....

[jasonthurley]

Does this happen on every session or just this one?

[iain_m]

Hi Jason

Thanks for checking this out.

I can confirm that I don't see the problem when using clip gain. The problem seems to be specific to the process of creating a new audio file via audiosuite processing.

(The reason I was using the gain plugin in the first place was to print fades, as described here.)

Unfortunately I am seeing this in every session I've tried.

Could I ask whether you used the session I uploaded, or tried it in a new session of your own?

Thanks again.

[jasonthurley]

We used our own sessions, hence why I asked if it is in all your sessions or just this one.... Did not process to new track though... I will test that in a bit.

[jasonthurley]

I just read your fade issue... I dont understand why when you consolidate the file the fade isnt consolidated... Im pretty sure when I process a clip and have the fade selected it processes the fade into the clip....

Maybe this is different in 10 as I use 9 at home but 10 here at work and am not always hands on as Im a systems engineer.