Strange Glitching Error When Tracking

[cvdpoel]

Hey guys, I'm experiencing with my live tracking rig and figured I'd make a post to see if anyone can help me out. (I'm not sure if it's a UAD issue or Pro Tools issue so I'm posting this on their forum as well)

When tracking, I'm finding that in Pro Tools I will randomly and periodically have these strange glitches happen where there's a small crackle and then the audio jumps like 3-5 seconds ahead. Those missing 3-5 seconds aren't in the raw audio files and there's no gap in the waveforms either.

This issue only happens on my live/laptop setup and not on my home/desktop setup - the difference between those two setups is that I'm using Thunderbolt (with a Thunderbolt 2 to 3 adapter) on my laptop and Firewire on my desktop.

Here's the rundown of my gear:
Presonus Digimax D8 -> Firewire, ADAT, and Word Clock cables -> UAD Apollo Quad Firewire with Thunderbolt Card -> Thunderbolt 2 Cable -> StarTech Thunderbolt 2 to Thunderbolt 3 Adapter -> Laptop

I'm running the latest version of Pro Tools and my UAD software is up to date as well as Windows and every other software I can think of that would affect this.

Here are the specs for my laptop as well:
Model : Dell G7 7588 GSeries
Serial Number : H6D****
Chassis : Dell Notebook
Mainboard : Dell 0FDMYT
Serial Number : /H6DGLP2/CNC************
BIOS : Dell 1.10.0 03/05/2019
TPM - Trusted Platform Module : INTC Intel 2.0 (PCR 24)
Intel vPro : 12.0.22.1310
Total Memory : 15.81GB SO-DIMM DDR4

Processors
Processor : Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz (6C 12T 4GHz, 3.7GHz IMC, 6x 256kB L2, 9MB L3)
Socket/Slot : FC BGA1440

Chipset
Memory Controller : Dell Core8H (CofeeLake-H 6C) Mobile Host Bridge/DRAM Registers 100MHz, 2x 8GB SO-DIMM DDR4 2.67GHz 128-bit, Integrated Graphics

Memory Module(s)
Memory Module : Hynix (Hyundai) HMA81GS6CJR8N-VK 8GB SO-DIMM DDR4 1Rx8 PC4-21300SO DDR4-2666 (19-19-19-44 4-63-21-7)
Memory Module : Hynix (Hyundai) HMA81GS6CJR8N-VK 8GB SO-DIMM DDR4 1Rx8 PC4-21300SO DDR4-2666 (19-19-19-44 4-63-21-7)

Video System
Monitor/Panel : AU Optronics Generic PnP Monitor (3840x2160, F2TW2€B156ZAN, 15.6")
Video Adapter : NVIDIA GeForce GTX 1060 with Max-Q Design (1280S 10C SM6.5 1.26GHz/1.48GHz, 1.5MB L2, 6GB 7.6GHz/8GHz 192-bit, PCIe 3.0 x16)

Graphics Processor
CUDA : NVIDIA GeForce GTX 1060 with Max-Q Design (1280S 10C SM6.1 1.26GHz/1.48GHz, 1.5MB L2, 6GB 7.6GHz/8GHz 192-bit)
OpenCL : NVIDIA GeForce GTX 1060 with Max-Q Design (1280S 10C SM1.2 1.26GHz/1.48GHz, 1.5MB L2, 6GB 7.6GHz/8GHz 192-bit)
OpenCL : Intel(R) UHD Graphics 630 (192S 24C SM2.1 1.1GHz, 512kB L2, 6.4GB)
D3D 11 : NVIDIA GeForce GTX 1060 with Max-Q Design (1280S 10C SM12.1 1.26GHz/1.48GHz, 1.5MB L2, 6GB 7.6GHz/8GHz 192-bit)
D3D 11 : Intel(R) UHD Graphics 630 (128MB)
OpenGL : GeForce GTX 1060 with Max-Q Design/PCIe/SSE2 (6GB)

Storage Devices
Disk : Seagate ST1000LM035-1RK172 (1TB, SATA600, 2.5", 5400rpm)
Disk : Seagate ST1000LM035-1RK172 (512.1GB, SATA600, 2.5", 5400rpm)

Logical Storage Devices
Storage (D:) : 466GB (NTFS, 4kB) @ Seagate ST1000LM035-1RK172 (1TB, SATA600, 2.5", 5400rpm)
Games (E:) : 466GB (NTFS, 4kB) @ Seagate ST1000LM035-1RK172 (1TB, SATA600, 2.5", 5400rpm)
Windows (C:) : 463GB (NTFS, 4kB) @ Seagate ST1000LM035-1RK172 (512.1GB, SATA600, 2.5", 5400rpm)
Hard Disk : 700MB (NTFS, 4kB) @ Seagate ST1000LM035-1RK172 (512.1GB, SATA600, 2.5", 5400rpm)
Image : 12GB (NTFS, 4kB) @ Seagate ST1000LM035-1RK172 (512.1GB, SATA600, 2.5", 5400rpm)
DELLSUPPORT : 1GB (NTFS, 4kB) @ Seagate ST1000LM035-1RK172 (512.1GB, SATA600, 2.5", 5400rpm)
ESP : 646MB (FAT32, 4kB) @ Seagate ST1000LM035-1RK172 (512.1GB, SATA600, 2.5", 5400rpm)

Peripherals
LPC Hub Controller #1 : Dell HM470 (Cannon Point) LPC/eSPI Controller
LPC Legacy Controller #1 : T1 FE-00
Audio Device : Dell ICH300 (Cannon Point) cAVS
Disk Controller : Dell ICH300 (Cannon Point) Shared SRAM
Disk Controller : Dell ICH8 Mobile SATA RAID Controller
USB Controller #1 : Dell ICH300 (Cannon Point) USB 3.1 xHCI Host Controller
SMBus/i2c Controller #1 : Intel ICH SMBus

Printers and Faxes
Printer : Microsoft Software Printer Driver (300x300, Colour)
Printer : Microsoft XPS Document Writer v4 (600x600, Colour)
Printer : Microsoft Print To PDF (600x600, Colour)
Fax : Microsoft Shared Fax Driver (200x200)
Printer : Brother PC-FAX v.3.2 (A3/LGR) (200x200)
Printer : Brother MFC-J6920DW Printer (600x600, Colour)

Scanners & Cameras
Scanner : Brother MFC-J6920DW LAN (Scanner, 2400x2400, USB)

Biometrics
Voice : Analog NUI Voice Virtual Sensor (Voice)

Network Services
Network Adapter : Bluetooth Device (Personal Area Network) #2 (Ethernet)
Network Adapter : Killer E2400 Gigabit Ethernet Controller (Ethernet)
Wireless Adapter : Intel(R) Wireless-AC 9560 160MHz (802.11ac (VHT), AES-CCMP, 758.35Mbps / 650Mbps)

Power Management
Battery #1 : Samsung SDI DELL W7NKD84 56.00Wh/3.56Ah

Operating System
Windows System : Microsoft Windows 10 Professional 10.0.19041
Platform Compliance : x64

[Darryl Ramm]

Well for starters... thanks for providing such great detail, but the 5,400 rpm HDD are not good. There is no reason to not have all SSD drives here, especially in a laptop that is being used to record on location and is prone to being bumped, dropped etc. Open up that G7 and replace the HDD with two Samsung 860 Evo SSDs or similar.

What actual partition are you recording to?

What is disk cache set to? If "normal".. set it to some value large enough to cache the session... but this still may not help when most of the IO are writes.

... but while a 5400 rpm HDD is awful... and you really need to fix this, it may not explain this problem. I would hope that Pro Tools would fall over with an AAE error if it could not write to the disk fast enough. So you may have more going on here. Crackles are always clues to look at clocking issues. I don't get why you have a Firewire cable mentioned, can't see how that would be used. Can you reproduce this problem with the laptop in the studio? if so you can start experimenting by getting rid of work clock... disconnect as many things as possible, just clock off the incoming ADAT, and see if that improves things. If not get rid of the ADAT box and just test with the Apollo interface itself.