1.7.07 locks up computer - Win11 Pro / 13900KS / 64GB RAM

[chaosdsm (Shawn)]

PC is a mid level video editing / content creation machine - complete PC spec’s at bottom of post. Laser is not attached, I run laser from MicroSD cards. I have been working on 1.7.04 for quite a while now, everything works great in it for me.

Having ignored the 1.7.06 notifications for a while, when I got the notification that 1.7.07 was available, I decided to download & install.

After installing, opening & creating 3 new projects, the PC started having problems when I tried to save g-code to MicroSD. I thought I had successfully saved the 3 new projects to the card, however, when I tried to safely eject, Windows said that the card was in use. 5 minutes later, I tried again & the PC was frozen. After trying several things, I was able to get the start menu open, and chose to power down & restart PC. Almost 3 hours later, the PC finally shut down & rebooted.

Thinking it had to be a hardware problem, I ran 2 days worth of diagnostics & benchmarks, but EVERYTHING showed the PC to be in perfect working order.

Ran 1.7.07 again for another new project, & more problems, but this time I was able to shut it down via Task Manager > End Task.

Reboot, and same problem, which mostly SEEMS to be linked to “Save GCode”

SO, I uninstalled 1.7.07, rebooted, and re-installed 1.7.07, and still same problem.

Finally, I again uninstalled 1.7.07, rebooted, but re-installed 1.7.04 this time. I rebuilt the 3 projects from scratch, plus 2 new ones, saved GCode to card, and no issues at all with 1.7.04, so it seems to be a problem in 1.7.07.

Guess I’ll stick to 1.7.04 for a while.

PC Spec’s: Gigabyte Z690 Gaming X DDR4 motherboard * Intel i9-13900KS CPU with Kraken X63 AIO * WIN 11 PRO 64bit * MSI 4070 Ti Gaming X Trio GPU * Corsair Vengeance 64GB(4x16GB) DDR4-3200 RAM * Two Solidigm P44 Pro 2TB M.2 NVMe SSD (one OS / one backup) + Intel 665p 1TB M.2 NVMe (storage) + Samsung 990 Pro 2TB M.2 NVMe (storage) + Samsung 980 1TB M.2 NVMe (Ubuntu) + Samsung 990 Pro 4TB M.2 NVMe (currently storage, will be OS drive in next refresh) * Seasonic Vertex GX-1200 1200W ATX 3.0 PSU * Samsung 50 inch QN90A 4K TV/monitor * FANS: Bitfenix Spectre Pro 200mm front intake + 2x Noctua NF-A14 Industrial PPC 140mm exhaust on Kraken X63 radiator + 1 Noctua NF-A12 120mm rear intake

[fins]

Does it also occur when you save to the machine and then copy to the SD card? Have you tried a different SD card? I appreciate that it works fine on the previous version, but just confirm it was the same card in that instance.

[chaosdsm (Shawn)]

Should have mentioned that, I have 4 cards that I rotate between 3 different card readers. SanDisk Industrial 8GB MicroSD (4pk) with Kingston and SanDisk MicroSD readers and a SanDisk SD card reader with MicroSD adapter

The other two cards are out in my shop & are identical to the two pictured. As part of my hardware testing, I tested with all 4 cards and all 3 readers. All cards had the same behavior in 1.7.07. I did not try save GCode to PC.

All 4 cards are operating close to spec at 36-37MB/sec write & 94-97MB/sec read regardless of which reader is used.

[fins]

Certainly comprehensive! To me, this sounds like the behaviour that an asynchronous file write going wrong would produce. Do you use drive or file system or performance tweaks?

[Ganitta28 (Roswell)]

Put your SD card in your PC without adapter, then format it in low level (slow mode) and try if it works better, you can also try a repair of the card with the command chkdsk don’t get the wrong drive!
I put you a link on the procedure.
SD cards wear out over time just like dd and ssd

CHKDSK

Comment exécuter CHKDSK sous Windows 10 ?

[fins]

It’s pretty clear it’s neither the card, nor the reader, given 4 different cards, 3 different readers.

[chaosdsm (Shawn)]

@fins Only performance tweak on my PC (other than water cooling the CPU) is having my Windows paging file on my secondary 2TB Solidigm P44 Pro NVMe M.2 storage drive.

@Ganitta28 the cards & card readers are definitely not the issue.

Vs the few benchmark tests I’ve found online for these cards, my 4 cards are performing much better, with the online results I see, being in the range of 24-28MB/sec write & 78-84MB/sec read - while all 4 of my cards are averaging about 36MB/sec write & 96MB/sec read with peak speeds reaching 37MB/sec & 97MB/sec.

For real-world benchmark, I transferred 6,458,806,272 bytes (6.015GB) of data (about 5GB of video clips, rest were some photos, documents, and LightBurn projects) to the memory card, this took exactly 3 minutes on the slowest of the 4 cards, which is an average write speed of 35,882,257 bytes per sec, or 34.22MB/sec - where 1,048,576 bytes = 1MB or 36MB/sec if your a memory card manufacturer who goes by the calcuation of 1,000,000 bytes = 1MB… the fastest card did it in 2min 56sec.

[fins]

Try the new 2.0.0-RC4 release; there was something in the patch notes that could possibly relate to your issue. In Linux land, it would be symptomatic of a failed sync, which I’d debug with lsof/strace etc. Do you use any AV other than windows defender? I can see an antivirus hooking a write call and failing in this manner.

[Ganitta28 (Roswell)]

Without mastery power is nothing!
Translation is often misleading!
I have a simple nuc i3, usb3 and an nvne, same stats for 50 euros
Chances are that it doesn’t come from the pc but from the user.
Surely the antivirus or the indexing engine of windows or less likely your sd is not configured for a quick deletion.
I never use an SD card reader, this leads to the use of an additional cable and necessarily adds connection points on the usb, the usb is not perfect far from it, like wifi it has packet losses (bits) which are partly due to the oxidation of the contacts.
Kind regards

[chaosdsm (Shawn)]

AV is Bitdefender, it’s set to ask before scan of any add-on devices, but real-time scan of my internet connection.

@Ganitta28 ??? You’re hugely mistaken… or just trolling… I’m not 100% sure yet… but if you’re saying a nuc with i3 is equivalent to my PC then I’m 99% sure it’s trolling.

Anyway, there’s no additional cables only a card reader that plugs directly into motherboard USB slot. Every USB slot, cable, reader, etc… that I’ve seen within the past 10 years and/or currently own have 24k gold plated contacts, so no oxidation or corrosion without exposure to acids.

Packet loss is virtually non-existent over USB these days, though most losses that do still occur are mostly attributable to cheaply made memory cards / flash drives / USB cables > faulty cables > RF interference > damaged connectors > bad drivers > software issues.

All that said, there’s always the occasional bad apple no matter the quality level of the product… which is why I spent 2 days testing hardware & software (with AV disabled & internet disconnected for nearly all of the testing) to determine that the problem was indeed LightBurn & not anything else. No other writes to the cards from anything else created the issue, only writes from LightBurn and only from version 1.7.07.

[Ganitta28 (Roswell)]

I just said that I had the same transfer rates, nothing more!
The thickness of the gold plating varies from 2 to 50 μm, it doesn’t take much connection/deconexion of usb to ruin this layer and let the copper-plated PCB appear, it’s like for emi, the pcs are shielded etc but when you ground the system, the lasers gain 90% stability.
FYI antivirus always has a task in the background, such as proactive searches etc.
Moreover, without modifying the Windows kernel on the administrator command line, it is impossible to permanently disable the windows defender antivirus and its spyware scan, it comes back with each update, new microsoft protection strategy.
best regard.