← Feed

Posts

• Post #4503249

  TIL there are vacuum pumps with 3 phase motors on them

• Post #4494585

  One of these days I should commission some artist to do a piece involving some kind of demonic summoning ritual with a pentagram and candles and whatever. But all of the arcane symbols drawn in the circle are distributed element filters, Wilkinson dividers, planar magnetics, etc.

• Post #4478303

  Welp, I need to re-record the ngscopeclient 100baseT1 demo video lol. After some tuning, I can now run the TX *and* RX protocol decodes , and constellation diagrams, in real time. The only thing I can't do is eye patterns of both lanes (that slows it down from 12.5 to around 11.75 Hz refresh rate)

• Post #4476040

  Today I learned that if you try to read a Vulkan SSBO in a compute shader with a binding greater than the max binding index declared when the compute pipeline was created, radv will crash with an unhandled page fault error and take down your entire X session. Debugging that, when my session got wiped every time I ran a buggy build of ngscopeclient while trying to find the problem, was... annoying.

• Post #4442305

  Thinking about traveling demo setups for ngscopeclient and thunderscope. Requirements: * Smaller is better * PCIe gen2 x4 or thunderbolt 3 host port for TS * Decent internal GPU or free pcie slot for an nvidia card Open to laptop, mini pc, SBC, or any other form factor but ideally it should fit in a carryon sized pelican case along with whatever else I'm demoing it attached to. What is the lowest cost/smallest form factor minimum viable setup if i don't care about GPU performance but...

• Post #4428169

  Why is shader optimization so addictive? I have other things to do... But it's 2AM, I'm thoroughly caffeinated, and staring at the Radeon GPU profiler reordering SSBO memory transactions for improved coalescing. And successfully shaved about 1.7 ms off the first shader in the PAM4 edge detector (4.71 -> 2.98 ms, a 58% speedup). The overall filter went from 11.22 to 8.8 ms, a 27.5% speedup. There's definitely still more room to tune, too.

• Post #4411415

  Silly idea: has anyone ever made an embedded device that can run HDLC over serial console for initial provisioning or OOB management or something? Like for when you want the ability to do file transfers and more fancy things than you can do with a simple shell, but don't want to burn the area and power/BOM cost of an ethernet phy

• Post #4410877

  I'm offering my oscilloscope probing class in Europe for the first time! If you're in or near the Netherlands, consider signing up. Space is limited to 10 students to ensure everyone has enough lab time; two spaces are already taken. https://www.eventbrite.com/e/oscilloscope-probing-training-tickets-1996971622881?aff=oddtdtcreator

• Post #4407401

  Squeezed one last optimization in after the v0.2-rc1 release that should land in the final v0.2. Filters now return a bitmask of flags that include the ability to advertise whether they can be appended to an existing command buffer (i.e. they are entirely or primarily GPU based and do not call begin() on the command buffer), and whether they can be appended to (primarily GPU based, and record into the command buffer but do not submit it). The scheduler now tries to identify as many blocks as p...

• Post #4406552

  @ireneista@adhd.irenes.space yep I thought the same

• Post #4399217

  ngscopeclient v0.2-rc1 binaries are now up on GitHub https://github.com/ngscopeclient/scopehal-apps/releases/tag/v0.2-rc1, as well as our local artifact mirror at https://dl1.ngscopeclient.org/ngscopeclient-releases/v0.2-rc1/

• Post #4385735

  The smoke just keeps getting worse. Last reading was 54.8 ug/m^3 PM2.5 which is an AQI of 149 (I had to go back and edit this post several times as I was writing it because the number kept climbing).

• Post #4347436

  Constantly coming across interesting things while making demo setups for showing off ngscopeclient filters. Here, I'm looking at the fall time of a 10 MHz reference clock coming off my Leo Bodnar GPS NTP server. This isn't quite as clean as the 10 MHz off the SRS FS752 GPSDO on the other side of the lab, but the NTP server lives on the main IT rack next to the scope in the VM server that I'm using for this test, so it was a convenient test signal source. I expected it to be constan...

• Post #4346857

  Nothing like making a pass through every feature of a tool writing documentation to find bugs, lol. I've lost count of how many filters I've found breakage in while preparing example screenshots for the v0.2 manual.

• Post #4341238

  @Lunaphied@provably.online @ar@is-a.cat I've seen a lot of people using sata as a cheap transceiver breakout

• Post #4340064

  @jpm@aus.social adding test points on the USB is a good idea though

• Post #4339489

  Making a demo / training board for showing off ngscopeclient and demonstrating probing techniques. It's got a STM32H750XBH6 paired with an XC7A35T-2FGG484I. Overkill compute wise, but the FPGA has transceivers which I needed. The FPGA can synthesize all kinds of digital signals for decoding. Any suggestions on fun analog things that won't cost too much area or BOM?

• Post #4266584

  It looks like I'm going to have some downtime in The Netherlands in late September / early October due to airline pricing (staying an extra two days after the event I'm attending saves $1200 on airfare, cutting the ticket price by more than half and more than justifying the extra meal/hotel cost) Anyone interested in a get-together Monday Sep 28th, Friday Oct 2, or Saturday Oct 3? Or have suggestions on things to do in the greater Gouda / Den Haag / Rotterdam area?

• Post #4259999

  Welp, I did it. FFT *and waterfall* in real time on 4 Gsps of streaming waveform data. This is going in the demo bin for sure.

• Post #4259570

  I need something bigger than a pair of 24" 1080p display on my lab bench. Big GPU profiling traces just don't fit in 1080 scanlines when you are looking at double digit numbers of threads and lots of Vulkan queues. I can't fit any more width, but maybe I'd benefit from like a 30-something inch 4K in portrait mode?

• Post #4255153

  @r@glauca.space yes, not being fluent in mandarin does make a lot of these parts off the table for me

• Post #4254906

  @r@glauca.space Without hard TRNG and AES it's not useful to me for my typical use case of embedded management stuff

• Post #4250526

  Once again daydreaming about a "baby zynq" microcontroller with a STM32H7 class processor and memory subsystem attached to a few kluts of low power FPGA with hard APB/AHB/AXI (depending on how big they want to make it) ports, a bunch of dedicated or muxable GPIOs, etc. ST apparently has an FPGA-esque peripheral in some STM32s but it's *sixteen* LEs which is a joke. You can't do any kind of useful custom peripheral in that. Ideally i would want to be able to implement my own p...

• Post #4249427

  Fun engineering challenge if anyone has ideas: The DisplayPort auxiliary channel appears to reuse opcodes for request and completion packets, and uses generally the same framing. This means that, to a first order, if you have a single packet, or a capture that starts in the middle of a long burst of transactions and don't know if you are at an even or odd packet boundary, it is not trivially possible to unambiguously decode the traffic. Is there a heuristic that you can apply to the first...

• Post #4226078

  @quantumdude836@universeodon.com That I'm not sure, I am not an expert at power systems. I know they're both 3-phase and that multiple topologies exist, but not which topology is used in which particular parts of a given power grid.

• Post #4224650

  Posting some typical north American power infrastructure in response to a discussion earlier today. This pole is in a downtown area and is mixed transmission, distribution, and communications. Top lines with largest spacing are transmission, feeding to a substation the next block over. Probably 100+ kV. The ones on the cross arm are distribution, probably around 10 kV. Both 3 sets are 3 phase.

• Post #4223068

  "His friend said he didn't have a problem with drinking. They were right, he's quite good at it!"

• Post #4222933

  Tired: vibe coding Wired: vibratory compaction ✨🛻

• Post #4213395

  @Lunaphied@provably.online It's probably possible. I don't know how many old sessions out there, other than my test data corpus dating back many years, are impacted though.

• Post #4191884

  Found and fixed a bug that could cause the MockOscilloscope driver in ngscopeclient to report bogus sample rates and memory depths when working with offline sessions. Unfortunately the bad state is persistent if you save a session when it happens, and it had been lurking for quite a long time. There's no permanent data loss, since if you have waveform data in the session you can figure out the actual sample rate and memory depth from that, but you need to manually patch the "rate"...