After my last post about OpenStep 4.2 on my ThinkPad 560E, one of the dangling TODO items was figuring out why the network setup that was made of all known-working parts didn’t work. I’ve now figured it out.
[TL:DR: The OpenStep 4.2 Etherlink III driver does a bad job with PnP and Media Autodetect on 3C589 cards. To make one work you will probably need to use 3Com’s DOS configuration utility to configure the EEPROM in the card.]
Continue reading
I’ve now tried out a few of the things I bought the Thinkpad 506E I posted about a while ago to try, and there are some interesting notes to share before the semester gets underway and I run out of energy again.
I decided amid some resultant discussions (Hi HN!) from my last post on this machine to give myself a conduct of not physically opening the hardware while I play with it, unless the HDD dies or the like. Doing so is making me exercise some long-dormant skills, which has been extremely fun.
So far I’ve amassed a pile of compatible accessories, booted into NetBSD and imaged off the original HDD contents, installed OpenStep 4.2, fixed the drivers, updated to Patch4, and very briefly taken it online before an irritating networking problem arose. I’ve also run into a problem getting RhapsodyDR2 on, which will likely be the next time-sink.
Continue reading
I’ve been running a TP-Link Archer C7 flashed with OpenWRT at home since early 2016 (and a TP-Link 1043ND with OpenWRT for years before that), but since I moved into my current place over the summer it has been falling over every couple weeks. It hasn’t been logging anything (I have a flash drive mounted that it persistent logs to) but goes down until hard reset, most likely just because of the load of two heavy stream/video-conference/file-sync users (…and probably not because of my kitten chewing on the antennas. Probably.) Rather than updating/diagnosing I decided that was a good excuse for a new faster router.
TL;DR: The Asus RT-ACRH13 is an excellent current-production OpenWRT host for ~$65 with only minor install challenges, and represents a significant upgrade over the Archer C7.
Continue reading
A not long ago there was some noise in places I follow about Zynq FPGA boards surplussed from their role as controllers in retired cryptocurrency mining rigs, for way less than the price of even the bare FPGA SoC. I impulse bought one EBAZ4205 from “College Shop Store” on Aliexpress for $19.08 shipped to try them out, since it seems to to be the most common and documented flavor, and it showed up yesterday. Short version: they look awesome for the price.
Continue reading
Since I moved in to my current house in May, I’ve been thinking about putting a (n additional) workbench in front of the inscrutable 7ft wide projection screen (?) thing some previous tenant built in the basement. I sketched out what I wanted early summer, and drew a proper design and cut-list in September – but by then the grueling part of the break-less fall semester had started, so it never got to the top of my list. This week I finally got around to picking up the parts and building it. This has been a rate-limiting step for other projects, since it was partly needed as a home for my CNC router.
Continue readingAs of about a week ago, I’ve apparently passed my Qualifying Exam research proposal, which was my last hurdle to PhD candidacy.
My PhD project is on TDCI (Time Domain Continuous Imaging) – an alternative imaging technology that folks in the research group I’ve been working with for almost 15 years now have been building, with my help, for the last 6 years. The basic premise is that digital sensors are not, in fact, re-settable film, and we should leverage them accordingly. By rough concept, TDCI capture is like recording the waveform of incident light from each sensel of the sensor, then computationally exposing that data into an image after the fact, to maximize information capture and so that the sensor and shutter behavior can be tweaked after the fact.
The specific deliverables I carved out as a PhD project were building a high-quality capture device (By hacking a mirrorless body), adding non-uniform exposure behavior to TDCI integration (specifying functions and regions for integration), and building a decent user-facing tool for rendering TDCI images from TDCI captures (A DSL + a GUI to specify common options) – because these were some fun tool-building to carve out of the larger research needs.
The deck with notes that I used are uploaded here. Some of the notes are …not entirely proper… because they were second-screen things for my consumption.
It’s a slightly odd situation because I’ve been working things that feed into this project for 6 years (and 11 papers), and technically started UK’s CS PhD program in 2012 (admittedly, I just signed up so I could keep taking classes while I finished my MS work, finishing the degree is a little “eh, might as well”), but it looks like I’m getting away with it. Also odd, thanks to the COVID-19 situation, it was done via zoom, in pajama pants.
The only points in the presentation I did things I was immediately not happy about were:
Plus I have the concern that Rafi (my Co-Chair) soft-balled the hell out of me and pitched a bunch of questions I’m reasonably sure he knew I had prepared spiels for from my materials. That’s not his usual behavior by reputation or observation, and it makes me slightly paranoid.
Now to finish out the deliverables and actually become Dr.
I got one of the free-with-YouTube-Premium Stadia Premiere kit + Pro trials just out of curiosity (since I’m waiting to cancel until Play Music actually stops working, and why say no to $100 of free toys), and after playing for it for an evening, while I’m very technically impressed … I’m completely baffled as to why anyone would pay for this thing, or especially “buy” individual games on it.
It is fast and surprisingly responsive, and the (insane) distributed “phone or computer + controller + Chomecast all talk to the internet and also to Bluetooth and manage to stay in sync” wizardry is an amazing technical achievement, as is the low-latency, reasonably low-artifact streaming.
…but the subscription/rental library is tiny, expensive, and non-portable. It sucks a massive amount of bandwidth (it seemed to be holding at about 20-25Mbit/s down during my AAA test). The fact that I need a minimum of three independent devices (since you can’t do most of the configuration or library management with the controller + Chromecast) to play on the TV is awkward, and the layers of account management and device syncing are pretty wonky for a single user, I can’t imagine dealing with it in a multi-user household.
I played a little bit of Celeste (as an input lag test; it was not half bad, though playing it on a controller is not my preference), a little bit of Hitman (To see how heavy duty graphical games would do), and a little The Gardens Between (hadn’t played it, in my Pro trial, looked neat) – and they worked, but nothing about the experience was particularly compelling.
The whole free-kit-for-Premium-subscribers thing feels like a desperate attempt to dump their hardware stock to build enough user base to recoup the back-end costs for another doomed Google product that will die as soon as the current back end ages out and the workforce moves on to career-advancing shiny new things – after Buzz, Reader, Plus, Play Music, the steady churn of ever-worse chat tools, and half a dozen other products that were useful enough to take all the air out of a market before being unceremoniously dumped Google has lost all credibility for paid rentals or ecosystem investment.
All the freebie premiere kits are going to be Goodwill gold in a couple years though – the controllers are decent (Not $70 decent, but decent) and seem to work as normal HID devices, and there are two nice USB power bricks with cables in there.
I needed some cheap little vregs recently and had run out of and/or lost all of my useful-value 780x linear parts, so I decided to look at what people in this century use.
I found some little buck boards roughly the size of a TO-220 package that looked exciting. These particular ones are QSKJ Mini DC-DC Buck Step Down Module model “QS-1205CME-3A”, Vendor page here, mine were 5pcs/$9 from Amazon.
Upon analysis they have serious issues with regulating under load, so the hunt for something decent continues, but the form-factor and advertised feature set are really compelling.
Pros:
Cons:
…and that makes it basically useless for most applications. Test data below the fold.
Maybe it could be resolved with appropriate external capacitors and/or offsetting the adjustable to regulate right at a known load or something, but not being drop-in really reduces their charm.
Anyone know of a similar offering that doesn’t suck at output regulation?
(Rel: Anyone know if any of the low-end electronic loads are worthwhile? I’m not looking to spend real-lab-instrument money, but it’s come up often enough lately that I want to be able to dissipate a couple 10s of Watts through a at least stepwise-controllable resistive load).
I ran into a delightful irony trying to help a student get set up for the embedded systems lab I’m running this semester: Neither Keil MDK ARM (a first-class ARM development environment) nor the Stellaris ICDI Drivers (TI’s programming/debug interface for their smaller ARM processors) will run on Windows 10 ARM.
I think this is also the first time I’ve seen a Windows 10 ARM device in the wild.
(For those who know ARM stuff, yes, Keil is kind of crusty, but we like the textbook we’ve been using, it uses Keil style directives and TI details, so switching to CCS with TI parts or a different ARM family would require major retooling.)
