glowingly

I do not know if Boston is unique in this regard, but there was a lot of company housing in the region.

Where I worked in Longwood (Boston), the offices surrounding me were massive, historic buildings with plaques commemorating their origin. They were originally built for housing nurses and doctors who worked in the area's hospitals (there is a concentration of hospitals in Longwood).

So in this case, these were built and used as housing before a later conversion to office space.

These are currently utilized hospital buildings. IMO, they are unlikely to be converted back to residential at this time.

I have followed this approach as well. I use a larger display from further away. I did get a deeper desk to enable this in my current setup. Previously, I mounted my monitor on the wall and moved the desk out ~10" (~25cm).

Either way gives me a greater range of suitable monitors, and they are all cheaper than going for unicorn displays.

Nowadays, I think it is doable with modern displays.

My 165Hz 1440p 32" (31.5") LG with 8 bit color and two external speakers consumes ~20W, measured at the wall.

My 14" laptop screen is about 19.8% of the area of that display. I have loosely measured it from ranges of ~<1W at min brightness to ~8W at max brightness. It is 1440p, 60Hz, 14", HDR (DV). I usually run it at ~20% brightness, which seems to be around 3W.

My office has the usual LED lighting at night or a large window + sliding glass door during the day. Not a dark cave by any means!

IMO (i.e, not an expert, only fiddled around with IC design at .18 micron in undergrad), no fast/cheap try-observe cycle.

Software for web? I (often) can see the finished product as I'm typing the characters in my IDE. Even for PCB design? The output can be acquired and tested under a week with low cost.

IC fab? Not cheap and not fast. All built on machines and processes that discourage experimentation on running lines, since even the slightest mistake can scrap weeks or months of work. Work that involves high pressure, high energy toxic gasses, and lots of them. Also some non-toxic elements/chemicals. Capital costs are extremely high (IMO, this is also part of why working with HW doesn't pay as much - capital costs are high, allowing employers to have serious leverage).

IMO, that is similar to: why the James Webb space telescope cost so much, and why many Mars rovers cost so much. They have to get it right, at almost any cost, because it costs so much to deploy them. Unless a fundamental change in costs (w.r.t. Earth --> Deep Space) occurs, the cost of failure will remain high and so will the costs of development.

They have USB controllers and enough PHYs for 4 USB 3.x ports on AM4. I don't know about serial, but they did have SPI and I2C.

AM5, I don't know.

For me it's probably Ghidra or Jetbrains IDEA in 2023. Both are desktop, thought I could see the debate on whether or not developers are considered mainstream consumers. I use them on Linux and Windows, so I definitely get value out of their cross platform capabilities.

I use Eclipse-based tools at work (again, I can see the debate). It seems like Samsung's Smarthings (IoT platform) used to use Groovy, but has recently migrated away.

I also know you said desktop, but a weak argument could be made in favor of the (also weak) JVM connections of Android. I'd put forth that some Android usecases are basically former desktop usecases.

Wordpad has read docx files for a while now.

edit: I just checked my Win11 VM, and Wordpad is included by default. It also reads and writes docx and odt.

Turn the ad off via the settings.

Settings->Interface->Notify me about additions or changes[...]

Same page also has another option for defaulting to library, community tab, friends, profile - instead of the store.

For the ones that used later gen Nanos (same form factor as early Shuffles) as music devices during running/jogging, the Watch supplanted that role with the added benefit of combining the Nano and fitness bands (GPS, heart rate).

I agree, a full phone was annoying large and hard to carry (tried forearm straps at one point) and often ran without one. Though I know people who still ran with a phone anyways.

MediaTek is pretty decent in terms of open source support for their WiFi chips and SoCs, to the point of continuously supporting a long time OpenWRT developer (Felix Fietkau) to work on their OpenWRT device drivers. This is an effort that is independent of their closed source drivers for the same devices.

Doesn't mean everything MTK does shines under the sun, but in Linux WiFi, they are probably the most open and well supported in the modern age.

I have the Power Monitor widget and it does derive from BAT0, according to the source code for the widget.

On my laptop BAT0 a bit slow to update, with an update rate of about 0.2Hz.

I've also noted my desktop's UPS reports different power draw values on its status display and the UPS' USB interface, by about 10% or so.

If MI300 counts at 146 billion transistors, than I think we can count Cerebras' Wafer Scale Engine 2. It sits at 2.6 trillion transistors.

https://www.anandtech.com/show/16626/cerebras-unveils-wafer-...

GP could be using a double negative. "Less lack of housing."

This somewhat mirrors my own development.

I started in C, and am still in it for systems/OS development at work.

In UI development, we had older C toolkits, but are replacing them with newer toolkits that are C++. I started off trying to just do C business logic with some C++ glue code for the UI. Sticking to what's familiar. In the end, it was just easier to do everything properly. All C in areas that are C. All C++ in areas that are C++. Same with C#. Trying to mix them is just more work than necessary.

At least I get to add C++ to my resume. There is a little joke at work, w.r.t. our existing codebases and code choices. They are career-driven choices, not technical choices. The result is a mess of a codebase, but I'm fully embracing that now.

So.... what happens when someone decides it's a smart idea to replace all of those commonly used functions that had physical buttons on the steering wheel with more touch buttons? Let's even say, capacitive touch zones? :D

I do agree that a good physical button arrangement on the steering wheel is quite useful.

On my car, the layout is quite good, IMO.

* All of the audio is controlled through a D-pad on the left side.

* All phone/voice handled through three physically different buttons on the lower left, accessible without changing how my hand grips the wheel.

* All car-computer/cruise-control through a D-pad on the right.

* As a result, any "type" of command is specifically associated with one hand, and one area of the steering wheel in particular.

As I've learned through a number of long term rental cars, the physical button layout can be bad.

* Many have the audio spread across both hands.

* Many require one to break grip to even do simple audio adjustments (my car doesn't).

* They often overload each area with too many buttons and controls.

* The car GUI can also have many slow, confusing layers that one navigates to get commonly used functions, like digital speedometer, estimated range, etc.

This isn't even getting to strange things, such as having the older generation UI completely present under the veneer of a newer, slower UI.

The car brand I have isn't faultless either - all of their newer cars exhibit the same bad habits. Some bad habits have been slowly walked back a bit in the latest 2022/2023 cars, but the damage is still present.

Sort of like bad ideas in Windows. The worst UI elements (IMO) in Win 8 were walked back a bit in 8.1, then successively more rollbacks via a number of Win10 updates. But the baseline is still worse in some ways vs 7 and before (e.g, start menu is less meaningfully customizable than 7, which is still less than XP, IMO). We see tiny hints of that with Win11 updates. I expect 11.1 or 12 to mostly remove the worst offenders.

^X is …?

It totally obviously means shift + 6 + X.

Too bad that doesn’t exit the program.

Macs in 2022 have an extra ^ on the correct button.

My native Gsync monitor, a now discontinued LG 32GK(?)650 is quite fast as well. I’ve complained a tiny amount about monitor reviewer blind spots, and this is one of them. The speed of every interaction on this monitor is very very fast. It is why I am very unwilling to let it go, even if it keeps me locked into the Nvidia side of things, since my monitor predates native Gsync modules that support DP Adaptive Sync and HDMI VRR.

Better than a gym I turned down: that one required a fingerprint, then handprint scan. WTF? I didn't buy any of the excuses for it.

I agree with the idea, but it may depend on how the X11 application was written.

My experience with X11 over a network (across a city, geographically and distance-wise) was very poor. Slow updates, just extreme latency in doing anything. Not impossible to use, but definitely unusable by any reasonable metric. We're talking dozens of seconds for most GUI interactions.

I eventually piped it through RDP (mRDP, iirc), which itself was piped through a VPN. This vastly improved performance and made it so that I didn't have to be physically present at my workstation. Just as well, since it was during the early covid timeframe.

My LG monitor has a brick that goes to 65W, but the measured power consumption is ~20W at the wall. ~12W in short idle (screen turned off less than ~1 min ago). ~0W in deep sleep, which is where it spends most of its time.

I am running on minimum brightness, since I'm indoors.

As an aside, many 4k/1440p videos are not really 4k or 1440p, vs how we would classify a 4k/1440p monitor. Video is mostly 4k/1440p light levels, but less than that w.r.t. color information.

A common video subsampling setup is 4:2:0.

https://en.wikipedia.org/wiki/Chroma_subsampling

In addition, common video compression techniques (at least the simple ones that I could still understand) basically reconstruct blocks of pixels. Reconstruction is lossy, at best.

If I had to guess, something similar could be done in games as part of their optimizations.

My experience with largely with 14" laptops. I am moderately tall, IMO.

I will admit to using a Thinkpad.

For other laptops I've had, I use my thumbs + tap-to-click. Also a healthy dose of KB. I am buried in the console, so I will admit that a more graphically driven user may not appreciate thumbing the touchpad and navigating by KB.

Sometimes, the options for edge-inertial-scrolling, tap then hold to drag (i.e. click dragging with a mouse), etc seem to mostly be limited to synaptics touchpads and sometimes features disappear with OS or driver updates. Really annoying, when those features are what make marginal touchpad usage shine, IMO.

Needs a relatively rare and expensive W680 board to work. I’ve looked for them. Rare and hard to actually find one for sale, especially in ITX[0]. In past generations, Core i3s also had official ECC support with certain workstation boards (though not consistently - there is at least one generation where the i3 did not support ECC in recent years).

[0] If it was mATX or ATX, I’d just use a non-APU Ryzen with an ASRock AM4 board - they had the most consistent ECC support in the AM4 socket, though are completely lacking it in AM5. This is the route I eventually went.

Do any of the Intel fabbed FPGAs have a free development IDE? Last I checked, it was limited to older Altera-era FPGAs. I know the pricepoint of the Intel fab FPGAs is outside of even my work projects, but I was surprised to see how few FPGA families were supported by Quartus free edition, and was wondering if I missed something obvious?

What I am curious about, is some Nvidia cards like the A2000 have ECC, but only enough chips for a regular roundish number of RAM, like 6GB or 12GB. So when ECC is enabled, 6.25% of the RAM is used for the ECC bits. [0; 1, in the notes]

Since desktop ECC gets around this by having physically more RAM ICs (usually 9 instead of 8, for example), what is the impediment from having a similar solution to Nvidia? I'd readily take a hit to memory capacity* and performance in exchange for ECC.

Why can't the memory controller already do this?

I should note, I'm mostly thinking of my NAS. I know ZFS can be run without ECC and some consumer solutions do. However, it seems ZFS should be run with ECC. I've already experienced observable bitrot with older images and video files, I'd rather not let it progress.

[*] in this case, 12.5% if we follow typical desktop ECC allocations

[0] https://www.nvidia.com/content/Control-Panel-Help/vLatest/en...

[1] https://docs.nvidia.com/cuda/cuda-c-best-practices-guide/ind...

I have run into something similar with really old MinGW compiled applications. Old MinGW used to start allocating the heap at an non-recommended, but technically working address. After Windows 7, however, MSFT enforced memory safety rules a bit more firmly, making this address non-valid and breaking the program. Visual Studio provides a tool to rebase the heap allocation address, so that old 15+ year old program works again in Win10. I chose another address, admittedly one that might fall into the same category as before, but all of us at work hope to be off this old program by the time that happens again. Or we'll just virtualize it again.

En France? Non!

https://www.videolan.org/legal.html

Patents and codec licenses

Neither French law nor European conventions recognize software as patentable (see French section below). Therefore, software patents licenses do not apply on VideoLAN software.

NB: libaacs is not shipped in VLC.

Never bothered me to login. FF 104 with ublock origin. Black on white.

Relevant posts:

I measured the standby power consumption on my set, and it uses an extreme 23 W, or 200 kWh per year For context, this would be approx 10% of the average consumption of a German...

My Apple TV uses 1.2 W in standby

Have anyone else measured their set and seen similar figures? Any way to lower it? I tried setting Eco to high, and that lowered it by a whopping 0.5 W in standby to 22.5...

edit: According to the specs, it's supposed to draw 0.5 W in standby? What is going on? https://www.sony.com/electronics/support/televisions-project...

Thanks, I had already turned off remote start (it lowered standby by 0.5 W). Remote Device Settings did not affect it much either

I reset the TV to factory settings, set it up without logging in to Google, and declining all privacy policies and "extra" services. Still idling around 23 W

After some trial and error, disconnecting the ethernet cable seems to be the trick. Then the TV is able to idle at 0.5 W after a few minutes

Something must be terribly wrong with the software though, that keeps the SoC active if it has internet connectivity? I found in the Norwegian specs that "network standby" should draw only 1 W

I'm no longer using the smart features after I upgraded to an Apple TV last year, so in principle this is not a problem for me to lose Chromecast, but it is quite irresponsible of Sony to have such buggy firmware

[EDIT: I forgot to include the concluding post:]

Indeed, I disabled Chromecast from the app menu and then the SoC is powering down to the 0.5-1W levels after a minute or so

So to summarize: If the TV has a network connection and Chromecast is enabled, the SoC is not powering down such that the set idles at roughly 25W instead of the 0.5-1W per the specifications

This is not good if it applies to more people than me, as each set would consume about 200 kWh per year in standby

German households consume 130 TWh electricity per year, or 1500 kWh per person. So this is sizeable in an European setting, and would help tremendously with the electricity saving targets that may come this winter due to the war

Now, how to get the attention of Sony and Google to fix this? This may even be in breach of the ecodesign guidelines!

by eisa01

Now that you mention multiple endpoints, I noticed (admittedly, this is in USB1/2, which seems to be a totally separate system from USB3) my KB shows up as multiple HIDs. Probably for RGB and nkey rollover. But I wonder how it is implemented? Is there an internal hub (adding one more layer of hubs, ontop of whatever is done internally in my PC before the physical USB port - a lot, if lsusb/lspci-tree-list is to go by)?

Older monitors could also get left behind forever. There was a firmware update for my somewhat buggy monitor, but Benq will only include it with new production monitors.

I do wish reviewers paid a bit more mind to the myriad of GUI/usability issues w.r.t. monitors (waking from sleep, DP disconnects, slow menus that are plagued with inconsistencies depending on the monitor's state, etc). Sometimes, if I'm lucky, a reviewer will make a side note about it in their comments section.