This sort of thing really annoys me. Part numbers are for use of engineers, not for the marketing dept. If you change the specs, change the part number.
I want all 7400s to be four NAND gates, regardless of how they are implemented. As long as the results are correct, you might as well put a little ARM controller pretending to be four NAND gates.
For analog parts, I agree any change to the data sheet should receive at least a different suffix letter.
In these cases of the TI parts, some of their most important specifications, like maximum supply voltage, noise and slew rate, have been changed, and not by a few percent, but by even a factor close to 2.
For so great changes, it is really not acceptable to use the same part number, especially when the part numbers have been in widespread use for many decades, so most users who are familiar to them will not bother to check again their latest specifications, where they could notice that they are no longer what they knew.
I can see both sides of this. I really want different part numbers for the same reason you do.
However we deal with a lot of regulated products and to just open a case at one of the Government Paper-Pusher Regulators will cost us $5,000 to just change the part number. We are a small company and $5k hurts.
I'm sure it does and you have my sympathies, but your situation would not be a reason to let Texas freakin' Instruments off the hook. They're not exactly "a small company", and I wouldn't be surprised if the $5k would have been cheaper than dealing with the response to this, so this just comes across as incompetence on their end.
... How do the "paper-pusher regulators" feel about getting a completely different part unannounced? I would guess unhappy, tbh. Like based on the thread it's not trivial changes.
It annoys me too but part numbers are not a spec but more of a strong hint. The attitude of the industry is that it’s up to you to read data sheets carefully and test. Even for a 2N2222 or whatever.
Keeping in mind, though, that this is a jellybean part. You're supposed to be able to order "a" 5532 without specifying the supplier, because many vendors produce "a" 5532, and they're all the same. Different vendors' 5532s are supposed to be able to be treated as the same SKU — literally dumped into co-mingled stock in warehouses — with no ill consequence!
(And yes, until TI's recent move, that was true of the 5532. All the other vendors' 5532s had matching datasheet specs, including the 22V max input voltage. Because a design that was built for "a" 5532 was usually built to run it up to 100%; and that a vendor couldn't offer their part as a swap-in if it couldn't do that.)
But now, if your purchasing department (or the supplier they purchase from) happens to order TI 5532s — or if the warehouse they're sourcing from has comingled any TI 5532s into the general 5532 stock — then your product is now broken, with no real recourse except to change your entirely supply chain to one that specifically excludes TI.
The EEVBlog[1] video about this has a nice example of only a single chinese manufacturer offering the same stuff as TI now does, even with the same PNP instead of NPN topology. All the others are comparable to the original.
> Different vendors' 5532s are supposed to be able to be treated as the same SKU — literally dumped into co-mingled stock in warehouses — with no ill consequence!
That may be true for a small webshop or a brick-and-mortar electronics store (what few of those still exist). Or be true for end users / manufacturers of equipment that includes such a part.
But (afaik) that's not how it works for large reputable distributors like Mouser, Digikey & co. You don't order a generic "5532" there, you order a 5532 from <specified manucturer> there. Part from manufacturer A may, or may not be interchangeable with same-numbered part from manufacturer B. There's even some parts that have same # but very different function between manufacturers. In other words: buyers, designers do your homework.
Likewise in a design, if you specify "5532" that should read as "any manufacturer's 5532 should do". If not (or unsure / untested), one should specify the part including its manufacturer. Or a list of acceptable manufacturer/part# combo's.
Ofcourse changing the spec significantly for a jellybean part like discussed here (and one with many 2nd sources), that's just evil. Change a part like that, give it its own part #.
>You're supposed to be able to order "a" 5532 without specifying the supplier
This is not true.
>because many vendors produce "a" 5532
This is true, in the sense of a "5532-type part". But you will note that all the 5532 variants have different manufacturer's part numbers (prefixes and suffixes) to prevent this confusion. They don't just do that for branding.
>and they're all the same.
This is emphatically and trivially not true, and it tells me you haven't done the work of carefully comparing data sheet specs across suppliers. Try it, you'll learn something.
>Different vendors' 5532s are supposed to be able to be treated as the same SKU — literally dumped into co-mingled stock in warehouses — with no ill consequence!
That might happen somewhere, but authorized distributors do not do this and volume manufacturers do not do this. You might have an internal part number with an authorized suppliers list that includes more than one variant of 5532 that has been vetted for production.
>And yes, until TI's recent move, that was true of the 5532. All the other vendors' 5532s had matching datasheet specs
Again, emphatically and trivially not true. Take a careful look at the NJM and On Semi data sheets. Spec by spec. Do the work and be amazed.
>the warehouse they're sourcing from has comingled any TI 5532s into the general 5532 stock
Authorized distributors do not do this. It gets hairy when you're sourcing NOS from grey market dealers for old designs or in severe part crunches like 2020-2022 era, but that's a different story.
>no real recourse except to change your entirely supply chain to one that specifically excludes TI
This concept is backwards. You would have an internal part number for 5532-type op amp, and it would have an authorized vendors list that would only include vetted parts. "Any 5532 but TI" is asking for trouble from someone else.
And parts do change or get updated and if you are buying from authorized distributors for production you and your supply chain and quality people will get product change notices. At that point it's your job (or the component engineer's, if you're fortunate enough to have one) to validate the new version or find a suitable alternate.
It has worked this way since the days of the vacuum tube, so there probably is some legal precedence somewhere. I think part of this with the NE5532 is just that these days most EEs spend most of their time with digital where the data sheet and its spec is absolute; in the linear world the spec sheet is an ideal and an average because the parts themselves are an ideal and the real world never lines up with the math. Back when the NE5532 came out it was still common to see power supplies that were unregulated or barely regulated or cheap regulators with poor tolerance that would vary a fair amount with wall voltage and massive tolerances on passives, data sheets and EEs took this into account, most parts would survive max voltage but there would be a higher failure rate, so run at 70% or 80% and you don't have to worry about it.
In these days of cheap SMPS and EEs that are trained with a strong lean towards digital and much improved IC fab, the max seems to be treated as the max safe voltage for good reliability and life, and you don't have to worry much about the tolerances of everything else so much. Back in the 90s when I was learning this stuff, the old EEs scolded me when ever I ran at max voltage and would patiently explain it all to me and that even if it can operate at that voltage, you can't be certain your PS will still be putting out that voltage in a year, parts drift as they age and accidents happen and the world is not ideal. They were right.
They’re just not really standardized at all, especially semiconductors. Not in the sense you’d expect naively. Some were a long time ago, and supposedly the old Japanese sc parts were, down to die geometry and process. But otherwise, the part number means “this is like the part with a similar number first made by someone else”, not “this is an exact replacement in every way”
Specs are just that... specs. A 7400 is one thing, but if you designed a circuit that barely avoids oscillating with a 10 MHz GBW, an unsolicited "upgrade" to the opamp that raises its GBW to 12 MHz may be more of a disaster than a favor.
Frankly it's not OK to "upgrade" plain old TTL parts, either, since a faster 7400 might expose a race condition that never caused problems before, or cause EMI problems that didn't exist before.
> If you change the specs, change the part number.
They took it from SW.
You know this joke with "Windows is a single platform" ? Or the joke with "use rust if you can compile it" ? Or "your browser version is not supported" ?
Gen 4.0 x16 is 32 GB/s in each direction, but the way this is implemented is not the way you'd go about this if you wanted high performance.
Edit: Their benchmarks are also run using ZRAM, which compresses pages before writing to swap. Not sure what the performance overhead of that is, but it's probably quite a bit.
First of all, it's a userspace program hooking the nbd driver, which is known for being slow. It also uses a bounce buffer in userspace before transferring to the GPU. So when the kernel needs to swap a page, it has to first copy it into a userspace facing buffer. The userspace program that has to wake back up and issue the cuda operation to copy the page into device memory.
nbd also doesn't really do a good job of supporting high queue depth or merging adjacent accesses. So if the kernel is issuing a bunch of 4K page swaps without any coalescing, you're going to end up with at least million kernel/userspace context switches per second just to handle 4 GB/s (4 GB / 4K page), let alone 64 GB/s. And that's just the NBD portion, forget the mess that is the NVIDIA driver. PCIe can move a lot of data, but in order to get anything even resembling the full bandwidth, you have to have use DMA engines with long page lists. Having to set up a transfer for every 4K page over PCIe will not reach full saturation of the bus.
Swapping to NVMe is a very optimized path -> the swapper can submit lists of pages directly to the NVMe driver and the controller can DMA them directly out of RAM, no copies or context switches CPU side at all.
This could probably be improved by migrating to the ublk driver as it might let you avoid the userspace bounce buffer. It'd also be able to have multiple write queues to at least set up CUDA copies in parallel.
It's true that Linux kernel is the throughput bottleneck. Unfortunately, the optimizations described above aren't sufficient to get within even 10% of hardware bandwidth.
Even if the swap system overhead drops to just a data copy, the memory management layer prevents swap from scaling to higher bandwidths.
The issue is not data movement; it is in the page unmapping step (which requires expensive TLB shootdowns). Larger kernel changes are required.
Linux's swap system is undergoing some large refactors lately. Hopefully some insights either from our work or Hermit (NSDI '23) can make it in to the mainline. I think Hermit's `rmap` optimization in particular is a candidate for upstream use.
but flash endurance isn't a strong argument here. you probably have O(TB) of flash, and aren't going to produce PB of swap writes any time soon. if you do a lot of swapping to a small flash device, it'll happen sooner.
I'm typing from a quite old 4GB laptop, which swaps heavily to a 250G SATA ssd. sure, it's not great, but it also costs zero. currently 9GB of swap is used, and it's not really noticeable. if I open 20 more tabs, it can introduce pauses.
google says this drive was released in 2014, and SMART says POH is about 10 years.
SMART also says wear leveling count is 665 and total written is 165327189538 LBAs (78834 GiB, or 338 drive-writes). I'm not expecting it to die soon, though using a 4G laptop is a bit of a stunt these days...
the point is that a system that has sustained heavy swapping for years has not generates so many writes to worry much. a modern system with 10x speed and 10x capacity (and probably less RAM deficit) would have even less effect. even for QDR with it's few-hundred cycle endurance spec...
I guess you haven’t tried AMD’s composable kernel on Gentoo, or qtwebkit. I have a special env for the former called half-the-threads because it eats 2.5GB per thread. I removed the latter as soon as I was able to. I even add 32GB (half my RAM) of ZRAM for CK, and the Gentoo ebuild has a check for enough RAM per thread that stops the build if unmet, it wasn’t there before and I’ve had my system lock up because of OOM which OOMD wasn’t quick enough to catch.
All of this is to say that, it does have a potential impact on flash, if you rebuild often, which tends to happen on Gentoo.
This is not HPC, they don't need to 100% use all cores.
If they knew how to do a state machine, a glorified chat app needs how many threads? Maybe 4: UI, network, audio, video. And the last 2 are only needed during calls.
They have 12 just because they don't know to do better.
Btw with teams (the application) off there's a com.microsoft.teams2.agent there anyway. What's it for, spyware?
Cars are now using eSIMs. Cutting an antenna wire only limits the effectiveness of the communication. You can wirelessly send and receive data with a solder terminal on a board if you're dedicated enough.
Yes, but the YouTube ed channels are such a treasure in and of itself. We had the “tech” to produce content like this for almost a century, but it took the Internet and democratization of content creation to come up with gems like smarter every day, veritasium, extra history, etc
My fear is that this is also being reshaped with ai, mostly for good now but I feel like the personal touch and passion of these creators is being diluted with the advent of generated content.
Maybe we are in a valley of the uncanny valley and the ai tools will become so good that they can successfully translate someone’s passionate vision faithfully, then it could be another renaissance.
In among all the MrBeasts and JackSepticEyes on Youtube there are some incredibly creative people.
Two that my 5-year-old loves are OddAnimalSpecimens who could easily have been on BBC children's programming in the 1980s, and Terragreen who would have been his ITV counterpart :-)
Probably the most entertaining child-friendly programme you can watch right now is whatever Jake Carlini is doing. Some wee guy in a house in Austin, Texas is coming up with better stories, better production values, and better life values than any of the "proper" children's TV productions, except maybe Sesame Street.
Thanks for the recommendations - I’m also big fan of 3blue1brown and PBS science, but as a recent dad am on lookout for content for my son to watch when he comes of that age - he’s just 1month now, hopefully by that time AI has not enshittyfied everything
We're a multilingual household, so another that gets a lot of love is Sendung mit der Maus which was originally a TV series but now is on Youtube as well - including some very old episodes. My son's German is way better than mine though, and these days so is his Gaelic - mostly I deal with people in English and I've kind of started to lose that skill.
If you like big 4x4s (and who doesn't?) then Matt's Off Road Recovery is pretty good. Utah looks lovely, and of course they're culturally fairly free of rude words so that's pretty okay for children.
Quiet Nerd is another of my son's favourites, he builds little electric-powered campers and drives them out into the woods near where he lives.
Thinking about it I have to revise my statement somewhat. I have seen The Great War, Technology Connections etc and my Youtube algo is after 15 years very tuned to me.
The issue is somewhat that this stuff needs to be pushed more into peoples feeds and not pregnant spiderman videos.
> The youtube channels are nowhere near the style and depth of documentaries like the ones above...
My friend, if you enjoy long format, deep diving documentaries written, produced and narrated without AI about Space, Physics, Human evolution or planet Earths history, then I insist you head over to the History of the Universe YouTube channel and start watching!
This specific video is probably my favorite (I'm a sucker for contemplating "time"and what it actually is) and was the one that got me hooked on their channel. They go way deeper into the details without becoming a formal lecture and it's genuinely captivating. https://youtu.be/ZSmNii0uOmw?si=3Jaty3XcMGlryhh2
Depends on what you follow. For example, look up The Great War.
At least where I live, basically everything that's on discovery, national geographic and the history channel now is just "experts" talking (reading a script) about "hitler's secret sex life" or some such thing, interspersed with a re-enactment shot or one of the "experts" walking around a slightly relevant building.
That is fair, I have checked out some The Great War videos. Not to mention podcast stuff like Hardcore History. They are very good an in depth. There is also full MIT/Harvard courses on Youtube. So yes it is all about what one looks up.
Any particular recommendations? I’ve been meaning to queue some up to have in the background playing when the kids are around hoping to stumble across something that that might pique their interests
Throwing out my recommendation for History of the Universe as well as its sibling channels. Honestly, the people behind these channels produce higher quality documentaries, both in substance and style, than 99% of the "professionally" created stuff I've watched in at least three last 5 years.
Huygens Optics - https://www.youtube.com/@HuygensOptics - Retired optics guy, lots of interesting stuff on making lenses and other optics phenomena and physics.
Dr. Jorge S. Diaz - https://www.youtube.com/@jkzero - Really good videos on the early history of quantum mechanics and related physics around that time.
Idealized Science Institute - https://www.youtube.com/@idealizedscience -Educational non-profit aimed at helping teachers and students, mostly physics. Typically featuring great physical demonstrations.
Tasting History - https://www.youtube.com/@TastingHistory - Recreates historical dishes, as well as serving interesting history about or surrounding the dishes.
Modern History TV - https://www.youtube.com/@ModernKnight - Lots of interesting videos about the middle ages in Europe, how people lived their lives as peasants or knights etc.
I think if you watch Connections 2 & Connections 3 (and even Start Trek: The Next Generation) then you can see the progression that all documentaries went by the 1980s. "Story" used to be more important, but then "eye candy" became far more important.
Connections was so influential that my university (Purdue) introduced a 3 semester series of courses on the history of technology.
Agreed that Connections 2 and (especially) 3 were pale shadows of the original series.
However the related The Day the Universe Changed, which took a slightly different conceit (the episodes aren't necessarily linear in time) and focus (science and philosophy rather than technology), is excellent, and I'd put it right next to the original Connections series.
(There is, it turns out, a fourth season of Connections, "Connections 4", released in 2023 on Curiosity Stream according to Wikipedia. I've net seen any of it. It consists of only 6 episodes. <https://en.wikipedia.org/wiki/Connections_(British_TV_series...>.
Modern audiences are expected to be glued to twelve different things at once. Producers are being told to adjust to this reality. Watch any movie now and they are all compensating for the distracted audience.
Movies used to be watched in a place for that purpose. Now its the toilet. Now the phone itself is ringing. A message comes in. Time to upgrade. Ding! All while some key scene in the movie is taking place.
Well, even if it’s the living room, it’s not like being glued to a seat in a theater.
I used to be associated with content groups at a former company and in the almost 15 years I was there we saw clear trends in type of content and length of content viewers consumed.
> Perhaps it's just me, but modern documentaries are rather dumbed down?
I hear this often and believed it myself for a while, but I cannot find much evidence for it. It's just the presentation style that people are nostalgic for.
These old documentaries are quite dumbed down as well (and sometimes wrong!). If you were to just read the literal text you'd think it was an excerpt from a children's book. The next time you come across an old video you feel this way about turn on the captions, mute the audio, and block out the rest of the screen except that tiny rectangle at the bottom.
I agree there was pride in the work that shines through, but don't be fooled into thinking it's any "better".
I think the number of good documentaries has remained constant as the total number of documentaries has risen. So sure, the percentage of documentaries that are good has fallen as we are innundated with crap, but as previous posters have noted, there are still several good productions every year.
There was a lot of garbage back then too, they just aren’t a memorable so that’s why you only remember the best ones. There are a lot of great documentaries out these days. The early to mid-2010’s had a huge surge in particular, though there is plenty happening now as well.
I don’t know how someone could call The Act of Killing or Five Broken Cameras “dumbed down.” Or 20 Days in Mauripol just a few years ago
Unlike with a lot of award shows, one can actually do pretty well watching all the Oscar nominees for documentary any given year. Guaranteed at least half are good or great.
Even Golden Age TV documentaries can seem dumbed down compared to actual books. Even at the time, in the 1960s and 1970s, thinkers expressed concern that the medium of television was inherently likely to delight audiences with spectacle more than truly educate them.
My parents have a book published in 1849, "The Chemistry of Modern Life" and it's interesting to see how they transition very deliberately between "technical" and then "dumbed-down" descriptions of things.
It's as jarring as Star Trek's habit of "30 seconds of technobabble followed by a metaphor involving a balloon" trope they keep hammering.
> Connections, Cosmos, Civilization, The Ascent of Man and Attenborough's Life on Earth
Civilization and The Ascent of Man were both commissioned by Attenborough - he had a major impact on the broadcast landscape well beyond natural history.
We watched all of The Ascent of Man in middle school. The only thing I remember is that when the narrator, Jacob Bronowski, gestured with the back of his hand (fingers up), he always tucked his middle finger behind his ring and index finger. I assume this was so that he wouldn't "give the finger" to the audience.
(I've been wasting neurons on this for fifty years...)
I feel it myself, I am dumbed down too. Having trouble even formulating this as I never type formally anymore.
Our state TV SVT buys in documentaries from BBC, Showtime, PBS and some of their own production. Some of their own are still good. The BBC ones are absolute garbage dumbed down now.
The world the aristocrats warned about in the 60s and 70s are here now.
What are 2 or 3 feature length documentaries you watched produced in the last 5 years? I’m curious what you’re watching and what made them so bad. I saw several excellent ones over the last few years.
They’re out there and they’re not obscure IMO.
> What are 2 or 3 feature length documentaries you watched produced in the last 5 years?
Since GP mentioned TV, I suspect the ones he's complaining about are the ones I complained about: one-hour TV documentary-style shows like BBC Horizon or PBS Nova.
The ones I've seen from recent years contain interesting stuff, but the presentation is too rapid, to flashy, too repetitive yet not enough time to let things settle.
> Perhaps it's just me, but modern documentaries are rather dumbed down?
It's not just you. Most modern TV documentaries, especially series, are dumbed down and sped up. Fast cuts, lots of woo, not too much to challenge your brain, don't want it to get strained.
Gone are the days where someone conveyed the information calmly while not driving a car somewhere irrelevant. No more lingering shots allowing you to process what you just saw and heard.
Thats because we have a trove of in depth specialist and deep youtube content including all those old documentaries to mine through these days
Youtube and the internet is a goldmine and way bigger than old 80s/90s content, im over 50 and remember the 80s well enough.. a few great well produced documentaries are not a comparable to gigabytes or petabytes of videos and podcasts we have today
The cultural format of exchange has changed and the consequences of that - so called tiktok attention deficit folks means perhaps no one watches this content but I think that too is a generalization and great content is watched probably by a greater proportion of smart curious people today than back in the 80s on your phone nonetheless- we have a pocket tv with an almost unlimited amount of content
Im an information junkie and just today I spent 3 hours watching a documentary series on the incan civilization follower by a Stanford video on LLMs and then watching Blaise Arcas’s interesting ideas on computational life and intelligence
Is that to be the end result of the pursuit of knowledge, creating something? There is the true dumbing down, insisting on a vague kind of productivity as the point of life.
eh, sometimes its just passive consumption, I'm not claiming it makes me a better person..but it rattles around in my head and latter bubbles up as creative solutions I think.. I would be an information junkie in whatever world I was in regardless..
>Perhaps it's just me, but modern documentaries are rather dumbed down?
A pet peeve of mine is the sound effects added to nature documentaries. I had to explain, once, that the ants do not actually sound like robots no matter how far you zoom in, despite the whirring of servos that the editors decided to add in.
Yeah, everything being black on modern motherboards might look cool with RGB lighting, but makes it harder to work on. I like the older green PCBs with white PCI slots.
I also lament the demise of color coded connectors at the back. I knew to plug my speakers into the green 3.5mm jack. Now everything is black, so I need to look at the manual again to see which of the 5 connectors is the right one.
I remember being a kid when standardized port colors came 'round (what was that, part of the AC'97 spec or something?). I thought that was dumb: I knew that the speakers plugged into the third hole from the top, and that was good enough for me. ;)
Back then: I would have loved black-on-black, labelled-in-black, with black cables and and black highlights on a black background. The accessories would be black, too: Black keyboard, black featureless keycaps, black mouse, with a black mousepad, on a black desk, in a black room with black walls and black windows.
Black.
I couldn't get black back then, of course. Computers were beige. The necessary floppy and optical drives were beige. Cables were beige. Keyboards were beige. Motherboards were some moral equivalent to beige. It might be possible to get one or two components in black at some points, but the rest were going to be beige so therefore the whole thing might as well just be resolutely beige.
That really annoyed me.
But I'm not a kid anymore; I'm old. I just want stuff that works well, and that is expandable enough to do some fun and unusual computing stuff with, and that I can see so that when I'm futzing around with it then my job is easier than it would otherwise be.
I don't want RGB or a tempered glass aquarium that shatters when part of it touches a tile floor the wrong way. I don't care about having multiple choices for the color of the anodizing on the heatsinks for the RAM. I don't want water cooling when a big slow-moving fan and some heat pipes does the job very quietly, with improved simplicity therefore longevity. I'm not trying to win a cooling benchmark; I'm just trying to keep the CPU within its specified thermal range while it does work for long periods at its maximum speed. I don't care what color the fans are as long as I can't hear them.
If I want to play with RGB by making or buying some party lights, then I know how to do that. Party lights for the room (or the whole house!), not the guts of the PC. :)
Otherwise: The computer is on the floor under the desk and the USB hub is on top of the desk, and that's all I need to deal with. It is purposeful and functional. There's no style points here, but I just don't care about that anymore.
(I'll be outside yelling at clouds if anyone needs me for anything.)
Agreed. My computing philosophy: If you aren't looking at the display, you're computing wrong.
I have a black case (some 10+ year old Fractal Design model) and an all black keyboard with no labels. Back in the pandemic, I was fortunate enough to score a videocard that happens to be light up RGB unicorn poop. I hate that part about it, so that helps remind me to keep the side panel of the case on. (I could, but I'd rather not disassemble it to unplug the LEDs.)
Maybe progress in terms of pure GHz measures or similar, but new and better CPUs are still being released, even outside of Apple. The CPU I'm on right now (AMD) was released in Q3 2025, and almost every CPU released today offers better value for the money than the previous generations.
reply